Semiconductor Fabs I: The Equipment

Semiconductor Fabs II: The Operation

Preface

I tried to include as many links as possible to allow the reader to go down rabbit holes as they see fit.

I try to include analogies in case the explanation is poor or the topic esoteric.

I don’t work in an advanced fab, but have some glimpses into them, so I rely on my ideas, conjecture, and the literature for a more accurate representation of what state-of-the-art fabs look like (although I’m sure there are features I could never dream of).

I first go over the data side of things, since a lot of that is fundamental to how the automation operates.

Why All the Data?

Fabs are incredibly hungry for data. Insatiably hungry. Data helps to connect patterns, solve problems, troubleshoot issues, and just plain understand what the heck is going on at the atomic level where the transistors and interconnects are made. I call it the fab data monster and Nano Banana 2 thinks it looks something like this:

More data is almost always a good thing because it allows you to fit your conclusion to the data. Just kidding. But not really. If an engineer has no freaking clue what’s causing a problem, blindly scouring the data may help uncover some anomaly that can clue them in on root cause. (But if you’re having to blindly scour the data, you probably don’t have enough data or your FDC systems aren’t developed enough.)

That said, false positives and false negatives are legitimate concerns that have to be considered. False positives may result in troubleshooting efforts in the wrong area and wasted time, money, and effort. False negatives may result in the ultimate root cause being overlooked and wasted time, money, and effort pursuing the wrong area.

What is the Data?

How much data could one fab possibly need? And what could they possibly be measuring that culminates in petabytes (1000 TB, or 1,000,000 GB) of data?

Tool Signals

Here’s a “short” list of potential equipment signals fabs can keep track of. For example, engineers may want to keep track of the temperature, pressure, and power of component1, while voltage, current, and resistance are relevant to component2.

• Temperature

• Pressure

• Angle

• Distance

• Position

• Voltage

• Current

• Resistance

• Power

• Quantity

• Status

Now those are just single characteristics that don’t add up to much storage space on their own. But what happens when we take measurements across multiple components across multiple tools across the entire fab?

Let’s assume the following for the NMP fab:

• 1000 tools

• 250 signals per tool

• 500 kB data per signal per hour (empirical data at a collection frequency of 1 data point per second; this includes the context needed to interpret the data (note that 1 Hz frequency is fairly slow and newer tools and fabs can accept data frqeuencies of up to and even exceeding 100 Hz, or 100 data points per second)

The math is then pretty easy:

total = 1000 tools × 250 signals/tool × 500 kB/signal/hour = 125 GB per hour ≈ 1000 TB per year of data

But wait! Those are just raw signals that the tool is reporting to the fab data monster. Statistics can be performed to get some extra info for each signal: mean, median, standard deviation, minimum, maximum, etc. Just those alone results in five times more data per signal than before (assuming you are constantly updating across the same time period, but generally a time period is defined and a single data point calculated for said period).

That’s a lot of data that is just passively created and recorded, some of which will be looked at, a lot of which won’t be. Regardless, it’s nice to have in case you need it.

Test Results

In-line Measurements

Wafers regularly get measured—either randomly, as determined by some algorithm, or intentionally due to the criticality of the process it just went through—throughout the line to ensure quality control at all processes. The measurements may be thickness after some film was deposited or etched, critical dimensions, number of particles on the wafer, or more specific measurements that are left up to the reader to determine.

These results are generally boring and not looked at because, well, they rarely fail, at least in more mature fabs where the technologies’ manufacturing processes have been optimized for years. Regardless, the tests are required for various reasons and results must sit in storage for some time.

Electrical Measurements

Some fabs (all fabs? Not exactly sure here.) will test their wafers in-house towards the end of the line to shorten the feedback loop if an issue is identified or get test results quickly so they can make changes permanent; it would be weeks if they chose to wait until the wafer got what is called its “final test” results, which measure the chip’s performance at its intended purpose.

In most, but preferably all, cases, the electrical test results are the fab’s gold standard for the quality of the wafer: if it’s passing and within the historical distribution of that parameter for similar devices, great! If it’s not, then something appears to have changed either within the line or with that specific lot. If the next lot of wafers that gets tested has similar out-of-the-distribution results... get investigating!

A few not-related-to-the-main-topic notes here for the more technically curious:

• Matching electrical data is pretty much how all changes ultimately get approved. Say I want to make change X to process Y. I will propose the change, get approved to run some experimental wafers, run them, review the test results, and if they’re good, request approval to fully implement change X for every wafer that runs through process Y. Changes may included small tweaks to processes or entire process flow changes.

• Some parameters are strongly related to certain processes within the line, which helps to narrow down what went wrong instead of having to spot check every tool. For example, if the threshold voltage is all out of whack, that points to a problem in the gate oxide area, which is a particular machine or two in the fab. The engineers reviewing the test results can then notify the gate oxide engineers and have them look into it by digging through the tool signals or maintenance history.

Histories

Histories allow engineers to look back and see what happened on a certain date or to a certain lot. Tool signals are a form of history (what was component X doing at Y time?), but other histories are also important.

Maintenance

Fabs want a way to easily document events in a machine’s life, whether automated or input by a person. These leave digital bread crumbs of-sorts that can be checked to see what happened, why it happened, etc.

Here are some examples of helpful automated comments:

SPC chart X failed with value of Y1 and Y2. Limit is X. Last Z points have been in control. [This is an event that I can anchor to and look around at: what maintenance, if any, was done before the chart fail? What was the response to the chart fail? Was something repaired or replaced?]

Machine alarm X with description “Y” occurred. It has happened Z times over the past 30 days. Review recommended troubleshooting and solutions here: [link]. [This is an event that I can anchor to just like the last one, plus I get to see how bad the issue has gotten.]

And here are some examples of helpful typed-by-a-person comments:

Removed parts A1/2/3, B, and C to better diagnose lift issue. Found that part A1 is sticking at its end range of motion, which corresponds to the side of the lift that’s having the problem. Parts A2, A3, B, and C are all good and have no obvious issues when testing. Regardless, replaced all three part As since everything is open. Original B and C will go back in. The new A1/2/3 serial numbers are 123, 456, and 789, respectively. Next steps are [list of next steps]. [This tells me exactly what the issue was and what was replaced, so future me can just reference back.]

Machine alarmed for X. Found that part D was completely powered off. Verified that all relevant circuit breakers were on and there is no power discontinuity up to the part, so appears that D has failed. Wafer 13 was 10 seconds into step E of recipe F when the failure happened. All wafers placed back into the FOUP. Replaced part D and verified it has power and functions normally. Machine was vented to atmosphere and opened for replacement. No other issues noted on machine. [Same as the above.]

This is all data! It may not be numbers, but it paints a clear picture of the what happened on a machine during a certain time period.

Lot Processing

Lots will get data and information automatically “attached” to them throughout their life. Examples include what machine the lot processed on for a certain process, what time it started and stopped, were there any abnormalities while processing, what associated data (as in in-line measurement results) is there. The list goes on. Like the in-line measurements, this data is really only looked at when there’s an issue.

Automation (of Everything)

Automation is a beautiful thing. It helped get us cheap and abundant everything, including semiconductors.

Automation here is referred to as, well, anything automated, and no, I’m not being a smartass. A majority of fab automation has to do with the actual wafer processing and making it less error-prone and more efficient, but there are plenty of other uses.

The Wafer Life Cycle Without Humans

The life cycle of a wafer—from its start as just a bare silicon wafer to its end when it’s full of chips—can be looked at to get a better understand of what fab automation is and isn’t. I’ve provided significant detail both for nerd-sniping and to get a good picture of how many decisions are actually automated in the fab on a second-by-second basis. While reading, think about how time-consuming and error-filled a fab would be if humans had to make all the decisions and perform all the calculations.

Here’s the basic flow for how a lot runs through the line, along with some non-ideal situations arising throughout to show what automation can and can’t do:

1. Go! (But no collecting $200—the fab needs that money!) The 25-wafer lot is assigned a lot number—call it 123—and device—call it ABC. The manufacturing execution system (MES) knows every single process that 123 has to go to and what it needs to do at each.

2. Lot 123 reaches its first process: thermal oxidation in a vertical furnace. 123 is ready to run and there’s an available furnace, so the MES dispatches 123 to the furnace, which initiates automated pre-flight checks that makes sure everything is copacetic:

   1. Is the machine actually available to accept this job request? The MES will query the equipment management system and tool itself to verify it can accept the proposed job.

   2. Are all of the necessary SPC charts in control? Common SPC charts for qual wafers (wafers that run a test to ensure the machine is performing properly) are:

      • Thickness: The qual is targeted to and representative of the process that the job will run, e.g., if the job’s process grows 10 nm of oxide, then the qual will also grow 10 nm or can be easily and reliably extrapolated from a different thickness.

      • Particles: The qual ensures no particles are being generated by the machine.

      • Contamination: The qual ensures there’s no metallic contamination coming from the tool, often in the form of particles.

   3. Is 123 allowed to run on this machine? Some fabs will automatically qualify like machines if everything matches, while others require individual machines to be qualified for each process. The former is much faster, but risker; the latter slower, but safer. See Intel’s (in)famous Copy EXACTLY! method.

   4. Checks current tool settings against a pre-defined list. Do all of the settings match? Incorrect settings can cause misprocessing or failure to interdict on a process that’s going poorly.

      • Analogy: Assume a vehicle has customizable air-fuel ratio alarm settings, where 14:1 air:fuel is the ideal, 13:1 is the lower limit, and 15:1 is the upper limit. If it goes above or below the upper or lower limit, the car alarms and notifies the driver. Before driving, it would be wise to ensure these limits haven’t been temporarily adjusted by a pesky, speed-happy teenager who forgot to change it back in their excitement of going fast. If either limit were changed to something much larger, the next drive could be in the harmful-for-engine range without the driver knowing.

3. 123 makes it to the furnace and runs the correct recipe that automation told it to run. There is no need to make sure that recipe A1B2C3D4E5 is selected vs. recipe AIB2C3D4E5 (see what the difference is?) because automation commands the correct one to be chosen.

4. 123 gets put on hold after the furnace because the post-thickness measurement was too high. Savvy automation systems would automatically remeasure the lot to ensure the measurement was legit (2 hours of the lot not moving, no manpower required), while stone-age automation would require a person to manually remeasure themselves (4 hours of the lot not moving, plus valuable manpower required). The measurement turns out to be bogus and the lot continues on.

5. 123 continues on and reaches a certain process (call it process 27) in the flow that requires feedforward data. The general feedforward flow goes like this:

   1. Data from process 26 for lot 123 is “sent” to process 27. Process 27 knows how to adjust itself based on the process 26 data. Adjustments can include a variety of parameters depending on the process, such as time, power, pressure, etc.

   2. A handful of wafers from 123 (call them 123-1, 123-2, and 123-3) are sent ahead of the rest of its brothers and sisters in case a sacrifice to the fab gods is in order, i.e., they will be the test wafers for the rest of the lot to ensure everything is fine. 123-1/2/3 all complete process 27 with the process 26 data and measure to check the results, resulting in process 27 data for wafers 123-1/2/3.

   3. Wafers 123-4 to 123-25 are then sent on to process 27 with combined data from both process 26 (for the entirety of lot 123) and process 27 (wafers 123-1/2/3 only).

   Pretty simple, right? Right?! The granularity of customizing this flow is practically infinite (depending on the sophistication of the fab automation, of course). Custom adjustment values can be set based on machine (from both process 26 and process 27), device, measurement tool, etc. Full moon? Adjust accordingly! Good or bad vibes in the air? Adjust accordingly!

   

6. 123 makes it to process 53, where an engineer has defined an experiment that will test a few different conditions. Individual wafers will be separated into distinct lots, process, then recombine into a big lot. The general process goes like this:

   1. Wafers are separated into “sublots”, e.g., wafers 123-1/2/3 become lot 123-A, wafers 123-4/5/6 become lot 123-B, etc.

   2. Lot 123-A goes through process 53 with any experimental conditions that the engineer defined. Repeat for the remaining lots. There is almost always a baseline lot to compare to.

   123-A/B/.../Z continue on until they reach the recombine point, which could be the next process or multiple processes down the line. Now imagine if the split occurred at a feedforward process and how complicated that would be!

7. Wafer 123-13 breaks randomly and other wafers are exposed to the particles that come with the break. Good automation would recognize that there was a wafer break and automatically inspect the lot for particles on other wafers, then make a decision to clean the lot if particles were found. Bad automation would put the lot on hold and wait for an engineer to review the incident, request inspection, review inspection, then make a decision. (In case I’m not being clear enough, I think pretty much anything in fab automation land is possible with sufficiently talented software engineers. It’s just a matter of assigning the necessary resources.)

This flow isn’t an exhaustive list of all of the potential pain points, but illustrates a good chunk of them. Now imagine having to do all of the following by hand:

• Assign a lot number and device. Was everything entered correctly?

• Verify that the machine can accept the job. Are all the settings—the tens of thousands that exist—correct? Are all of the SPC charts in control? Is the machine qualified? Was the correct recipe selected?

• Review the thickness measurement fail. Is it real or fake? Does it need to be remeasured? Do you need to tag the lot for later inspection or testing?

• Make feedforward process adjustments. Was the math done correctly? Was the correct parameter adjusted? Was the full moon accounted for?

• Separate the lots correctly. Were the correct wafers chosen and settings applied to each?

Now rinse and repeat some of these multiple times for the hundreds, if not thousands, of lots running the fab at any given point. It’s unsustainable and overwhelming, hence the need for automated systems and processes to do most of the work. QED.

Other Automation Functionalities

Fault Detection and Classification

Averroes does an excellent job of explaining FDC systems and there’s no need to reinvent (or re-explain) the wheel.

Part Management

I’ve never seen or heard of anything like this, but if I could vibe-code up a part management system, it would look something like this:

• Detect what maintenance is coming due and what parts are needed for said maintenance. If parts are on-hand, excellent; if not, order them while taking into account current stock, lead times, historical delays.

• Learn how often a part fails and extrapolate out to ensure there is always a spare available. For example, if machines 1-12’s partX fails at a rate of once per month, then 12 fail per year on average, so there should be at least one available at all times, but preferably two in case failures occur around the same time. Remember: two is one and one is none—equipment downtime is often more costly than paying for and keeping an extra spare on hand!

• Original equipment manufacturers will often obsolete parts because of obsolete components, upgrades they’ve made, or issues they’ve found. OEMs could alert the company, which would trigger the following actions:

  • Flag what tools are at risk, if there is a risk, and notify the engineers.

  • Update the part number in the company system to reflect the new OEM part number. Companies have their part number, which is separate from the OEM part number. For example, NMP LLC’s part number for partX is 1111-222, while an OEM PN is 1234-12345.

  • Change all existing references to the old part number to the new part number (like a pointer’s memory getting updated)

Praise Be to the Automation Engineers

Automation here is similar to automation elsewhere: it makes people’s lives and the manufacturing process safer and more efficient. And it’s freaking awesome. I can sit at my desk and do a good chunk of my job without moving because some automation guru coded up a wonderful script that helps me out.

And while the fab requires everyone to operate smoothly, the automation engineers are the heroes working in the background that nobody really thinks about. Here’s to them.

See Also

• Equipment Anomaly Detection for Semiconductor Manufacturing by Exploiting Unsupervised Learning from Sensory Data

Three types of lines exist in the policy and behavior sense:

• Soft: These are okay to cross, but not preferable. There may or may not be a tangible action taken afterwards, but the person whose line was crossed should take note.

• Firm: These are somewhere between soft and hard lines and should result in some tangible action being taken that is less drastic than the hard line.

• Hard: These are not okay to cross and (should) result in some tangible action being taken that is more drastic than the firm line.

Most lines are rarely set and rarely thought about in detail. Most line setters use the good ol’ “I know it when I see it” test, waiting for something to happen before they decide what to do. This is a poor practice because of the pernicious force known as normalization of deviance.

When lines aren’t set before they’re crossed, it forces a decision to be made at the time of crossing (if it can even be recognized that something was crossed!), during which many things can happen:

• The line setter convinces themselves that the line wasn’t really crossed and everything is fine. This will land the setter in not-so-nice territory if this occurs enough times because the line effectively moves back each time.

  • Ex: Ruben, Lou’s boyfriend, playfully pinches her, then playfully punches her, then seriously pinches her, then seriously punches her, and so on. Each time she convinces herself that her domestic abuse line wasn’t crossed, ultimately leading to her getting full-on abused.

• The line setter acknowledges the line was crossed, but because taking action is uncomfortable at the time of crossing, vows to wait until it happens a second time because the first time may have been a one-off. This increases the likelihood they give a third chance to the offense when/if they apply the same thought process to the second.

  • Ex: Diane blatantly lies and talks about Joe both behind his back and to his face. Joe explains away the behavior as Diane having a stressful time and continues being “friends”. Diane continues the behavior while Joe accepts and normalizes it as Diane’s personality. Joe’s self-esteem decreases as he continues to spend time with Diane.

• The line setter acknowledges the line was crossed, but convinces themselves that the line really should’ve been just a teeny bit further when they originally set it.

  • Ex: Harlan’s original salary threshold for taking the Giving What We Can pledge was $100k/year, but now that he’s reached it, it feels a bit low. After all, he deserves to treat himself a bit more for all the hard work he put himself through to get to the coveted six-figure salary. Plus, he may have a baby in the next few years! And everyone knows how expensive babies are! Harlan resets his salary goal at $120k, which will be plenty when the time comes.

By setting a line and its corresponding action early, the action becomes the default until proven otherwise. This is similar to trigger-action plans.

How to Set a Line

Here’s the general process of setting a line:

1. Figure out the general line. Whether it’s domestic abuse, talking smack, donating money, or rights being restricted or outright revoked, it must be defined.

2. Define the criteria for both the soft, firm, and/or hard versions, but especially the hard. The soft line being crossed serves as a forewarning to the hard line being crossed, giving ample preparation time for if the hard line is eventually crossed. The criteria must be well-defined with little room for interpretation.

3. Decide how many times each can be crossed before the action is taken. It’s fine to give someone a stern reminder that they crossed the line in case they forgot, weren’t aware of the line, weren’t aware that it was soft/firm/hard, etc. It’s not fine for it to happen more than the set number allows, especially if previous actions were taken.

4. Define the actions for each line. This can also be done in conjunction with deciding the number of times it can be violated, since more drastic actions should have fewer subsequent violations and thus a lower number of allowable violations.

5. Define what circumstances would have to be present for the action not to be taken. What evidence would it take to show that the hard line was crossed, but the action shouldn’t be taken? (This is a bit contradictory to how hard line is defined above, but the hard line action is simply the default, not a blind requirement that must be executed. Setters should double-check they didn’t miss something before taking the default action.)

6. Communicate the lines and actions to people who either may be at risk of crossing them or will help with maintaining accountability of executing said actions.

7. Prepare for taking the action when/if the time comes. Preparation may be mental, physical, or environmental.

Ensuring Accountability

The line means nothing—and in reality, is likely a large cost—if the action is never performed when it should be. Assuming the fourth and fifth steps are done honestly and comprehensively, it should be clear what decision needs to be made when the line is crossed.

Thus, an accountability method must be put in place to enforce the action being taken.

A few ideas that all rely on the honor system to some extent:

• Require a cost greater than that of said action be paid. If the action costs $10, make the cost of not doing the action $20.

• Publicly or privately announce the lines and ask a trusted person to be your accountability partner. They know your lines and make sure you follow through on the actions, else a cost will be incurred (see previous idea).

• Automate the action. For example, write a script that looks to see if the friend who borrowed money ever paid it back by a certain date. If current_date > deadline_date & money_repaid = false, then send an automated email unfriending them.

Line Examples

Here are some hard line ideas and associated actions (in no particular order; assume the case is straightforward with no nuance):

Government Overreach

• Soft line: A government violates a law with the expectation that the lengthy legal process will allow them to reap the benefits before a ruling is made

• Soft line action(s): Protest

• Hard line: A government blatantly violates a constitutional amendment or refuses to comply with a court order

• Hard line action(s): Apply for a visa or similar in another country

Relationships

• Hard line: Romantic partner commits adultery

• Hard line action(s): Break up/divorce

• Hard line: A friend doesn’t repay an $X loan

• Hard line action(s): Stop being friends with said person

• Soft line: A friend makes disparaging comments about you, but claims it’s “just a joke”

• Soft line action(s): Tell them to not do that again, but continue being friends with them

Workplace

• Firm line: Boss makes an immoral or illegal request, but doesn’t retaliate when it’s refused

• Firm line action(s): Submit a whistleblower complaint; submit an ethics violation with the company; begin a new job search; resign

• Hard line: Annual raise is 0%

• Hard line action(s): Begin a new job search

• Hard line: $Xk/year annual liquid compensation

• Hard line action(s): Donate X% to charity

• Soft/firm/hard line: Achieve $X net worth

• Action(s):

  • Retire

  • Say fuck you to your terrible boss

  • Start looking for a new job that pays less, but is better otherwise (stress, hours, culture)

Health

• Soft line: Weight above X

• Soft line action(s): Begin weight loss actions (eating less, exercising more)

• Hard line: Heart attack or stroke

• Action(s):

  • Change health-related habits (diet, exercise, stress)

  • Start a medication

  • Retire

  • Start looking for a new job that is less stressful

Much like unathletic sports fan shit-tweeting about professional athletes, many horror fans will trash characters in horror movies making “stupid” decisions without considering other factors, like how much adrenaline is pumping through their veins or how scared they are that they are about to die a brutal death at the hands of a maniac. Sure, some (most?) stupidity is required for the plot to work and action to be had, but there’s a non-zero amount that can be explained by people being people and all the baggage that comes along with it.

So to give these critics a chance at proving just how smart and capable they are, I propose a business where customers (hereafter referred to as players) get to experience a horror movie in real life.

Choose Your Movie

Scenarios may be custom-made or based off of movie franchises such as:

• Halloween

• The Texas Chainsaw Massacre

• Friday the 13th

• Wrong Turn

• Scream

• A Nightmare on Elm Street

• Terrifier

• Predator

• Alien

Everything inside the map is based off of the scenario: costumes, environment, goals, weapons, traps, etc. Players will simply be placed into the environment with no idea what’s about to happen because movie characters never do. They simply know they must escape and there’s only one way to do so.

Goals

The goals are the same across all scenarios: either escape the area or “kill” the villain. Killing the villain is simple (see Killing the Villain). Escaping is a bit trickier because there are multiple subgoals that aren’t always obvious. For example, subgoals may be in the following order:

1. Find the key to the garage

2. Find the key to the car

3. Find gasoline and fill the car’s tank up

4. Start the car

A time limit is in place to ensure there is adequate pressure to move at a reasonable pace.

How to Win or “Die” Trying

Traps

The map may be littered with traps that the villain knows well, but is just as able to fall victim to.

The traps depend on scenario and are realistic in the accessibility sense (Jason Voorhees can’t set up a claymore behind a door because he doesn’t have access to claymores) and capability sense (Jason Voorhees can’t build a tripwire connected to a swiveling .50 BMG machine gun because he’s not smart enough).

Traps may include:

• Knife on the stairwell: As a player is going up stairs they trigger the trap, allowing a plastic knife to swing down and hit them. Hopefully they’re short or quick enough to dodge it!

• Standard trap door in the floor that only opens down (to a nice, safe foam pit). This can be on a stairway, in a hallway, or outside with shrubbery or grass placed over it. Players that fall victim to this trap may be rescued by other players.

• Anything from the Home Alone movies:

Killing the Players and Villain

Various weapons can be found around the map that interact with both the players’ and villain’s sensor-covered costumes. Each weapon has a sensor in it to allow “hit” registration on the person. Hit points vary by weapon type, intensity of use (e.g., how hard it was used), and where it hits on the body. For example, a gun headshot is an instant kill, a knife stab to the hand is X, and a sword chop to the elbow with sufficient speed is massive and renders that arm unusable while also starting a timer for that player to either stop the bleeding or die. The treatment must be effective and applicable to the wound.

Weapons are selectively available based on scenario, so not everything may be present. Potential weapons may include:

• Knife: The plastic kind that retracts into the handle

• Gun: Shoots foam darts, but very accurate

• Blunt or Sharp Object: An accelerometer measuring speed helps to determine damage

Quitting

Players who want to quit out of stress, fear, or soiled pants may yell out the safe word/phrase (preferably “This is so much more difficult than I thought and I regret my harsh words!”) at any moment, which will be heard by game staff through microphones. Lights in that area will immediately cut off to prevent knowledge gain while a staff member blindfolds and leads them out to the waiting room. The other players may continue to play if they are sufficiently far from the quitting player; too close and everyone goes on pause.

Setting

The ideal building shell for this is a large, empty warehouse where the supporting infrastructure can be built without revealing any information. Players enter into the administrative area, sign the necessary paperwork, don the costumes, get explained the rules, then head out into the arena.

Administrative

Fees and Returns

Players are charged $X to enter the game, with anywhere between 1-6 players depending on the game style. Players who survive get paid out $(X × Y × S), where Y is some multiplier determined by the business and S is the fraction of surviving players. For example, a five-player game where four players survived and the house paid 5:1 would result in a payout to each player of $(X × 5 × 4/5) = $4X. The multiplier increases proportionally to the number of players who attempt, but do not win, the game.

It’s important to note that only players who survive get paid and players who die get nothing. This encourages players to act just as selfishly as they would in real life when there’s a legitimate threat against their lives. It also incentivizes working together to ensure as many people as possible survive because it a) increases the odds of others surviving (unless there are idiots in the group), and b) increases payout amount.

Confident players can wager extra money up to a limit and multipler decided by the business when initially registering; however, the business is not obligated to accept it as they can do their due diligence on the person to either reject, counteroffer, or accept the terms. For example, Chad is extremely confident he will win, so he proposes putting down $(10 × X) at 3:1 odds, both within the Y × X and Z:1 odds limits set by the business. There is both an individual limit and group limit on the potential payouts.

Another potential idea of attracting more customers is having a jackpot that gets paid into by every losing player’s registration fee and gets paid out to the winner.

Players who quit are considered dead.

Knowledge Transfer of the Game

In order to prevent knowledge transfer, and thus unfair advantages, between different players and groups before, during, and or after the game, the following precautions are taken:

• Players sign non-disclosure agreements (NDAs) to ensure they do not provide any info of the game to others. This is a more extreme, yet necessary because of the potential payout, version of how escape rooms kindly ask you not to publicly reveal information. (Obviously this is difficult to enforce, but still serves as a deterrent.)

• Players relinquish any electronics to staff to prevent mapping, pictures, videos, audio recording, etc.

• Players who are killed are immediately removed by staff and taken into a waiting room where they are not allowed to see anything.

• Players are only allowed one attempt per year to limit rapid learning of the map and scenario. Their government-issued ID is checked against the business’s database before registration.

Liability

Players sign a waiver before playing ensuring they understand the physical and psychological risks of playing the game.

FAQs

Can’t You Just Do VR?

VR is definitely an option, but in-person fear is more intense and thus representative of the experience of characters in movies.

I consider 2011-2015 to be the golden age of street workout. There was a beautiful simplicity in the scene at that time. Teenagers, mostly Slavic, were achieving world-class levels of strength using the most basic of playground or homemade equipment. They didn’t have ultra-specialized tools or customized training and nutrition programs. I saw videos of seemingly everyday guys doing skills in their bedroom that gymnasts spend years developing and perfecting (there’s a specialization factor here, of course, but the point still stands); others were training on playgrounds between giant Soviet bloc-style apartment buildings while young children stared at them in awe. Their success was a testament to their ambition and hard work, a direct insult and contradiction to those who claim they need chalk and perfect knurling and humidity control and supplements galore and 10 hours of sleep every night to get strong. These guys were unapologetically blowing past the limits that Big Fitness had put on them.

It felt like a fuck you to the establishment. Like looking someone in the eyes offering you money, turning them down, then proceeding to do what they said you couldn’t without the money. It was so pure and focused on getting stronger, on unlocking feats of strength nobody else had. The videos weren’t fancy, just simple 360p shots compiled together to form an amalgamation of motivation and envy.

It made you feel like you, too, could achieve this. If these guys from these poor countries with suboptimal training conditions could do it, why couldn’t I, a man with plenty of time, access to nutrition, and financial support? I did achieve some amount of success. I actually got pretty strong, partially as a result of voraciously consuming these videos and trying to be like them, albeit without the subfreezing temperatures and playgrounds built for calisthenics.

The Content Creation Made Them Weak

And then the age of content creation took over. Athletes no longer shared videos for the love of the game, but the love of the money they got from peddling their course or e-book that promised the customer the magic pill that would help them achieve some skill in just a few short weeks or months! Enshittification had reached eastern Europe and poisoned the leaders.

More and more athletes were advertising their courses. More and more YouTube thumbnails were optimized for the algorithm to maximize clicks and retention. More and more videos started with “YO GUYS IT’S ME AGAIN FIRST HIT THAT LIKE AND SUBSCRIBE BUTTON AND NOW LET ME DRAG OUT THIS VIDEO FOR A BIT BEFORE SHOWING YOU WHAT YOU CAME HERE FOR” instead of just straight showing what we came here for.

And while the strength of the community continued to grow—more people were achieving Victorian crosses, more people were getting closer to the reverse planche, presses were starting lower—it still just felt...wrong, as if the original spirit of the sport was gone and replaced by some grotesque capitalistic version of its former self (and that’s coming from someone who freaking loves capitalism!).

The Event Horizon Made Me Sad and Happy

It appears as though the event horizon has been crossed for not just street workout, but other interests, too. What were once pure hobbies dedicated just to the hobby itself have become ways of extracting value from viewers.

This is sad and unfortunate. I would spend hours crawling VK’s planche and handstand pages, oogling and oggling at what the people on there were capable of. The admins of these pages would accept contributions or crawl the website for potential reposts, collecting the top pictures and videos of their page’s topic while lurkers like me enjoyed from afar. The pages are effectively dead now. I can’t tell if they’ve just migrated to another page, another forum, or if it’s actually dead, a relic of the past.

I lament everything else in the world that has fallen victim to the temptress of money and fame at the expense of its origin. I don’t know what it’s like to be financially struggling and have to choose to do this, and I’m doubtful I’d be strong enough to turn down significant money, especially if no better alternatives were in the vicinity. Heck, I don’t even know if these guys do or don’t enjoy the content creation they do now—maybe they love it and are happy to have discovered a niche that allows them to pursue their passion and get paid for it, despite the literature suggesting otherwise. (And just to be clear, the alternative of them continuing to live in not-so-great economic conditions for me to satisfy my nostalgic desires is not what I want—I want a world where their standard of living is so high that they don’t have to game the YouTube algorithm!)

And in some sense, this is a proxy of the reach of the internet and the value that people can create using it. These athletes are both creating a market and responding to it. The infrastructure and opportunities just weren’t available back then (as far as I remember), and now they are, providing at least the option to capitalize on it. And while I’m happy for them, it doesn’t mean I can’t miss the good ol’ days.

Video Repository

Golden Age Videos

Here are some videos from the golden age (GA):

GA1: Behzodbek had some of the lowest Malteses known to the community

GA2: Dmitry was one of the all-around strongest athletes; pushing, pulling, you name it

GA3: Kolya had a crazy strong front lever and one of the OG Victorian kings

GA4: Ruslan was one of the first to have a strong iguana planche

GA5: N1k didn’t have the cleanest moves, but my goodness was he powerful

GA6: Dima was known for his pulling strength on his farm (?)

GA7: Artem Z. was a pioneer of the “cool edit”

GA8: Roma Myagkov was one of the strongest vertical and horizontal pullers of his time

GA9: Just some random Kazahks havin’ fun holding their bodies in strenuous positions

GA10: Valera Kischenko was known for his bedroom rings sets

GA11: Max’s long proportions make his levers all the more impressive

GA12: The hefesto is basically a bodyweight biceps curl

GA13: Viktor is one of the top planchers of all time

GA14: Like Max, Artem M.’s long limbs make his lever strength impressive

New Age Videos

And here are some videos from the new age (NA; note that some of these are from the same creators from above!):

NA1: Same creator as GA13

NA2: Same creator as GA6

NA3: Same creator as GA14

NA4: Same creator as GA1

See Also

• Old The Best Forever playlist: A collection of the strongest street workout athletes’ videos.

Imposter syndrome is defined as:

a psychological occurrence in which an individual doubts their skills, talents, or accomplishments and has a persistent internalized fear of being exposed as a fraud. Despite external evidence of their competence, those experiencing this phenomenon do not believe they deserve their success or luck. They may incorrectly attribute it to the Matthew effect, or they may think that they are deceiving others because they feel as if they are not as intelligent as they outwardly portray themselves to be.

I identify with this strongly and attribute it primarily to comparing myself to the people I work with who are crazy smart...or knowledgable? Or experienced? Which one is it?

20 Years Got Me Trippin’

The senior member in my group has been in the same position for 22 years now, working 60-80 hours per week every week since then. In standard “work years”, that’s 33-44 years of experience. (There’s something to be said about this scaling not being 1:1, but that’s another topic.) He’s been involved in almost every single engineering decision the group has made. He has extensive history built up both digitally and mentally. He’s practiced the art of OODA loops every day, and for 20 of those years it was under the mentorship of someone he claims is leagues better than him.

Further, the beginning years were the first of a kind. All of the equipment was new—not just to the group members, but to the industry as a whole. Couple that with an undeveloped group and a high-demand manufacturing atmosphere that was expecting results last week and the result is an environment that forces learning and adaptation at an incredibly fast pace. Knowledge and experience is shoved down the throat of engineers in heaps, giving them little time to breathe, little time to digest. Yet they still did. Information was received and carefully documented. Lessons were learned and quickly implemented. Procedures were written and fanned out.

I forget these facts a lot. I forget that the level of learning during that time was exponentially more than is available to me now. I forget that necessity is the mother of invention. I forget that knowledge begets knowledge, that identifying the correct location of four puzzle pieces with an empty space between them can automatically reveal the missing piece’s structure and color. I forget that I only see a small portion of the daily behind-the-scenes happenings. I forget that he was responsible for creating a majority of the documentation, and in doing so learned the material extremely well.

Experience has a significant impact that most others often fail to give credit for, instead attributing it to just being smart.

Experience-Intelligence Relationship

This brings me to my thesis statement: while a minimum level of intelligence is required to operate in certain environments, sheer experience can make up for lack of it. I view this as a very simple equation of:

y = -x + C

which only exists in a portion of the first quadrant below. The y-axis is experienced needed, the x-axis is intelligence, C is some arbitrary constant offset, and the person must be above the line in order to operate effectively. In other words, less intelligence is needed if one has more experience and more intelligence is needed if one has less experience. The larger the C value, the more intelligence is needed for a given experience level or more experience needed for a given intelligence level.

In the wise words of Kant:

Experience without theory is blind, but theory without experience is mere intellectual play

Quick on the Draw

Vast experience in a topic allows for patterns to be recognized more easily and heuristics to be developed and applied more successfully. Something happening (call it problem P1) that is ever so slightly different than what happened with problem P2 a year ago is still a useful clue into the root cause. Combine this recognition with a sharp memory and/or detailed documentation and it’s hard to not solve the issue (and solving the issue at all signals some amount of intelligence). As P3, ..., PX occur, heuristics for finding root cause can be refined and updated, leading to faster or clearer resolution.

Habits matter here. Instinctively applying heuristics at the onset of an issue prevents time from being wasted with fruitless troubleshooting and gets directly to the core issue. The more times the heuristic is applied, the more instinctive it becomes, the faster the problem is solved, and the more intelligent the decision or finding appears.

Conclusion

A few takeaways and overall points after putting these thoughts to paper:

• Experience requires a minimum amount of intelligence

• Experience can create the illusion of intelligence

• Focusing on obtaining knowledge can quickly enhance intelligence (or the appearance of)

• Intelligence is not a magic pill for lack of experience nor vice versa

See Also

• You Are NOT Dumb, You Just Lack the Prerequisites

Background

In 6th grade I discovered the North Shore, Vancouver’s coveted mountain biking area comprising Mt. Fromme, Mt. Seymour, and Cypress Mountain. For whatever reason, Fromme was what intrigued me the most and what has stuck with me to this day. The trails were legendary and awe-inspiring, even by today’s standards using today’s technology. Guys were hucking the craziest of stunts on hardtails and 26” wheels. Everything was experimental and tested the limits of both the bikes and riders.

And yet sadly, there are few places out there that compile all the materials on a single trail in a single place. Some of it’s on Pinkbike, some on YouTube, some on Facebook, some in people’s garages on hard drives gathering dust. This is my attempt to gather as many materials as possible in a single place for reference and posterity.

If you have more materials that aren’t shown here but you think should be, please contact me and I’ll add them.

Bounties

I offer what I consider generous bounties for information or material (videos) related to each trail. Expand the trail sections to see the payout and requirements. Please contact me with the bounty requirements attached in order for me to review.

I will not release any bounty information or material to the general public unless the information provider explicitly approves it.

GMG

Bounty information:

• Payout: $1000

• Requirements: An uninterrupted, unpublished (i.e., I can’t find it anywhere on the internet), top-to-bottom POV run, including all stunts and features. I will prorate the payout if it’s a partial run that I deem novel. The video must be downloadable.

• Notes: I suspect that there are existing, high-quality POV runs for the upper section of the trail. For the lower, I wouldn’t be surprised if the NSX guys had uninterrupted footage, although I don’t know what the helmet/body cameras were like back then.

GMG stands for Gabe (Szczurek), Mike (MacGregor), and Gerry (?), the three guys who built the trail.

Spectrum Techwear 2007 video (GMG starts at 5:12)

Mountain bike YouTuber Dangerous Dave takes on OgoFloGo (start at 0:17):

NSX 3 (start at 36:15):

NSX 4 (start at 36:29):

NSX 5 (start at 6:24):

Some random guy from a Facebook video:

The Flying Circus

Bounty information:

• Payout: $500

• Requirements: An uninterrupted, unpublished (i.e., I can’t find it anywhere on the internet), top-to-bottom POV run, including all stunts and features. I will prorate the payout if it’s a partial run that I deem novel. The video must be downloadable.

• Notes: N/A

Link: The Flying Circus on The Trail

Mr. Skinny

Bounty information:

• Payout: $300

• Requirements: An uninterrupted, unpublished (i.e., I can’t find it anywhere on the internet), top-to-bottom POV run, including all stunts and features. I will prorate the payout if it’s a partial run that I deem novel. The video must be downloadable.

• Notes: N/A

Disconstructed Wheel

Bounty information:

• Payout: $100-1000

• Requirements: An exact latitude and longitude of the location of the Wheel, even if it’s not still standing. Proof in the form of a video is required and can be verified using the videos. The payout will be less if the wheel is not still standing and more if it is.

• Notes: As far as I can tell, the Wheel wasn’t built specifically for NWD—it appears Sam built it before the film. That being said, there’s almost literally nothing on the internet that I can find on it. Pictures are sparse, there’s one video from NWD, and only rumors of its location. I even reached out the Sam’s sister to see if she knew anything, but never heard back.

Comments about the location:

rideunited (Apr 29, 2010 at 6:42) The wheel trail is in Nelson B.C.

footplant (Jun 5, 2008 at 13:04) Blue river BC.... its the hampsterwheel....if thats not blue river... then..... its a new one else where

Coastal Crew: Kranked 8

Bounty information:

• Payout: $500

• Requirements: An uninterrupted, unpublished (i.e., I can’t find it anywhere on the internet), top-to-bottom POV run, including all stunts and features from both lines. I will prorate the payout if it’s a partial run that I deem novel. The video must be downloadable.

• Notes: As far as I can tell, the trails were built specifically for Kranked 8 on the Coastal Crew’s property since they’re (mostly) not featured in any other videos. In the second video below the CC says “we have a dialed zone where we had built some unique lines in for our segment in Kranked Revolve. We as riders, builders and filmmakers felt that these lines deserved more than just a handful of shots in a movie.”

Extra Bounties

The following trails were probably built just for the scenes they were used in, but that doesn’t mean the remnants of them aren’t still remaining deep in the woods.

First, Wade Simmons’ NWD 7 one-shot segment:

Bounty information:

• Payout: $100-200

• Requirements:

  • $200 for an uninterrupted, unpublished (i.e., I can’t find it anywhere on the internet), top-to-bottom POV run, including all stunts and features from both lines. The video must be downloadable.

  • $200 for an exact latitude and longitude of the location of the trail, including access instructions, even if it’s not still in rideable condition. Proof in the form of a video is required and can be verified using the NWD video.

• Notes: N/A

And second, the Wade Simmons’ et al.’s NWD 8 segment:

Bounty information:

• Payout: $100

• Requirements: $100 for an exact latitude and longitude of the location of the trail, including access instructions, even if it’s not still in rideable condition. Proof in the form of a video is required and can be verified using the NWD video.

• Notes: N/A

Background

I’ve been a competitive person for most of my life, but with no outlet to express and develop it. The rest of my family is fairly lax about most parts of their life, and the areas I was competitive in didn’t overlap with them. The same goes for my friends. None of them took seriously and competitively what I took seriously and competitively.

Having a community like this is ultimately detrimental—or at least limiting—to personal development and progression. (Of course, it’s still important to have family and friends who are not competitive in the same areas as you.) Without some overlap, development is or becomes limited by the individual; this can only take one so far. A partner is required to make progress at the higher levels in specific areas. Competing against oneself offers little challenge and little reward besides more self-discovery (which is not a bad thing!). Competing against others forces challenge and reward in the pursuit of achievement and victory, continually edging each person up little-by-little.

Benefits

Information

A core philosophy of mine is that feedback and learning from others is almost always valuable: if the advice or takeaway is beneficial, then it can be implemented later, personalized, or built upon based on situation at hand; if it’s detrimental, that’s a lesson learned (hopefully in an easy way) that won’t be used in the future and can still possibly be changed for the better; if it’s neutral, it does no harm but still provides information about what does and doesn’t work. Information is gained and future selves benefit in every case. Competition is the same way:

• Win: Winning handedly shows what works and in some cases, what doesn’t for others (average preparation by the winner and poor preparation by the loser still leads to a win); barely winning reminds the winner of the role luck plays or what skills still need to be developed.

• Lose: Losing handedly shows what needs to be worked on (either work on what the winner did well or what the loser did poorly); barely losing reminds the loser of the role luck plays or what kills still need to be developed.

• Tie: Tying shows what needs to be worked on.

Again, information and a plan of action are learned in every case. A few examples:

• Tony and Christopher are in the front pack of the mile race with no one else around. In an effort to mentally break Christopher, Tony surges at the halfway mark, dropping Christopher and establishing clear dominance over the field. In the final kick, Tony struggles and narrowly beats Christopher. Tony learns that the mid-race surge can work, but that his kick is weak. He adjusts his training to improve his endgame. Christopher learns he is psychologically weak and has trouble dealing with discomfort during races. He adjusts his training to include more mental sessions.

• Carmella and Paulie are the top salespeople at their company, both in the running for employee of the year, which is primarily based on total number of sales. Paulie is close behind with a few months to go, but Carmella’s lead grows significantly in the final weeks of the year. Paulie reflects on his lackluster performance and finds he doesn’t perform well under pressure and with a lot on the line. He keeps this in mind for the next year by focusing on bigger accounts and not worrying about others. Carmella notes how stressed Paulie got and sees the same weakness, albeit to a smaller extent, in herself. She does the same thing as Paulie by focusing on relaxing under pressure and doing what she does best: closing sales.

• White’s queen has managed to survive, while black’s king is alone on the board. After many moves of dancing, black forces a stalemate to the dismay of white. White learns his endgame is poor and needs to be worked on; black learns his endgame is decent, but takes away from his earlier blunders and the lucky-to-him mistakes white made in haste.

Those examples are only for intra-discipline competitors, but learning opportunities are abundant between inter-discipline competitors, i.e., competitors from one discipline can easily learn from others in a different discipline. Carmella can learn from Tony about making a big push in the middle of an extended effort, as well as finishing strong when other competitors are close behind. Paulie can learn from Christopher about maintaining mental strength despite being behind and the other person appearing to be comfortable with the current pace. There is a lesson to be taken from everything.

Motivation

In some runner’s training journal (I have searched and searched, but cannot find it; I think it was Cerutty or Kellogg), he mentions that both winning and losing result in motivation. Winning affirms training and the preparation that taken place for the big day. Losing produces frustration/anger and a desire to do better next time. There are two special circumstances to be aware of, one for each outcome:

1. Continual winning fosters a false sense of security. The winner starts to think they can’t lose. They get lazy, competitiveness decreases, and suddenly their rank is slipping. Some can muster the momentum to start training sufficiently again, while for others it may take a bit of time. The simple way to prevent this to understand the fact that at any time someone can become better, whether it’s due to their work or a personal fault.

2. Continual losing causes distrust of the training and a general lack of motivation. Imagine training hard (reminder: hard ≠ better!) and race results, either in place or performance, not showing for it. The initial thought is that hours, days, months, even years of hard work have been for naught. No one likes feeling this way, so care should be taken to avoid continual losing, especially to the same opponent or a major rival. (Again, dosing is everything here.)

Outside of those two cases, the statement is generally true. Winning should be experienced more due to the impact it has on one’s psyche, but losing is important in its own right, too. More on implementing winning and losing in the Implementation section.

Signaling

Being known as a competitive person has its advantages. Getting picked over others is more likely when the position requires someone who will do whatever it takes to win. Why would a company choose inept David for that director role when they know Frank hates losing, whether it’s a sale or chili cooking contest? The choice is obvious.

Being competitive often benefits those on the same team. Spouses and children will benefit from their working spouse’s/parent’s competitiveness in the form of raises; players in the form of wins and lessons; companies and stockholders in the form of productivity; employees in the form of more business and more raises. Learn to control competitiveness based on the situation and watch popularity skyrocket.

Detriments

Unhealthy

As mentioned above, being ultra-competitive and losing too often can wreak havoc on one’s mental health. Not good enough, not training hard enough, continual bad days, etc. All reasons for losing. All ways to lose self-confidence or try to overcompensate in an unhealthy manner.

Becoming okay with losing is an integral part of competition. It allows for clarity in the mind after a loss, giving the loser time to mull over his successes and mistakes and create a plan to address them. Agitation and anger cloud the mind and often cause bad decisions to be made.

In rare circumstances, competitiveness can ruin reputations and even friendships. There is a time and place for competition—pickup basketball at the local church with some randoms is probably not that. Being known as “that guy” does no favors.

Individuals

Mark Twight makes note of men’s competitiveness in the gym, stating something along the lines of “put a score on a guy’s [yes, women are intentionally excluded from this sentence] performance and he will work his ass off”.

And it appears there are documented differences between the sexes. Cashdan’s Are men more competitive than women? investigates:

whether women and men differ in (a) what they compete over, (b) whom they compete with, and (c) their competitive tactics, including use of aggression.

The results are unsurprising (following taken directly from abstract):

In Study 1, university students kept diaries of their competitive interactions during the term. Sex differences, few overall, were as follows: (a) men’s diaries contained more same-sex competition, (b) women competed more about looking attractive whereas men competed more about sports, and (c) men used physical (but not verbal) aggression more frequently than women. In Study 2 strength of competition was also measured by questionnaire. Women and men felt equally competitive overall, but men felt more competitive about athletics and sexual attention whereas women felt more competitive about looking attractive. In men, but not women, competitiveness for financial success was correlated with various aspects of mating competition. Young men were more competitive than older men in a variety of domains and were also more physically and verbally aggressive, but no age difference in aggression was found for women.

From Niederle and Vesterlund’s Gender and Competition:

Laboratory studies have documented that women often respond less favorably to competition than men. Conditional on performance, men are often more eager to compete, and the performance of men tends to respond more positively to an increase in competition.

I believe that men and women’s differences in competitiveness (both in discipline and intensity) is largely determined by what society and peers (read: competitors) find valuable. Men are more competitive in sports (excelling in a sport is attractive) and sexual attention (being sought after by many women implies value). Women are more competitive in looking attractive (no explanation needed). The relatively recent rise of women achieving higher statuses in the workplace also corresponds to an increase in competitiveness. Career women are now found more attractive than ever, giving women an incentive to become more competitive in the workplace.

And like Robin Hanson has said, a vast majority of competitiveness is likely based on simple signaling.

Implementation

Development

Developing competitiveness for those who either a) don’t have it, or b) want more of it is a fairly straightforward process:

1. Learn and appreciate how fun it is to win

2. Learn and appreciate how un-fun it is to lose and the shame that comes with it

These are two high-level points that still encompass all of competition. How these are done is a personal choice, but some suggestions/thoughts:

• Success can be both relative and absolute. Getting last place in a high-level event is still last place (bad), but still at a high-level event (good). Choose which to focus on. Hint: it’s the former.

• Shame is a powerful motivator. Ashamed about lack of performance? Get better. Ashamed about losing to someone who barely trains and relies solely on natural talent? Get better. The list goes on and on.

• Look at what winning can result in, both materialistically and intrinsically: attractive partners, high salaries/bonuses, reverence, submissiveness, self-confidence. If any of these are desired, winning must happen.

• Look at what losing can result in, both materialistically and intrinsically: settling for an unattractive partner, lack of promotion, disrespect, apathy, lack of self-confidence, shamefulness. If any of these are to be avoided, winning must happen.

• Capture thoughts and feelings as vividly as possible directly after all wins and losses. Remember how intense the feeling of [blank] was and use it to fuel future training.

Community

Again, others are a requirement for increasing competitiveness. Finding them is a simple process:

• Does the discipline have some type of Schelling point, a place where competitors are bound to meet? For exercise the Schelling points are gyms, for gamers it’s tournaments, etc. If so, make it a point to go and meet like-minded people.

• Post advertisements online in relevant places: subreddits, forums, etc.

• Talk around with friends and acquaintances and learn about their interests. They may have something in common they’ve never spoken about.

It’s important to remember that a vast majority of people want to improve. No one wants to regress or stagnate in their skills. Making it obvious that they will get better by partnering up will encourage them to do so—a mutually-beneficial relationship.

Feedback

Once competitiveness is at its desired level and a community is found, it is time for training. Implementing feedback from others and reviewing training is a major part of the development process, but is often neglected or never used.

The feedback process is flexible, but should roughly follow this outline:

1. Find a partner who is of a similar skill level and drive.

2. Define goals

3. 1. Train

   2. Review training with partner on regular basis

   3. Implement feedback into training

      1. If goal has been reached, return to step 2

      2. If goal is still in progress, return to step 3a

Design

The win-loss ratio is a personal number, but should not be skewed too heavily towards one side. As mentioned in the Motivation section, too much winning can breed over-confidence and laziness, while too much losing can decrease self-confidence and create disdain for the activity.

A win-loss ratio of (6-7):(3-4) seems appropriate, i.e., out of 10 events, wins occur six to seven times and losses three to four. When in doubt, err on the side of a larger ratio, i.e., more wins. A handful of both wins and losses should be close and possibly due to (bad) luck, while others should not be. Close wins/losses produce hunger, easy wins produce confidence, and big losses (should) produce self-reflection.

Programming this ratio requires knowledge of opponents’ skill levels. From here opponents can be picked accordingly. Close wins/losses should come from someone equally as skilled, if not a bit more. Win and confidence is boosted since a superior player was just beat. Lose and it’s understandable, they were better. Easy wins should come from worse players (obviously) and big losses from way better players (obviously). Again, take lessons from all competitions.

While I wrote this in 2020 and some advances have been made, the principles remain the same.

Physics

First, why are ions shot into silicon (Si)? What does it do from a physics perspective?

The reason has to do with the band gap, the energy difference between the valence band and conduction band.

Within Si exist both electrons (negatively-charged) and electron holes (positively-charged). The hole is a bit of an oddity: it’s an unoccupied space where an electron could exist, but doesn’t at that moment. While not literally a particle, they can be treated and thought of as such. Current can be viewed as the flow of electrons or holes. When a bias is applied, electrons begin to move and occupy holes. When electron A moves to hole B, hole A is formed. This propagates and creates the flow of charge, or current.

What does this have to do with Si? Si has four valence electrons in its four states of 3s²3p². Those four valence electrons covalently bond with the surrounding Si crystal lattice structure, effectively filling that Si atom’s valence shell. So, when an ion is introduced into the lattice that does not have four valence electrons, one of two things happen: there are free electrons (if the number of valence electrons is >4, such as in phosphorus’s case) or free holes (valence electrons <4, such as boron). In the first case, four of phosphorus’s (P’s) electrons have bonded to surrounding Si atoms’ valence electrons, leaving the fifth one free and able to contribute to current, making it a donor impurity, because it donates an electron. This creates an n(egative)-type semiconductor. In the second case, three of boron’s (B’s) electrons have bonded to surrounding Si atoms, but one Si atom and the B atom both have incomplete valence shells (7/8, 6/8 filled, respectively). To fill the vacancy, the Si and B will pull a free electron from the lattice, leaving a hole in its place, making it an acceptor impurity, because it accepts an electron. This creates a p(ositive)-type semiconductor.

The Fermi energy level, E_F, is the energy level at which the probability an electron exists there is 0.5, based on the Fermi-Dirac distribution:

E_F for an intrinsic semiconductor (no impurities added) is given by:

where E_i is the intrinsic Fermi level, a constant, and N_C and N_V are the effective density of states. (Note that the link only shows you density of states, g(E), not effective DOS. The effective DOS derivation can be found here.)

While E_i cannot change, E_F can based on the number of ionized impurities present in the substrate. When an n-type dopant is added, E_F shifts up towards the conduction band, decreasing the E_C−E_F gap. Similarly, when a p-type dopant is added, E_F shifts down towards the valence band, decreasing the E_F−E_V gap. Because there are now electrons sitting at the new n-type E_F level, even a small amount of thermal energy can excite them into the conduction band, creating current. For the p-type, electrons can also be excited to E_F, leaving holes in the valence band, creating current.

It’s this change in the E_F value that drives modern-day devices (admittedly, some other things help, too). By combining two substrates—one n-type and one p-type—and creating a p-n junction, both of the E_F levels must match up. The p-n junction is the basis for transistors, proving the importance of ion implantation in devices.

Dose vs. Concentration

Two common terms used are dose and concentration. They are similar, but not synonymous. Dose is number of ions per unit area (cm²), while concentration is number of ions per unit volume (cm³).

Given beam current, I (charge/time), and time spent implanting, t, the total amount of implanted charge is simply Q=It. Dose, ϕ (number/area), can be multiplied by the elementary charge, q, and scan area, A, to also get the total implanted charge. Integrating the current with respect to time can also calculate dose, which is actually how the tool itself calculates dose (see dose measurement). Equating the first two and solving for dose gives:

Dose can also be found be integrating the concentration, N(x), with respect to depth:

The concentration of an implant can be approximated if the range, R_p, and straggle, ΔR_p, are known (range and straggle values are experimentally-determined and based on energy):

The N(x) equation above is a simple Gaussian function, but is not entirely representative of how actual implant profiles look. To better simulate real profiles, the Pearson IV distribution and moments are used, which accommodate for the skewness that often occurs.

The respective moments are as follows:

1. R_p: Average depth of an implanted ion from the surface of the substrate

2. ΔR_p: Variance of the depth.

3. γ: Skewness, or how much the profile leans to one side (on a distance x-axis, meaning towards or away from the surface). A distribution that is skewed left (long tail on left side) has negative skewness (higher concentration towards the surface), while a distribution that is skewed right has positive skewness (higher concentration deeper in the substrate). A perfect Gaussian distribution has no skewness because it is perfectly symmetrical.

4. β: Kurtosis, or heaviness (or fatness) of the tails.

The integral limits are from 0 to ∞ because a negative number would not make sense when referring to depth. Using all of those, the Pearson equation can be solved and specific values substituted in to the IV solution. For specifics, see pages 8-12 of Dr. Alan Doolittle’s lecture.

A typical concentration vs. depth Gaussian curve is shown below:

The concentration profile makes intuitive sense, especially if you rotate the image 90° and superimpose the mask and substrate onto it:

It is seen that the peak concentration occurs about halfway through the SiO₂. The concentration steadily decreases from there as it exits the SiO₂ and enters the Si. As seen in the graph above for range and straggle vs. energy, increasing energy will result in deeper implants and a shifting of the Gaussian curve to the right, while also flattening it due to increased straggle.

Junction Depth

The junction depth, x_j, is the depth at which the impurity concentration equals the substrate concentration, or:

This is the location of the p-n junction, where N_A=N_D. To find this depth, simply find the substrate concentration, trace a horizontal line on the concentration vs. depth plot, find the intersection of the two, then simply draw another line vertically down to the x-axis. This is the junction depth.

Desired junction depth can be achieved by choosing the energy and species based off their range and straggle values:

Based on this, there can be two junction depths if a more complete Gaussian curve describes the concentration and/or the substrate concentration is high enough (N_sub>N_imp(x=0)).

One rudimentary way of calculating junction depth is the groove-and-stain method. After grinding a shallow cylindrical groove into the substrate surface, a dye is applied that only reacts with one type of present impurity, staining that area. There will then be two distinct areas: one dyed and one plain. The junction depth can then be found using basic geometry:

Stopping

Once the ions enter the lattice, they have to stop somehow. Two mechanisms cause stopping: nuclear stopping and electronic stopping. The stopping power is then just the sum of both the nuclear and electron stopping components, measured in energy loss per unit path length of the ion:

More detailed derivations can be found here.

Nuclear Stopping

Nuclear stopping is caused by the collision of the ion into the substrate’s nuclei. If the incoming ion’s energy exceeds the substrate’s displacement energy, the substrate atom will break its four covalent bonds and become displaced from its original position. Depending on the ion’s initial energy, a collision cascade can occur: both the ion and first-displaced atom will knock into more atoms and displace those. This continues until both the ion energy and last-displaced atom energy are less than the displacement energy.

Heavier atoms, such as As⁺, will slow down primarily because of nuclear stopping. Their collisions with substrate atoms are more elastic, so energy loss will be less than the electronic component. For low energies (low velocity), nuclear stopping is the primary mechanism up to a certain point, but electronic stopping will eventually take over.

Nuclear stopping damages the lattice (even causing sputtering) and calls for an activation anneal.

Electronic Stopping

Electron stopping is caused by the interaction between the ion and both bound and free electrons (electron cloud) within the substrate. Given that all implantable ions are cations, they will all react similarly to electrons. The ion continually slows down due to the attraction to/drag of the electrons, eventually stopping.

High velocity ions slow down primarily due to electronic stopping. Their collisions with substrate atoms are highly inelastic. Electronic stopping dominates at moderate energies.

The approximation for electron stopping power is:

where Z₁ is the ion atomic number, Z₂ is the substrate atomic number, a₀ is the Bohr radius, N is the substrate atomic density (atoms per unit volume), and v is the ion velocity. (I can’t seem to find a source on this. Seems to be from one of Lin53, Lin61a, Lin61b from this references section, but even digital copies don’t seem to exist.)

Activation Anneal

Because of the damage caused by nuclear stopping, the lattice must be repaired to improve conduction and integrate the impurities into the lattice. The defects that result are primarily vacancies and interstitials. In a vacancy defect, a lattice point is missing an atom: this is caused by either the ion or substrate atom collision (as in the collision cascade). However, this missing atom leaves a place for an implanted ion to become part of the lattice structure, thus “activating” it (the activation energy has to be met). Interstitials refer to atoms that are present in the lattice, but not at a normal lattice point.

Annealing is a high-temperature process, and with high temperature comes movement from the atoms. The previous Gaussian estimation can then be modeled by including the ion diffusivity multiplied by the anneal time in the straggle terms:

While skewness remains 0, straggle and kurtosis increase as the curve flattens out.

Equipment

Sources

The most common dopants are boron, phosphorus, and arsenic. Aluminum and antimony are occasionally used, but their high diffusivity pose issues for devices.

Gases, such as AsH₃, PH₃, and BF₃, are used to form the ions. Gas is flowed into a small chamber that contains a basic thermionic filament. Electrons are emitted and collide with the gas, ionizing it and forming positive ions within the chamber.

Ionization is the process of adding or removing an electron from a neutral atom, e.g., F becomes F^-. In elements that lose an electron, the ionization energy is the amount of energy required to remove one electron from the atom’s valence band. To remove the first of He’s two valence electrons, 24.6 eV (eV is electronvolts, the amount of energy produced by accelerating one electron through one volt). Ionization energy increases for each valence electron removed. For example, to remove Ne’s eighth valence electron requires 239 eV, compared to 21.5 eV for its first. For an implant gas like arsine (AsH₃), around 15 eV is required to dissociate the molecule into a As⁺ cation (positively-charged ion).

Multiple ions can form from the same source gas. For example, BF₃ may ionize into ¹⁰B⁺, ¹¹B⁺, ¹⁰BF⁺, ¹¹BF⁺, F₂⁺, ¹⁰BF₂⁺, and ¹¹BF₂⁺. To find the relative percentage of each in the plasma, the gas spectrum, which shows the beam current vs. atomic masses, needs to be analyzed.

To increase the probability of ionization, magnets are used to make the electron follow a helical path from one end to the other, rather than just a straight line from anode to cathode.

This continuous ionization of the source gas forms a plasma within the ion source chamber. A vacuum line connected to a vacuum pump is connected to maintain millitorr levels of pressure. Once there is a sufficient amount of ions to form the beam, a large voltage is used to extract the positive ions from the chamber and direct them towards the mass analyzer.

Safety

Most source gases are highly toxic and carcinogenic. See PEL/TLV data for almost all implant gases here.

Mass Analyzer

There are two common types of mass filters: the magnetic sector and radio frequency quadrupole (RFQ). The magnetic sector uses a magnetic field to curve the various species orthogonally from their original path, using the equation:

Where B is the magnetic field strength, m is the species’ mass, V_ext is the extracting voltage, v=2qV_extm is the ion velocity, and q is the species’ charge. Depending on the species’ mass, some may go less than, equal to, or greater than the desired angle.

While the magnetic sector is a relatively robust method, it does pose issues. Take a P⁺ cation and a diatomic O₂⁺ cation. P⁺ has an atomic mass of 30.9738 amu (an electron has a mass of 0.0005 amu, which is negligible). O₂⁺ has an atomic mass of 31.9988 amu. Given a constant extraction voltage and magnetic field strength, the differences in radii will be only 1.61%. This minute difference can cause oxygen to be present in the beam, possibly ruining the implantation.

Acceleration

After the desired ions have been sorted from the undesired, they are accelerated, collimated, and condensed into the final beam.

The acceleration column is kept under as high of vacuum as possible to minimize the amount of stray particles the beam hits, which can cause it to be deflected. The final ion energy is a simple calculation:

Graphite apertures are commonly used in forming the beam. By designing slits correctly in successive apertures, parts of the beam that stray out are eliminated, leaving only the most parallel ions. They are replaced as needed.

Endstation

The endstation is where the actual implantation takes place. Because each substrate has millions of places (read: transistors) that need implanting and the ion beam is not large enough to cover the entire substrate uniformly, scanning needs to occur. (Sometimes the term rastering is used synonymously, but that is a specific type of scanning. “Scanning” will be used to avoid confusion.)

Beam Stop

Because the magnetic sector only distinguishes based on mass and not charge, some neutrals may be a part of the beam. In order to prevent them from being implanted, some implanters include a small bend at the end of the acceleration column, where an electric field deflects the charged particles one way while the neutrals continue on to nowhere.

Mechanical Scanning

In mechanical scanning, the substrate is attached via clamping or centrifugal force to a robotic arm and moves around a stationary beam. The substrate holder also needs tilting capabilities, otherwise the tilt angle will be 0° (beam is naturally orthogonal to substrate). The primary advantage is a constant tilt angle to ensure consistent implanting across the wafers.

Mechanical scanning is preferred for high dose implants, which use high beam currents.

Electrostatic Scanning

In electrostatic scanning, variable electric fields in both the x and y direction scan the beam across a stationary wafer. This requires a narrow beam, which is generally given by low dose and low current implants. Electrostatic scanning is avoided on high current tools because the beam tends to expand with higher currents, causing poor scanning control. The tilt angle will also vary as the beam scans over the wafer, causing channeling and uniformity issues.

Dose Measurement

To actively measure the dose, a Faraday cup is used. Using the equation from the dose vs. concentration section, the dose is able to be found using an integrator:

Plasma Flood Gun

When positive ions strike an insulator, e.g., SiO₂, an electric field can form. If this field becomes large enough, the device can be damaged or even ruined altogether. To prevent this electric field from forming, the positive ions need to be neutralized simply by adding an electron. To this, a plasma is struck using the same method as in the ion source: a filament emits electrons that ionize the nearby gas(es). Ar is commonly used. The plasma, being filled with electrons, diffuses into the beam path and the electrons combine with the positive ions directly before they reach the the substrate.

Varian, arguably the leader in ion implantation technology, has published a paper explaining the plasma flood gun and its operation here.

Vacuum

High vacuum is required to prevent beam deflection or contaminants being implanted. This paper discusses vacuum system design for ion implanters.

Process Considerations

Implant Masks

Masks are used to block certain areas from being affected by a process. In ion implantation, a mask is used to prevent specific areas of the substrate from being implanted. After all, the source and drain of MOSFET wouldn’t necessarily be a source and drain if everywhere around it was doped the same.

So, how to decide which mask to use? A few things about the mask must be considered: cost, stopping power, adhesion to the substrate surface, compatibility with the substrate itself (for example, Cu diffuses very quickly into Si, which can pose later issues in devices). Amorphous (non-crystalline) and high-density materials are best due to larger stopping powers. Common masks include SiO₂, Si₃N₄, photoresist (specific science here), Ti, and W.

Masks can also be used when forming shallow implants. Simply by knowing Rp and ΔR_p of the mask and substrate, the energy can be adjusted to push the ions completely past the mask and only a small depth into the substrate. This will take trial-and-error, as the ion energy as they leave the mask is unknown.

Photoresist Masks

Photoresist masks should be chosen based on a few parameters. First, the temperature at which it softens. If being used at a mask, it is likely to have already been post-develop-baked (also called hard bake). Any temperature above the hard bake temperature may cause reflow, which can easily ruin the entire process. There are multiple ways to avoid this:

• If possible, choose a resist with as high of a softening temperature as possible.

• Improve cooling to the tool’s (where the implanting actually occurs). Instead of using deionized water as a coolant, a mixture of ethylene glycol and water or simply pure ethylene glycol may work better.

• Choose a metal or dielectric mask that won’t have reflow issues. This will add more steps (deposition, etching, photoresist removal) in the fabrication process, but is much better than a completely-ruined process.

High energies can cause strong cross-linking of the resist, making removal more difficult than normal. However, because high-energy ions don’t make it to the bottom of the resist, a lift-off process can be performed to more easily remove the strongly-cross-linked parts of the resist.

Dielectric and Metal Masks

If the photoresist mask has been ruled out, the other two options are dielectric, such as SiO₂ or Si₃N₄, or metals, such as Ti, W, or silicides of either (TiSi₂, WSi₂). Range and straggle of the implanted species into the selected material should be chosen. Denser materials are obviously better masks.

Other materials have been analyzed using a range of energies and species:

Using data like this, both the mask thickness and material can be chosen based on the species and energy.

Channeling

Channeling occurs when the ion beam is angled such that there are few obstacles, in this case, atoms, in its path. Take B⁺. Its atomic radius is around 85 pm (”p” being pico, or 10^-12). Viewing the diamond cubic structure of Si from the [110] direction provides very wide openings, or channels, for the ion beam to penetrate. Ions can travel extremely far into the lattice if the beam angle is orthogonal to the axis.

There are a few methods to prevent channeling:

• Tilting: By tilting the substrate slightly, the beam is no longer orthogonal to the substrate, causing the ions to collide with the atoms and stop more easily. A common tilt angle is 7°.

• Surface amorphization: Depositing a thin layer of amorphous material will both slow and scatter the ions once they pass through.

• Substrate amorphization: Implanting inert species, such as Ar, just below the substrate surface.

The mask height and channel width must be considered when tilting. Basic trigonometry requires a tilt angle no larger than:

In practice, this limitation is essentially negligible. The mask (generally photoresist or a dielectric like SiO₂/Si₃N₄) is between 100-1000 nm and the channel width to be implanted is 10-2000 nm (no sources on either of these despite combing through Google Scholar, but I’ve heard these numbers thrown around). These give minimum and maximum angles of:

Again, these numbers are extreme. No manufacturer will be creating 10 nm source/drain widths with 1000 nm mask heights, nor 2000 nm S/D widths with 100 nm mask heights. More appropriate numbers would be 100 nm mask for 10 nm width and 1000 nm mask with 2000 nm width, giving:

Additionally, there will be a shadowing effect if any tilt angle is used.

To fix this, a twist angle is added. The twist rotates the substrates so the shadowed region will be exposed to the ion beam. A “quad” implant may also be performed by implanting at 0, 90°, 180°, and 270°.

To find the minimum angle (critical angle) to prevent channeling, the following equation is claimed (1, 2, 3):

where Z_ion/sub are the ion and substrate atomic numbers, E₀ is the ion energy in keV, and d is the substrate atomic spacing along the ion path in angstroms. However, the sources that use this equation provide no references to where this came from, nor do the square root argument’s dimensions make sense. For this equation to provide units of degrees, the argument needs to be unitless, but it’s not:

This equation has also been claimed for critical angle:

As seen, the dimensions do not cancel out, so neither equation holds in that regard. Dr. Stephen Campbell (author of Fabrication Engineering, one textbook where this equation is used) said via email:

That equation came from the first edition of the book, now more than 20 years ago. As I remember it, it was strictly empirical. You will find a few equations like this in the book. The truth about this field, however inconvenient, is that the development of detailed understanding of fabrication processes often lags a decade or more compared to the development of these processes. I don’t know of a theoretical analysis of this topic.

(If you have an original source for this equation, please contact me.)

Uniformity

Modern implanters have both good angle and dose uniformity (see “Representative Ion Implantation Systems” section here): less than 0.1° and one σ < 0.5%. The dose uniformity is primarily a function of the dose measurement system, which can adjust in real-time for any errors that occur (beam turning off, spiking in current, etc).

Doping Alternatives

While doping remains a necessity for device fabrication, ion implantation is not the only method of creating n- or p-type regions.

Diffusion Doping

The other primary method of doping is through diffusion via a furnace. The substrate is heated up and a gas source is introduced into the furnace (chamber or tube). The gases used for each common implant are POCL₃ for P, B₂H₆ for B, and AsH₃ and C₄H₁₁As (TBA) for As. Using POCL₃ as an example, it reacts with O₂, to form gaseous P₂O₅. P₂O₅ then reacts with the Si surface to form phosphosilicate glass (PSG), (P₂O₅)_x(SiO₂)_1-x, which serves as the P diffusion source. (The other source gases also react to form a glass layer on Si.)

There are two steps in the diffusion process:

1. Pre-deposition: The source gas is flowed into the chamber along with a reactive gas (generally O₂), forming an oxide and finally the glass layer. Some dopants diffuse slightly into the substrate.

2. Drive-in: The source gas is turned off and the dopants are allowed to diffuse into the substrate.

Diffusion is based on Fick’s laws and increases with temperature:

where D₀ is the element’s pre-factor (temperature-independent).

Spin-On Doping

Spin-on doping (SOD) is the least-common method of doping a substrate. A liquid source is applied to a substrate, then spun at a thousands rpm for a few seconds. This (ideally) provides a uniform film across the surface. The substrate is then heated to allow the dopants to diffuse.

Specific application notes and product information can be found here.

A garage-fab SOD process can be seen here.

See Also

• Lecture by Dr. Alan Doolittle of Georgia Tech.

• Applied Materials’ lecture series on ion implantation: Part 1, Part 2.

• Lecture by Dr. Nathan Cheung of U.C. Berkeley.

• The Effect of Implant Angle and Resist Shadowing in Submicron Implant Technology.

• Lecture by Dr. Lynn Fuller of Rochester IT.

• Ion Implantation: Science and Technology by Dr. James Ziegler.

• Anatoli Bugorski, a Soviet researcher whose head was hit with 76 GeV proton beam (”G” being giga, or 10⁹, so 76 billion eV!) and survived.

• Funny images made using an ion beam. These systems generally don’t use the same species as ion implantation, but rather He, Ne, or Ga.

• Stopping and Range of Ions in Matter. This program allows the user to run their very own implantation Monte Carlo simulations from their (Windows) computer. It is even used in industrial settings.

• Example ion implantation problems for students. Corresponding lecture found here.

• Summary of commercial ion implantation technology. Written by two of the industry’s top companies (Axcelis and Applied Materials). The history and development of implanters is discussed.

• Vacuum: Discussion of vacuum concepts, technologies, pumps, etc. Useful for an equipment engineer in the semiconductor industry.

• Axcelis White Papers

Vibes, or What Paul Graham Would Say

Taiwan is nice. By what definition of nice? you ask. To which I answer: all of them.

The place itself is nice. It’s just nice to be there and walk around and exist, even if you’re not there for a particular reason or doing anything particularly exciting or fun. This is what Paul Graham would have felt if he had included Taipei in his Cities and Ambition.

The people are nice. Everyone we interacted with had a smile on their face and love in their heart. We never felt direct pressure to buy anything or tip or really do anything at all; we were just left alone to do as we pleased. There’s also an air of trust throughout the country, a feeling that you can leave your laptop in a coffee shop or door unlocked and be just fine and whole when you come back.

The weather is nice. Although we did go during one of the best times of the year and about a week or two after a typhoon, the temperature was moderate and we didn’t experience any crazy weather. (If you’ve been during the summer, you may be scoffing at this paragraph.)

The parks within the city are nice—the landscaping is well-groomed and designed and there are the perfect amount of benches with and without shade. The grass is green and the flowers colorful. The museums are well-curated, informative without going into too much detail, and clean.

The food is nice, but that deserves its own section.

Is this place a utopia? An argument could be made.

Food, or Why I Now Feel a Deep Pain When I Eat Back Home

The food here is as amazing as it is plentiful as it is cheap as it is varied. Restaurants line every street and alleyway; some are crowded with lines and others are empty with bored shopowners twiddling their thumbs, but both have plenty of food at the ready for the next hungry customer. Nor are the restaurants the same. Sure, beef noodles and buns remain some of the staple dishes, but plenty of other options exist, from turnip cakes (lo bak go) to Chinese omelettes to Taiwanese burgers (koah-pau) to steamed buns to raw eel. The tastiness is also top-notch, although we probably filtered out most of the bad places using Google reviews.

Did I mention cheap? Taiwanese meals were many times cheaper than the American equivalent because everything's cheaper there: labor, ingredients, real estate. Competition is fierce and forces the market to drop prices to keep up. I love free markets.

And healthy(-ish)! Fruits, vegetables, rice, and meats, particularly pork and fish, are staples in diets here, which, all things considered, is fairly healthy. There are definitely fried foods. Desserts aren’t as obviously laden with sugar and trans fat. It’s incredibly easy to eat healthy food while out and about.

Comparative advantage can be seen in the flesh here by watching just how many Taiwanese people eat out for their meals. The food is cheaper because the restaurants and vendors buy in bulk and it’s more delicious because, well, they’re chefs and have practiced their craft for quite some time. In multi-generation homes (i.e., children, parents, and grandparents all live together), the grandparents will often get groceries from the local street markets during the day and have food ready for everyone when they get home. As an aside, I’ve heard in that in a southeast Asian country (I forgot which) all social classes go out to eat at street vendors or restaurants—it’s not just the poor out eating street and fast food while the rich dine in or the rich out eating at fancy restaurants while the poor scrape by on rice and beans at home. I got a similar feeling in Taiwan that nobody was above eating street food.

Like everywhere with any business, the places that had massive lines or were filled up with customers were often the most delicious, the most TikToked/Instagrammed, or some combination of both. A donut shop we walked by had almost 80 people in line! Thankfully most of these places seem to have mastered the art of the "here's your food now pay quickly and GTFO but in the nicest way possible so we can get more business and you can enjoy our food".

The night markets are pretty awesome. Smells, sights, and sounds galore. You can grab some stinky tofu in one booth, turn around and play some carnival games, go to the next booth for some okinomiyaki because it turns out stinky tofu isn't your thing and the person you're with almost vomited from the smell, then buy some shirts and trinkets in another section. There are also so many people from all walks of life: families and friends, young and old, locals and foreigners. Like I said above, night markets welcome all demographics and shun none.

All of this makes Taiwan a food lover’s paradise. Cheap, unique, and delicious food is everywhere, served up by friendly people.

(In case it wasn’t obvious why I feel a deep pain now, it’s because an average meal in my home city costs somewhere around $15 (I’m a cheap bastard and want $3 meals again) and I often can’t simply walk to it from my apartment (I’m a lazy bastard and want it right next door). The variety is definitely available, though.)

Before looking it up, I guessed the obesity rate of Taiwan is 15%. The actual number is 12.5%.

And for some specific recommendations in no particular order:

• Taipei:

  • Sihai Soy Milk: Gotta get here early to beat the rush. Everything is good.

  • Di Yi Beef Noodle Restaurant: Don’t ask. Just go.

  • Rahoe Fuzhou Pork Pepper Buns: So delicious. Most pepper buns with meat will probably do, but these were especially tasty.

• Tainan:

  • Neighbors Baking: A very nice woman who specializes in cream puffs.

  • 井佐製所ケーキ: Cake shop with a cozy interior and delicious tea.

  • Tianmianjiang: Homey spot with Taiwanese classics.

Fun Food Facts

Taiwan is a land of superstition and it shows in two food-related ways:

• Pineapple:

  In Taiwanese the pronunciation of the word pineapple sounds like a propitious blessing of good fortune and future prosperity. ... The pineapple has been referred to in traditional culture as the best gift for a house warming party and upon the opening of a new business or to wish one’s favorite political candidate success at the election boxes.

  Bittersweetly, lawyers, nurses, and police officers are exempt from this tasty tradition because if their job is prosperous, it means crime and sickness.

• Kuai Kuai culture:

  A phenomenon in Taiwan wherein people put snacks of the brand Kuai Kuai next to or on top of machines. People who do this believe that, because the name of the snack—”Kuai Kuai”—stands for “obedient” or “well-behaved,” it will make a device function without errors.

  Apparently TSMC is a big proponent of this. Maybe this is their secret sauce, not their insane work culture!

Cleanliness, or Where All the Trash Cans At?

Walking around any streets you’d think that Taiwan had invented the 2 nm transistor node before trash cans. They are few and far in between for a few apparent reasons that I can deduce or find online:

• Having to hold your trash makes you a bit more conscious about your consumption

• Having to hold your trash until you get home encourages proper sorting of trash vs. recycling. Taiwan’s recycling rate is an astounding 55%, compared to the U.S.’s 32%.

• Fewer trash cans means less smell, fewer overflowing cans, and fewer rats and pests per unit area

There’s an unspoken rule about being able to pop into any of the numerous 7/11s to dispose of small items. We came up with the system of using a small bag to hold our trash while we walked around. For households, trash trucks drive around playing their version of ice cream truck music to notify residents that they better get outside to throw away their garbage.

Litter was almost non-existent, even in the extremely crowded night markets where it’s easy to drop stuff and difficult to pick up. There was even a human street sweeper outside our hotel sweeping leaves into a dust bin.

Unsurprisingly, there are some bad apples who litter. And unsurprisingly, the local populace doesn’t take too well to it, opting to shame them (and I’m sure name them if they could) through screenshots of security camera footage. Taiwanese Liam Neeson ready to hunt down litterbugs might make for a good advertising campaign!

Read a bit more about Taiwan trash here.

Orderliness, or Getting In Line Almost Literally Everywhere

The Taipei metro is very orderly. New riders patiently wait in a single-file line marked on the floor while the old riders get off, then get on when the coast is clear. Nobody plays music out loud. Nobody talks on their phone. Few people talk to each other. All you really hear is the noise of train on the rails. It’s eerily quiet and calm compared to the complete and utter chaos that is New York City’s subway. But this is exactly how it should be by default; if I wanted crazy, I’d go to the club or bad part of town.

Transportation, or You Better Learn How to Ride On Two Wheels

Scooters reign supreme. Parking is easier to find and actually perform; the low(er) speeds of cars are less likely to kill you in a collision; infrastructure is made for you (there is a scooter-only section in front of cars at stop lights). I don’t think I’ve ever seen so many scooters in one place. It’s comical watching tens of scooters drive around a roundabout like they’re in a Mario Kart race.

Walking is second in command on the transportation hierarchy. The metro is never more than a few minutes’ walk away. Bikes are available for rent if you really need one. The climate (for us) was generally pleasant and made walking pleasant by extension.

All vehicles are super courteous to pedestrians, except for that one guy in Tainan who almost ran me over because he was looking at his phone.

Language, or Using My Hands and Smile For Most Interactions

Mandarin and English are high-class, Taiwanese is middle-class, and Hakka, the tribal language, is low-class, like you’re coming from a Native American reservation. At least this is what our food tour guide told us. If you want a high-class job, learn Mandarin. This goes against my intuition of the Taiwanese government wanting to keep the Taiwanese language at the forefront because it continues to make them culturally different than mainland China, further separating the identity of the Taiwanese people from Chinese people.

I’ve never read Chinese characters before. For whatever reason I incorrectly assumed there would be a pretty standard font, but nope, turns out they have their own Calibris and Arials and Times New Romans for us foreigners to struggle with! My first run-in with this was at a breakfast spot we had already visited. The first time we went a very kind Taiwanese-American woman helped us order, showing us the where the pictured menu items were on the actual text-only order ticket itself. Easy peasy. I came back a few days later, confident that I could replicate it. Nope. I spent 10 minutes meticulously comparing the menu to the ticket to find exactly what I wanted. Damn you and your beautiful fonts, Taiwanese typographists!

Most people barely spoke English, forcing us to quickly adapt by pointing, holding up numbers with our fingers, and having a beaming smile to make sure we were friendly and not at all miffed by their lack of English. The “worst” ESL interaction we had was when we ordered a breakfast item and she came back with a Google Translate screen that said “Chinese medicine”. Me, in my infinite awkwardness when in another country and not being able to speak the language, simply nodded yes and prayed that it’s what we wanted. It came out looking normal, tasted normal, and I didn’t experience any crazy side effects, so I think Google just had a brain fart.

Cartoons, or Na Na Na Na Boo Boo We’re More Fun Than You

Everything is cartoonized. Wet floor signs, condom advertisements, crosswalk signs reminding you to press the button, street art, bike vs. pedestrian lane reminders, some sign that looks to be about karaoke that I couldn’t translate, entrances to buildings and storefronts. You name it, there’s probably a cartoon version of it somewhere in Taiwan. Call it rule 35.

And all of it was lovely. It screamed “we can have fun while still conveying pertinent info” instead of screaming “we are a boring society and must convey pertinent info in the most boring way possible for reasons we haven’t really thought about”. Have some pizzazz for crying out loud! Don’t just use a wet floor sign, use a wet floor sign in the shape of a banana! Don’t just make a sign that says the name of your business, make a sign that gives the info with a fun little mascot! Don’t be boring, be fun!

Miscellaneous, or Random Info That Doesn’t Really Fit Anywhere Else

Taipei 101‘s tuned mass damper system is flippin’ awesome and well worth the $25 it costs to see in person. Here’s a video of it working during an earthquake. T101 also houses what used to be the world’s fastest elevator at a top speed of 60.6 kmh or 37.7 mph.

The dress code is super casual: good-looking and comfortable was the criteria most people seemed to use. I saw very few suits while out and about. Taiwanese Redditors seem to confirm this. For weddings, dressing up doesn’t matter nearly as much as how thick the hóngbāo you bring is.

Bidets are widely used over here. Nice hotels seems to offer heated-seat bidets by default.

The Grand Hotel in Taipei was awesome for our last night’s stay. They have tours of their once-secret tunnels (one of them even has an emergency slide, but alas, you aren’t able to ride it due to “safety concerns”—I expressed my disappoint to some chuckles and a polite “you’re not the first!”). We discovered that Grand Hotel weddings are highly sought after when we walked in upon 20+ wedding-dressed couples standing in line (of course) to take their pictures in front of the Christmas tree and staircase.

Cash is still king. Don’t get caught without a few hundred NTD on your person, otherwise it may be a bit awkward when it comes time to pay. Easy Cards are convenient and you can get some cool designs at any 7/11. Some (bigger) places accept credit cards.

See Also

• notes on taiwan

• Notes on Taiwan: Dan's notes were a big spur to get me to Taiwan and I reiterate and confirm a lot of what he says

• Community Trust Tests: Simple tests to reveal a lot about a place

• Travel to Cheap Destinations: More bang for the buck often leads to a better time

Semiconductor Fabs I: The Equipment

Preface

I tried to include as many links as possible to allow the reader to go down rabbit holes as they see fit.

I tried to keep info basic enough for the skimmers/laypeople to enjoy while still adding more technical details located at the end of each section in the collapsible element. As you’ll see, there are a lot of “for example”s. As the great Paul Halmos once said, “a good stock of examples, as large as possible, is indispensable for a thorough understanding of any concept, and when I want to learn something new, I make it my first job to build one”.

Interactive animations are denoted by the large “Click here to [X]!”. While rudimentary, they get the job done. These were vibe coded using Claude 4 Opus with extended thinking and a few iterations.

This information is current as of August 2025.

Sections or sentenced with an asterisk (*) I am less knowledgable about the information in this post than in Fabs I. I only have firsthand experience in a few (older) fabs, but do have some secondhand experience through colleagues and friends working at more advanced fabs.

Context

Go read Construction Physics’ (Brian Potter) How to Build a $20 Billion Semiconductor Fab. Seriously, go read it or you will (probably) be lost over the rest of this post.

What Do Fabs Want? Chips! When Do They Want Them? Now!

And that isn’t an exaggeration. Like most manufacturing operations, fabs want to pump out as many wafers as possible in as little time as possible for as little cost as possible with as few defects as possible. Sound familiar?

Fabs are often skirting around the impossible region depending on a variety of factors.

First and foremost, market conditions. If the market is desperate for chips, fabs are willing to spend more money to satisfy the customer demand. If a downturn is happening, the money belt gets tightened and fabs are more focused on going slow to make sure everything they output is of acceptable quality and can be sold for money.

Second, leadership. A fab’s culture comes from the top. Is the fab leadership more risk averse to the point of wanting to slow things down to make sure the final product is high quality and cheap? Or are they willing to risk it for the biscuit (read: a fat bonus) and go into warp speed (read: turn off some quality checks) to get more chips out the door faster?

Combine these factors with the Eye of Sauron that is the company’s top brass constantly scanning fab performance metrics (speed, cost, quality) and you get ever-changing priorities, especially if one metric was particularly poor, e.g., poor quality will lead to less risk taking at the cost of other metrics getting worse.

Onto the operations. A fab’s operations consists of people and systems dedicated to producing as many wafers as possible in as little time as possible without botching the process flow. They (mostly) don’t care about cost because that’s an engineer’s job. They only care about quality when it affects their metrics, otherwise it’s the quality group’s job.

Because I’m only familiar with one or two fab’s operations styles, I’ll pull what I can from the literature and the rest out of my as...repertoire of semiconductor knowledge. There may also be some “this would be super nice to have moments” in there.

Terminology, Also Known as Fabspeak

Defining key terms upfront is helpful for obvious reasons, so here they are in no particular order:

• Cycle time: Time required to produce a part or complete a process. From Semi Engineering’s Battling Fab Cycle Times:

Generally, the most common metric for cycle time in the fab is “days per mask layer.” On average, a fab takes 1 to 1.5 days to process a layer.

• WIP: Work in progress, or the number of wafers currently in progress at X, where X can be a specific step, a group of steps, etc. “Loadings” is similar. For example, fab X’s loadings/WIP may consist of product1 at 60% of total WIP, product2 at 30% of total WIP, and product3 at 10%. The start numbers (the next term) reflect these numbers.

• Starts: The number of wafers that are started each day, or WPD. For example, a 500 WPD fab will add 500 WPD to its total WIP each day. Hopefully they are completing more than 500 WPD, otherwise WIP will start to build up quickly!

• Product or technology: Each wafer is a unique product, also called technology. Product1 may be for your car’s braking system and product2 may be for your GPU. These require different processing steps but may still share the same machine at some point in their life cycles. For example, all cars go through the paint step but not all paint steps apply to the same car.

• Device: A subset of a technology. Device1, ..., deviceX may all be of type technologyX, but have slightly different functions. For example, if a Toyota Prius is the technology, then the Prius LE, Prius XLE, and so on are the specific devices. This is also like saying a specific manufacturer part number. Individual chips, also called die, on the wafer are all made up of the same device.

• Lot: A set of 25 wafers that all follow the exact same steps (unless some are getting experimented on). These all have a specific number for tracking purposes.

• Qual: A test wafer (or wafers) that verify a tool is operating correctly. For example, a deposition tool qual’s process flow may look something like:

  1. Measure pre thickness (so we know how much thickness was added)

  2. Measure pre particles (so we know what particles were already there)

  3. Process step on tool with target conditions

  4. Measure post particles (to see how many new particles there are)

  5. Measure post thickness (to see if the target thickness was reached)

• Overall equipment effectiveness (OEE)

• FOUP: An acronym for front opening unified pod, these are the carriers that lots travel around in from tool to tool.

Filling the Fab Full of FOUPs*

Planning groups work across a company’s fabs to ensure the best loadings based on current demand, forecasted demand, and each fab’s capabilities. Shift in the market? Adjust the numbers! Change in another factory’s capabilities? Adjust the numbers! Wind direction change? Adjust the numbers!

Planning groups work with each fab’s industrial engineers to make sure the numbers they’re looking for can run, bringing us to...

Modeling a Fab’s Throughput Capabilities

Before actually running any material, the fab must know how much it can run. You can’t just say I’m going to start 1000 WPD without knowing if those numbers can actually be supported.

That’s where the industrial engineers come in. Trained in the art—okaaaaay, science—of modeling, they’re able to tell:

1. How quickly each product can make it through each machine, and thus the fab

2. Which tools each product uses

Those numbers can then be normalized to determine the possible product mixes that can run, i.e, how much of product1, ..., productN can be run at any given time without interfering (too much) with the other products. We want everything to move smoothly!

Starting Fresh Wafers

Now that the loadings are decided, the fab can actually start running wafers. Fresh lots are started and assigned both a lot number and device name.

They then go out into the fab, get processed according to their flow, and get shipped out for testing (that’s outside the scope of this post).

It’s at this point that the MES takes over.

The Brains of the Fab: Manufacturing Execution System

As the heading implies, the manufacturing execution system (MES) is the brains of the fab, helping to coordinate, process, and keep track of every single lot that’s currently active and its history. Everything you need to know—past steps, future steps, current steps, etc—is in one central location. It’s from the MES that you can do many different things:

• Set up new process flows for new devices entering the fab

• Change (improve) existing process flows

• Run experiments on lots

• Prioritize lots

• Review lot history

It’s difficult to overstate the level of automation and complexity of modern-day fab MESs. It keeps track of an insane amount of granular data—from individual wafer IDs to the process for each to certain signals collected during the processing. It manages everything almost flawlessly with the occasional help from people when something causes an issue.

(Side note: I think a fun experiment to test a fab’s automation abilities would be removing all humans and seeing how long it would take for the fab to come to a complete standstill.)

Real-Time Dispatch (RTD)

The Planning Master has an excellent real-time dispatch post that goes into great detail and gives great examples of RTD rules. Here are some shamelessly-copy-and-pasted quotes from the post that summarize RTD:

The idea [of RTD] is to reduce WIP bubbles by spreading the WIP across the flow as evenly as possible. Another goal is to do the best local optimization so we can maximize throughput across the various [processing steps].

We can do this by combining global rules with local rules:

• Global rules – looks at all the WIP that is running in the FAB and prioritizes the lots.

• Local rules – looks at the WIP in a specific station and optimizes the throughput (for example by picking the same type of lots to run in a cascade)

Pretty self-explanatory. There’s a ton of math that goes into this on the backend to ensure optimal utilization. More on that in Math of Manufacturing.

Click here to test your dispatch skills!

RTD Rule Examples

The Planning Master also gives examples of some rules (I’ve added some that have “[NMP]” at the beginning):

Implant:

• We would like to have lots running the same gas one after the other.

• We would like to have lots with the same energy levels running one after the other.

• We would change gas if the number of wafers waiting in the implant station is more than 100 and 50%-60% of the lots needs a different gas.

Diffusion:

• Largest batch first. [if a batch process]

• Running test wafers as part of a production batch if possible.

• [NMP] We would like to have lots running the same temperature one after the other.

• [NMP] Can we wait X time to increase batch size?

Photo:

• Using masks already on the photo magazine.

• Promote lots in different processes that needs this masks, to increase batching on the photo tool.

• Changing mask will be done if the number of wafers waiting for different maks is more than 100 and 50%-60% of the lots needs different mask.

[NMP] Metrology:

• [NMP] Prioritize metrology tools that are physically closer to the tools that use them

See Also (RTD)

• An Integrated Approach to Real Time Dispatching Rules Analysis At Seagate Technology

• Comparison of Dispatching Rules for Semiconductor Manufacturing Using Large Facility Models

• Multi-objective real-time dispatching for integrated delivery in a Fab using GA based simulation optimization

• Deep Reinforcement Learning for Queue-Time Management In Semiconductor Manufacturing

• Short-Term Scheduling Model of Cluster Tool in Wafer Fabrication

• Bottleneck Management Strategies in Semiconductor Wafer Fabrication Facilities

Miscellaneous MES Features

Here’s a list of features that are or would be super helpful in a fab’s MES:

• Staging lots near their next processing tool. Sometimes the tool a lot needs to go to is already in use, but why not kill the time by making sure the lot is super close when the tool is ready to accept it?

• Automatically create and run quals for tools. Quals may run on an as-needed or periodic basis. If the MES can know that either is coming up, it can create the qual in advance so it’s waiting when the tool is ready.

• Prioritize certain tools by location. Let’s say a qual finishes processing in sector1 and there are two measurement tools it can run on, one in sector1 and one in sector8. We want it to process in sector1 because we get results faster and it frees up the overhead transportation system.

Manufacturing Math

(This section doesn’t go deep because a) I’m not super familar, and b) I’m not interested (enough) in becoming super familiar.)

With that being said, go read Chance and Robinson’s FabTime Wafer Fab Cycle Time Tutorial and Queueing Formula Examples. This is one of the better, more accessible texts I’ve read on some of the math behind manufacturing chips (although it’s transferrable to most, if not all, manufacturing processes).

SemiEngineering also discusses some of the math and strategies in their Battling Fab Cycles Times.

Priority Lots

Sometimes lots need to get expedited. Reasons range from it being a prototype device that the company or another group wants quick results on; a product that a customer is willing to pay extra for; or a lot involved in qualifying a new tool or improving a process.

The MES allows priority levels to be set to speed up the processing of the lot. This can be achieved a few different ways:

• Allowing that lot to skip ahead of other lots that were already waiting at a processing step. For example, if lot1-lot5 are waiting to process at step 64 and lot6, the priority lot, shows up, then lot6 will process before any of the waiting ones.

• Ensuring tools are empty when the priority lot arrives. Time per process step is both known and stable, so the MES can know at what time to stop processing lots on tools that the priority lot is going to arrive it. It can also restart tools that the priority lot has passed.

• The overhead transport system moving the lot before others. This is like an ambulance choosing the highest priority patient and then all the cars pulling over to the side to let it get through more quickly.

Click here to prioritize a lot!

Priority lots are no joke. Time is critical here because it can add up so quickly. An extra 30 minutes per processing step—which is on the low end if not managed properly—across 300 total steps results in an extra week of processing time! This is important enough that engineers will get called if there’s a hold up. Multiple lots of the same priority device will often be started at the same time in case something unfortunate happens to the leading lot.

Priority designations aren’t handed out lightly because if every lot is a priority then no lot is a priority. They also slow down other material from processing.

• Hunting For North Korean Fiber Optic Cables: The author uses various internet images and material to piece together how NK has its fiber optics lines laid throughout the country.

• Master of His Virtual Domain: What a top Clash of Clans player’s life looks like when working to stay at the top of the leaderboard.

• Jeffrey Epstein’s flight logs: You can search by guest, year, aircraft.

• Jeffrey Epstein’s camera roll as Google Photos: A cloned Google Photos but it’s Epstein’s camera roll, complete with by-name search and favorites.

• Los Angeles Fire Damage (visualized): Before and after photos of the devastating 2025 fires. Move the map and click on a dot to view a home.

• First Justice Department memorandum from February 2010 arguing that killing Awlaki would be legal: al-Awlaki was the first American citizen to be assassinated by a U.S.-led drone strike. This document outlines the legal justification of killing him.

• J.D. Vance Dossier: The full background research on J.D. Vance before he was chosen as Trump’s vice presidential running mate. A very interesting look into what is actually looked at (and how deeply) in high-level politics.

• You Get About Five Words

• The Lost Generation: Savage discusses being a white (millenial) man in today’s day and age, especially in certain professions, and how they’ve been left out to dry in favor of other candidates. Complete with some stark numbers about hiring rates and demographic changes over time.

• Categorizing Extreme Elites

• Cognitive Reflection Test (CRT): “The CRT was designed to assess a specific cognitive ability. It assesses individuals’ ability to suppress an intuitive and spontaneous (”system 1”) wrong answer in favor of a reflective and deliberative (”system 2”) right answer.” There are only three questions.

• Baird Industrials: The Epitome of a Sweatshop: A look inside an investment bank from the eyes of the analysts and associates in the trenches.

• Growing Independence: Jeff gives examples of how he is nurturing independence in his young children.

  I have three main motivations here. The first is teaching: eventually they’ll need to make good decisions on their own, and the sooner they start the more practice they’ll be able to get. The second is a kind of long-term laziness: once they can do things for themselves it’s less work for me. And the third is respect: they’re people and as much as possible they should get to choose how their lives go.

• Comment, Don’t Message: Jeff makes the case that publicly commenting is often better than private messaging because more people can see it (and thus benefit), others can chime in, and the response burden is less.

• Experimental Study on EUV Radiation Characteristics of 1 μm Laser⁃Excited Solid Sn Target Plasma

• Former ASML head scientist Lin Nan drives China’s latest EUV breakthrough:

• Exclusive: How China built its ‘Manhattan Project’ to rival the West in AI chips

• Madrid Trip Report: Joshua talks about his experience traveling in Madrid.

• Alice in Wonderland syndrome

• Turning 20 in the probable pre-apocalypse

• You Look Better When You Try Hard

• the 80th percentile displacement: why Russ Roberts (and you) hates modern popular movies

• Peter Luger Used to Sizzle. Now It Sputters.: Pete Wells’ scathing review of Peter Luger.

• On Israel, Trump Should Echo Reagan: Joshi makes a case for the U.S. weakening its relationship with Israel.

• Contradict my take on OpenPhil’s past AI beliefs: Yudkowsky invites the community to explain how/if he was wrong.

• Peng Zhao: Citadel Securities (not to be confused with the hedge fund Citadel, although they are closely related) CEO as of December 2025. He assumed the role at a very young 34 after being their chief scientist for less than a year.

• Claude Opus 4.5 Achieves 50%-Time Horizon Of Around 4 hrs 49 Mins

• Howard Marks (investor):

  co-founder and co-chairman of Oaktree Capital Management, the largest investor in distressed securities worldwide

  …

  He hopes to have average returns during a bull market, while minimizing losses during bear markets due to his belief that losses do more harm than any benefit investors obtain from gains.

• Robert Hanssen: An FBI agent who spied for the Soviet and Russian intelligence services for money.

• The Hillbilly Elite: JD Vance’s blog from his Yale Law School days.

• I think Substrate is a $1 Billion Fraud: Part 1: FCR explains the reasons Substrate could be to good to be true, including the fact that founders have been outed as frauds before and there is basically no evidence to support their claims.

• The Money Ick: Robin Hanson gives a bunch of examples of places where we used to feel weird about charging for money, but now do it without a second thought.

• Prediction Markets Now: Robin Hanson on the current state of prediction markets.

• Kid Door: Jeff builds a mini door for his young children to go through.

• Childhood and Education #16: Letting Kids Be Kids: Zvi talks about the child supervision and lack-of-independence pandemic sweeping the nation today. Starting at 10 years old my parents let me do whatever as long as I told them where I was going, if I changed places, and was home by 9:00pm. The leash got much longer as I got older.

• I said hello and greeted 1,000 people at 5am this morning: Golden retrievers are loved, and acting like one can increase the amount you are loved (at least while you’re acting like one).

• Mar-a-Lago face:

  a plastic surgery and fashion trend among American conservative and Republican individuals described as excessive or uniform plastic surgery interventions such as lip augmentation, Botox, and jaw contouring, coupled with heavy makeup, spray tans, fake eyelashes, and dark smoky eyes.

• Deal toy: “customized memento or gift that is intended to mark and commemorate the closing of a business deal in finance or investment banking”

• Tombstone (financial industry):

  a type of print notice that is most often used in the financial industry to formally announce a particular transaction

  …

  This public disclosure is done in a form that lists the participants in a specified order ... The order is so important that, in 1987, five top investment banks withdrew from a syndicate underwriting a $2.4 billion debt issue for the Farmers Home Administration, because they would have been listed under other, smaller regional banks.

• Most Badass Intern Story?: Investment bankers share their most badass intern story.

• VSCO girl:

  VSCO girls are described by some as “dress[ing] and act[ing] in a way that is nearly indistinguishable from one another”, using oversized T-shirts, sweatshirts or sweaters, Fjällräven Kånkens, scrunchies, Hydro Flasks, Crocs, Pura Vida bracelets, instant cameras, Carmex, metal straws, friendship bracelets, Birkenstocks, shell necklaces, and other beach-related fashion. Environmentalism, especially topics relating to sea turtle conservation, is also regarded as part of VSCO culture.

• Sankebetsu brown bear incident: Japanese bear attack saga.

• Sloth bear of Mysore: Sloth bears are no joke, and this one was particularly aggressive. There are some other pretty cool stories out there of professional hunters tracking down human-killing animals and writing about it.

• Jefferies Houston | Culture, Hours, Exits?: Another look inside an investment bank from the eyes of the analysts and associates in the trenches. One of the Dallas Jefferies associates died in 2025.

• Moelis Houston PSA: A perspective of what it’s like working for Moelis in Houston.

• Dumbest thing you’ve said in an interview: Wall Street folk share the dumbest things they’ve said in interviews.

• Examples of Superintelligence Risk: Jeff discusses how superintelligence could take over in some rather boring ways.

• 2025 in Reading and 2026 Goals

• JJ’s Razor: “Malicious or stupid, it doesn’t matter, because your options are the same”

• The Wadsworth Constant: “the first 30% of any video can be skipped because it contains no worthwhile or interesting information”

• Marchetti’s constant: “the average time spent by a person for commuting each day. Its value is approximately one hour, or half an hour for a one-way trip.”

• You won’t believe what gets an email flagged at Goldman: CNBC has the list: LLMs probably obsolete this list, but it’s still pretty funny with things like “Where did my {money}|{funds}|{account} go”, “Paying fees {through|thru} the {nose|a--|butt}”, and “don’t you f*cking understand”. This is just normal banker speak! Trust me, I watched Wolf of Wall Street!

• TSMC Arizona Outage Saw Fab Halt, Apple Wafers Scrapped: 1000s of wafers is brutal, especially on a new facility where they’re trying to recoup the cost quickly. My best guess is some bulk gas supply, like nitrogen, failed.

• Confessions of a Bicycle Race Promoter: Willis talks about what it was like to run Austin’s Driveway Series.

• McMansions 101: What Makes a McMansion Bad Architecture?: Masses and voids, balance, proportion, and rhythm.

• Who earns a higher salary than you and the jobs they work: A nice data visualization on who earns what.

• Flower Mound, TX: An example McMansion in Texas, including labels on everything that makes it ugly and unappealing.

• When the Missing Reasons Aren’t Missing: Estranger parents can look past the reasons their children give them for the estrangement.

• Unknown Knowns: Five Ideas You Can’t Unsee

• McMansion Hell: A site dedicated to posting about the ugliness of McMansions. There are even educational sections for those out of the loop on architecture.

• Inside Dubai’s dark underbelly as models lured to sick ‘Porta-Potty’ parties by men lavishing them with gifts: This is the centerpiece of a rumor about how the Dubai chocolate trend started.

• Pro rata: “in equal portions or in proportion”. “In venture capital, it can refer to the Pro-Rata Participation right and mean ‘the right to continue to participate in future rounds so that you can maintain your ownership.’”

• Nick Patterson (scientist): “a mathematician ... with notable contributions to the area of computational genomics”. He was also a child chess prodigy and worked at RenTech for a while.

• Heather Sue Mercer–Duke Football case

• Because It’s Easier:

  training should be IN-efficient, and difficult. Training is an attempt to change the condition of muscle and connective tissue, the energy system supplying them, and to cause adaptation and improvement. Evading the difficulty one consciously sought in order to produce that change does the opposite. Our nature seeks efficiency and if we aren’t stridently aware of that nature it can sabotage our consciously chosen processes.

• David Magerman: “an American computer scientist and philanthropist. He spent 22 years working for an investment management company and hedge fund, Renaissance Technologies.”

• Stephen Trauber: According to WSO, Trauber runs a very sweaty shop in Houston.

• Ding Xuexiang: “currently the first-ranked vice premier of China and the sixth-ranked member of the Politburo Standing Committee of the Chinese Communist Party.”

• Le Morne Brabant: A peninsula in Mauritius famous for its “underwate waterfall”.

• Kevin Warsh: A potential pick for the 2026 Fed chair position. He was on the BOG from 2006 to 2011.

• Timothy J. Heaphy: Lead investigator of the January 6th committee.

• Jason Gaverick Matheny:

  American national security expert who has been president and CEO of the RAND Corporation since July 2022. He was previously a senior appointee in the Biden administration from March 2021 to June 2022. He served as deputy assistant to the president for technology and national security, deputy director for national security in the White House Office of Science and Technology Policy, and coordinator for technology and national security at the White House National Security Council.

• Kojak:

  To find an empty parking space directly in front of the building you are visiting, regardless of the time of day, or busy urban location. From the televison series “Kojak”. The title character would race off to locations in Manhattan and always park right in front of the building.

• Gish gallop:

  a rhetorical technique in which a person in a debate attempts to overwhelm an opponent by presenting an excessive number of arguments, without regard for their accuracy or strength, with a rapidity that makes it impossible for the opponent to address them in the time available.

Google is where the reputations of businesses are both made and broken. A poor Google score or review is enough to turn consumers away without a second thought. Businesses understand this and do whatever they can to earn the precious five stars from each customer: pressuring you in person or via email to submit a review, creating QR codes to make it easier to review, giving you a free item, the list of both ingenuity and shadiness (and sometimes both!) goes on. Businesses’ response to a poor review can help them look good to potential customers or confirm the review’s accusations.

In a world with no reviews, consumers go into everything blind. They have no clue what to actually expect, only what the business has hyped up on their website. The businesses are also blind. They operate in a feedback loop that is difficult to get information.

The power ultimately lies in the consumer’s hands, just like South Park’s Cartman thinks. And with great power comes great responsibility.

(The rest of this essay assumes the reviewer is a reasonable, charitable, and kind person.)

Helping Everyone Out

Leaving as many honest, descriptive reviews as possible provides information for both the business and other potential customers to make decisions off of. Businesses can take the feedback and improve off of it, guarding against another potential review having the same piece of not-positive feedback. Customers can decide to not eat there, sending a silent signal to the business that they’re doing something wrong. But what? Is it the prices? The dirty bathrooms? The fact that they require your phone number and spam you even though they literally call out your order number? How does the business know what exactly they’re doing wrong?

The reviews! The businesses have to have feedback, preferably in the form of reviews, to know and improve on what they did wrong, and the only party that can give them that is the consumer.

Other businesses can also learn from reviews, both directly and via negativa. Business A can look at reviews of business B to figure out what they’re doing wrong and fix it before it comes to bite them.

In the end, everyone is better off for it. Customers get better businesses and businesses get more customers because they’re now better businesses. The cycle repeats itself until we’re all eating a three-star Michelin restaurants and experiencing top-notch service at all bicycle shops.

Rating Businesses

I’m still slightly undecided on how to rate businesses. Do you rate them relative to others in their class (e.g., steakhouse vs. steakhouse, but not steakhouse vs. taco joint)? Do you aim to form a bell curve? Are they actually normally distributed? Is five stars the default, with anything less than the expected level of service or quality of product resulting in stars being removed?

In the end, I think you have to rate on an absolute scale (which should roughly turn into a bell curve, although maybe not entirely centered). The New York Times food reviewer Pete Wells has a nice system that helps him rate the restaurants he visited:

1. How delicious is it?

2. How well do they do the thing they’re trying to do?

But that’s just food. What about for all businesses, like a bicycle shop or hair salon or law office? I choose a weighted factor approach of:

• Job Quality (70%): This is the reason the business exists. A bicycle shop exists to sell and repair bicycles. If they did a kickass job, regardless of other factors, then the review should primarily reflect that. This includes things like speed, price, etc. If the job was slow compared to what was advertised or the quality did not meet the price paid, then that is poor quality. (These things should obviously be known or estimated before agreeing to start the job so there aren’t any surprises or disappointments.)

• Service (20%): Did you enjoy doing business with them? Did it make you want to come back? Job quality can only compensate for poor service so much.

• Vibes (10%): Are the vibes cool? Do you like what they’re doing and want to support them?

These weights may vary person-to-person, but I’d argue not by much. If they do, the priorities are probably wrong.

Structure of Good and Bad Reviews

How a review is structured matters because you get about five words. The important points should be up front with the minor points at the end.

Excellent experiences that are worthy of four or five stars should start positive in order to reinforce what the business is doing well and serve as a quick snippet for why others should come here. Any minor negative points should be at the end.

Here are two examples of five-star reviews for NMP Cafe, one high-quaity and one low-quality:

• HQ (5 stars): Delicious coffee (I had the latte), kind staff, and a cozy atmosphere that’s great for both working and socializing. Music was a tad loud for my taste, but others didn’t seem to have a problem with it.

• LQ (5 stars): Fine coffee shop. Music loud.

Poor experiences should start negative in order to directly explain what the business is doing poorly and serve as a quick snippet for why others should not come here. Positive points should come after.

Here are two examples of two-star reviews for NMP Burgers, one high-quaity and one low-quality:

• HQ (2 stars): Burger topping bar had flies buzzing around and was generally dirty. Cashier grabbed inside of cup with fingers. Burgers and fries were otherwise good.

• LQ (2 stars): Unhygienic food storage and staff practices. Food otherwise good.

All this said, leaving an X-star-only rating with no text is still better than nothing because it’s some information. The owner may even be able to tie it back to the reviewer and learn from it.

In-Person Reviews

In-person, so effectively private, reviews should become more normalized. (These are in addition to online, public reviews.)

Opening up a real-time dialogue line between the customer and business rep allows for more effective communication to be had through answering questions, clarifications, etc. And there shouldn’t be any awkwardness! The customer is essentially giving the rep a chance to do better and make even more money from happier future customers!

My approach in the few times I’ve done this is to politely ask for a manager, start with a simple “hey, I’d like to give you some polite feedback on X” (and maybe make it clear I’m not looking for a free anything), then kindly explain my position. They’ve always been outwardly receptive and appreciative of the chance to listen and talk. Experiences may vary.

Go Earn Your Google Points

Do it for your family, friends, and neighbors. Do it for the business owners that want to do better. Do it for the guy who was gonna experience a nasty meal, but because of your review—yes, your review—didn’t. Do it for the business owners who are constantly asking for feedback on their product and the experience because they’re struggling, but never get anything. Do it for the chance to become an influencer or food critic. Do it for the clout. Do it for your future self.

See Also

• Social Defection: Rules separate us from the animals

• The tyranny of the “best” - be wary of what strictly dominates

I’ve only been to two cheap destinations: Romania and Taiwan.

I’ve been to three expensive destinations: French Riviera, Iceland, and New York (I, II, III).

Philosophy of Choosing Destinations

There’s one rule and one heuristic I use when choosing where to travel:

• Rule 1: Does it pass the Do-I-Actually-Want-To-Go-Here test? If it fails, find another place (duh!).

• Heuristic 1: When in doubt about where to go, choose the place that’s cheap.

That’s it. Rule 1 removes all irrelevant questions that come after and Heuristic 1 helps to pare down the choices when the decision is difficult.

Don’t You Want to Live Like a King?

I’d argue a lot of people travel to get away from their everyday life and experience something new, and since most people don’t live like kings, living like a king would be something new and a reason why people travel. QED.

Traveling to a cheap destination (CD) unlocks purchasing power that you can’t get anywhere else. Sure, you can splurge on a nice dinner in your home city, but it sure doesn’t feel the same as eating the same quality of food for a fraction of the price while traveling.

Everything is more accessible when traveling to a CD. A few examples:

• Food: A delicious, full meal in Taiwan cost about $3 if you know the right spots, whereas the equivalent in New York costs upwards of $20.

• Lodging: Five-star hotels are now accessible. While paid amenities within the hotel may be more expensive than expected, the free ones remain nice.

• Experiences: Intra-country travel (taxi, subway, trains) are cheap. You can do fun things—massages, tours, etc—for cheap.

The other side of your coin is depressing: you (probably) live a quality of life worse than that of your home when going to expensive destinations, because, well, everything is more expensive! Combine that with constantly questioning if you should buy this or that because of how expensive it is and that makes for a not-great vacation. This worry is eliminated when traveling to cheap destinations!

The Experience May Not Actually Be Priced In

Markets give us a lot of information, but they’re flawed in the sense that they rely on taste and not everyone has the same taste (or good taste).

Taiwan was cheap, but freaking awesome. Romania was cheap, but freaking awesome. Iceland was expensive and pretty awesome. The French Riviera was expensive and pretty awesome.

So why are cheap destinations cheap? Wouldn’t market forces cause them to become more expensive? Not necessarily, for a few reasons:

• Status: Expensive places—Iceland, French Riviera, etc—are often high status; cheap places are often low status. People who don’t care about status games are few in number and more willing to go to cheap destinations.

• Mindset: If you’re sufficiently open-minded and curious, then where you go mostly won’t matter because anything can be interesting. People who are that open-minded and curious are few in number and more willing to go to “boring” destinations.

• Cost: It can cost more to get to these destinations, turning off potential visitors who may assume the cost on the ground is similar to the flight price, thus forcing prices on the ground to stay low to counteract the expensive flights and attract more visitors who are able to look past the large upfront cost.

See Also

• Matt Lakeman’s trip reports

• *Don’t Be a Feminist*: The Origin Story

• The case for overseas dating and marriage: The man gets a loving wife and the woman gets a massive improvement in quality of life.

• Miracle at the Meadowlands:

  For the Eagles, the victory snatched from the jaws of certain defeat served as a morale boost, leading that season to a playoff berth and, two seasons later, the franchise’s first Super Bowl appearance. To Giants fans, it was the nadir of a long era of poor results, but the aftermath of this would lead to major changes that proved beneficial for the franchise in the long run. For the sport in general, the main legacy of the game was its contribution to the adoption and acceptance of the quarterback kneel as the standard method for winning teams in possession of the ball to end games under the appropriate set of circumstances.

• Kuai Kuai culture: Translating to “behave behave”, Taiwanese engineers will put green and unexpired Kuai Kuai snacks near their equipment in hopes that it will behave better.

• Betel nut chewing

• (Report) Evaluating Taiwan’s Tactics to Safeguard its Semiconductor Assets Against a Chinese Invasion

• Where’s Putin? How The Kremlin Hides His Location With Three Nearly Identical Offices

• Chicago Boys:

  a group of Chilean economists who rose to prominence in the 1970s and 1980s. Most were educated at the University of Chicago Department of Economics under influential figures like Milton Friedman, Arnold Harberger, and Larry Sjaastad, or at its academic partner, the Pontificia Universidad Católica de Chile. After returning to Latin America, they assumed key roles as economic advisors in several South American governments, most notably the military dictatorship of Chile (1973–1990), where many attained the highest economic offices.[1] Their free-market policies later influenced conservative governments abroad, including those of Ronald Reagan in the United States and Margaret Thatcher in the United Kingdom.

• Destruction of Syria’s chemical weapons

• Don’t let people buy credit with borrowed funds

• Two can keep a secret if one is dead. So please share everything with at least one person.

• At Pope Leo’s Urging, Bishops Issue Historic Rebuke of Trump’s Raids

• Brightline is Actually Pretty Dangerous: “Brightline is about 20x more deadly per passenger-mile (counting people inside and outside the vehicle) than driving”.

• Snow Shovels & Singlespeeds:

  A decade ago I probably cared more about optimization, maximization, efficiency and outcomes. Carbon bikes, fast times, race results. Now as a middle-aged athlete and human, I find myself increasingly more interested in the means than the end. That might sound like a cop-out in response to my waning peak physical abilities. But I think such an attitude is also just the result of a natural maturation as one goes through life.

• ‘Scarcity and growth are oppositional’: How streetwear legend Supreme lost its luster: The headline says it all

• Karin Immergut:

  American lawyer who has served as a United States district judge of the United States District Court for the District of Oregon since 2019. She has concurrently served as a judge of the United States Foreign Intelligence Surveillance Court since 2024.

• How much of your life are you selling off?

• A Common Habit That Costs Us Friends: Never reaching out

• On Free Speech:

  So to conclude: censorship in public spaces bad, even if the public spaces are non-governmental; censorship in genuinely private spaces (especially spaces that are not “defaults” for a broader community) can be okay; ostracizing projects with the goal and effect of denying access to them, bad; ostracizing projects with the goal and effect of denying them scarce legitimacy can be okay.

• Bullshit Jobs:

  postulates the existence of meaningless jobs and analyzes their societal harm. He contends that over half of societal work is pointless and becomes psychologically destructive when paired with a work ethic that associates work with self-worth.

• James Duane:

  an American law professor at the Regent University School of Law, former criminal defense attorney, and Fifth Amendment expert. Duane has received considerable online attention for his lecture “Don’t Talk to the Police”, in which he advises citizens to avoid incriminating themselves by speaking to law enforcement officers.

• High-Density Days

• When the Job Search Becomes Impossible: Three Phases of Burnout

• Susan Monarez: “American microbiologist and public health official who served as the Director of the Centers for Disease Control and Prevention”.

• You can try to like stuff

• How I Eat

• 10 minutes is ~1% of your day: How do your emojis stack up?

• Do you like dogs, cats, both, or neither?

• “It’s a 10% chance which I did 10 times, so it should be 100%”

• e to the pi Minus pi

• Work culture creep: The environment part seems huge to me. This motivated me to change my work outfit to something much more professional to allow me to shift to my “home mindset” by changing clothes when I get home. Report to be published early 2026.

• Make product worse, get money: A similar argument seems to get made by believers of planned obsolescence, where companies make products last just long enough for the consumer to say “okay, that lasted a long time better go get a new one”. The risk of it getting out that they deliberately planned the lifespan and the free market encouraging cheaper prices and/or longer lifespans seems to go against that here. Combine this with just how big the market is and it’s probably in the company’s best interest to attract new consumers than force existing ones to pay for a new product.

• Rich Friend, Poor Friend: “So this dynamic emerges where my rich friends never ask each other for help, pay for services using money, and never do anything unpleasant for each other, whereas my poorer friends are always doing stuff for each other out of necessity and becoming closer knit in the process.”

• dissolution

• 2025 U.S. Department of Justice resignations: Integrity isn’t dead! Thursday Night Massacre seems appropriate here.

• 51 days in a Russian jail: Sofiane Sehili reveals all on his trans-Eurasia record attempt... and its spectacular failure: Sehili is one of, if not the, best ultraendurance cyclists out there.

• How a chip is designed: See Semiconductor Fabs I: The Equipment and Semiconductor Fabs II: The Operation for how a chip is made.

• What Cost Variety?

• Splash (otter): Search-and-rescue otter trained for police usage. Apparently otters can “detect scents underwater by blowing bubbles and quickly re-inhaling them; the inhaled bubbles absorb odors from the surrounding water.”

• Just hiring people” is sometimes still actually possible

• Ophidiophobia: Fear of snakes.

• “Et tu, Ilya?”: Trying to make the case that Ilya was jealous of Sam’s achievements and that’s why he tries to oust Sam.

• Chuck Hagel: 24th United States secretary of defense from 2013 to 2015 in the administration of Barack Obama.

• Neil Wiley

• The Mainstreaming of Loserdom: Yeah, sorry, not having hobbies isn’t cool. Go outside, use your brain, or do something with your hands.

• Oliver (chimpanzee): “chimpanzee once promoted as a missing link or “humanzee” due to his somewhat human-like appearance and a tendency to walk upright.” See also humanzees.

• Claude 4.5 Opus’ Soul Document

• Underrated reasons to be thankful V: Dynomight’s fifth edition of his Thanksgiving classic.

• Federal prosecutors in Eric Adams case resign after being put on administrative leave: Integrity isn’t dead!

• Damian Williams (lawyer): “served as the United States attorney for the Southern District of New York from 2021 to 2024. He has been involved in the prosecution of numerous high-profile individuals, including Ghislaine Maxwell, Sam Bankman-Fried, Sean Combs, Mayor Eric Adams, and U.S. Senator Bob Menendez.”

• Richard C. Wesley:

  Wesley has described himself as “conservative in nature, pragmatic at the same time, with a fair appreciation of judicial restraint,” adding that “I ... have always restricted myself to what I understand to be the plain language of the statute. ... As long as the language is plain, we should restrict ourselves.”[6] He aims to write opinions that satisfy what he calls the “Livonia Post Office test”—that is, they are understandable to his neighbors back home.

• Mamdani, Trump Meeting Wasn’t Just Smiles

• The “tasting day”: why buying 5 babkas at once is an underrated source of meaning: I’ve done similar but with walking to each place—I call it a food crawl. Choose your food, find X restaurants within walking distance of each other, and get walking (and eating). I find the walking between leads to great convos, fun discoveries, better digestion (see verdauungsspaziergang), and less guilt about all the delicious food you just ate.

• Stop Applying And Get To Work

• Alpine Starts

• A Day’s Bookends

• Not stepping on bugs

• emails i’ve sent

• how to actually adjust your sleep schedule

• You’re WIBNO: “Warmly invited by not obligated”. I’ve been searching for something like this for a long time and have finally found it.

• tips on packing for a trip effectively*

• friendships shouldn’t be seen as ledgers of obligation

• Billionaires spending lots of money on things is consistent with how you (probably) live your life

• prompts to stare into the abyss

• project ideas i hope someone steals from me: The typesetting ones are awesome, but (probably) require sooooo much work. Now I wonder how a SOTA LLM would do at typesetting something in LaTeX or similar. A FancyBookGPT that outputs an entire book given text would be pretty neat.

• Snake Island (Ilha da Queimada Grande)

• The Planes, Soviet Trains, and Rare Automobiles of North Korea

• Andrei Lankov: Russian North Korean expert based in South Korea.

• A pattern to the best events I’ve run

• How to Clean when you Hate Cleaning: A straightforward guide to cleaning for those that either hate or don’t know how to clean.

• Is it time for Post-Stoicism?

• Various ICBM speeds animated: The whole “hitting a bullet with a bullet” explanation didn’t really click for me until I saw this—missiles move insanely fast. Couple that with multiple warheads per missile (see multiple independently targetable reentry vehicle (MIRV)), decoy missiles, etc. and you have a really difficult problem to solve.

• ICBM address: “hacker slang for one’s longitude and latitude (preferably to seconds-of-arc accuracy) when placed in a signature or another publicly available file.”

• Interiors can be more fun: Ideas on how to make interiors less boring and more fun.

• Favorite quotes from “High Output Management”

• Question the Requirements

• Sanjay Shah: “a British trader who was sentenced by a Danish court in 2024 to 12 years in prison for tax fraud, the heaviest penalty ever handed out in Denmark for a fraud case.”

• Two easy digital intentionality practices: The first is go for a walk without your phone (I think this can be more generalized into “do something without your phone”) and the second is to switch phones with the person you’re with. I found the first surprisingly hard not because of willpower, but sheer habit. My phone lives on my person and it feels weird to not have it with me when leaving the apartment.

• “You’re not sick enough for this medicine.”

• The Bleach Bottle is Empty: Learned helplessness can start in childhood and follow you to adulthood.

• Always mask at airports: Especially when in lines and during taxiing. The “if you don’t do it all the time it’s worthless” argument continues to fall flat: total viral load matters! Any reduction in the amount of contact, whether by space or mask, is better than no reduction, hence masking being worth it.

• Curtis Priem: Nvidia cofounder. “In November 2023, Forbes estimated Priem’s net worth to be approximately $30 million; if he had retained his shares in Nvidia, Forbes estimated that Priem would have been worth $70 billion.”

• Things I Learned from the Fatima Discourse™

• Ilya Sutskever deposition: Some extra details about the OpenAI coup and what led up to it. I find the constant objections and bickering humorous. It’s also interesting that Ilya doesn’t know who’s paying for his legal counsel. Maybe it’s a future superintelligence ensuring he doesn’t go broke on his path to creating said superintelligence.

• Birthday on the Charmoz

• You’re always stressed, your mind is always busy, you never have enough time: It’s amazing how screens have captured and held our attention with a one-way ratchet that’s incredibly difficult to break out of.

• A theory of performative engagement: or, how power actually works on Twitter and Substack: I constantly feel like this is the case with LinkedIn, Substack, and Twitter replies, especially those starting with platitudes like “X, thanks for posting this. Here are my very generic thoughts on it to increase the number of views my account gets in hope that someone rich and powerful sees”.

• The Kill Pause: Kirkpatrick discusses the tragic death of Balin Miller, a climber who died after rappelling off the end of his rope. It’s amazing what fatigue, competence, and impatience can cause some people to do.

• Finding It

Philosophy

My interest in exposing this website to more readers has fluctuated over time. At first it was exclusively a place for me to “organize thoughts, interests, and happenings in my life”. Eventually I changed mindsets, believing that some select posts may be useful for others (although I never attempted to promote this site anywhere, so my mindset was practically the same). I considered implementing some basic SEO on my site in hopes of attracting more readers, but chose not to for a few reasons.

My mindset has finally settled since then, and unless some money comes my way and I sell out, I expect this to be my final stance: if the content I produce is good enough, readers will eventually find my site without SEO.

The vehicle for spreading blog gospel is almost always the readers themselves. If the content is good enough, it will be shared in one way or another—forums, social media, email, other blogs. If the content is good enough, it will be read in full and its lessons and thoughts discussed and heeded or discussed and opposed. If the content is good enough, readers will desire more. If the content is good enough, readers don’t need to be tricked or click-baited into reading it—its reputation will do that. If the content is good enough, the author won’t beg the reader to sign up for a mailing list—the reader will search one out, and even ask the author if nothing is found.

My mindset is my goal: to create content that’s such high-quality that it promotes itself.

This idea is similar to Zvi’s Play in Hard Mode, specifically example IV:

You start a website writing articles devoted to the things you care about. To monetize it, you sell advertising through Google. It does not pay much at first. You keep at it, attracting a small but devoted readership. Some were already your friends, others soon join them. You look at what resonates so you can get feedback, but are careful not to take actions designed to maximize page views. Over time your writing improves and you learn much together. A community of sorts arises. You don’t quit your day job, but you teach others what you have learned.

SEO is a crutch used to mask inability at the expense of awareness and improvement. While I’m sure some SEOers are aware their content is atrocious and the only reason they get readers is because of SEO, I’m also sure others aren’t aware of it. They think their content is decent, that it’s not just regurgitated, unoriginal garbage designed to get as many page views as possible and line their pockets with that sweet AdSense cash.

This is detrimental to both the producer and consumer. The producer is misled by their bogus traffic stats, choosing to be complacent about their ability and continue with the same quality of content as what’s been getting them the traffic. The consumer is again led into a predatory site which offers little of value, but takes advantage of them and their time.

My Second Law

This leads to what I call My Second Law, which states that:

The less a website’s name sounds like the topic it’s devoted to, the better the information it contains and/or produces.

(Of course there are some major exceptions, but it seems to hold true for a large amount of cases.)

In searching for websites detailing prepping, the following websites were Google’s first page results after searching “prepping 101”:

• primalsurvivor

• survivalsullivan

• readylifestyle

• thesurvivalmom

• theprepared

• pewpewtactical

• survivalist101

• thepreppingguide

All were riddled with ads and links to other pages on the same website, and while the information was occasionally good, the vast majority wasn’t. I finally stumbled upon lcamtuf‘s Disaster planning for regular folks, arguably one of the best resources I found discussing all aspects of prepping without pop-up ads or incessant linking. The HTML is concise and offers no evidence of SEO methods, the writing straightforward and organized, and the information self-contained, i.e., no linking to other sites or intra-site.

This is the epitome of against-SEO-ism (whether the author intended it to be or not). The fat has been trimmed, leaving nothing but quality for the lucky reader that stumbles their way onto the practically undocumented site.

Spreading the Gospel

Those who produce high-quality work without implementing SEO deserve to be recognized for their efforts and patience. Share one of their links. Discuss with a friend. Do something that helps them get closer towards the readership numbers they deserve, but don’t have. After all, it’s not just for the producer, but the unaware readers that don’t know what they’re missing.

I dream of the day I can leave my bike unlocked on the back of my car while I run into the store. I dream of the day I can leave my laptop unattended in the coffee shop while I go to the restroom. I dream of the day kids can be told “just be home before dark” again.

I dream of these because I live in the fairly low-trust society of the United States. Of course, there exist enclaves of different levels of trust, e.g., Memphis is very low trust and the average Maine town is very high trust, but I’d argue that most of the big U.S. cities are on the lower end, or I’m just projecting and/or bringing my baggage with me when visiting other cities.

The U.S. Joint Economic Committee Republicans authored The Geography of Social Capital in America report (interactive map here) that discusses what they refer to as the “social capital index” as defined in the report. The data is fairly robust, but I’m not going for data-based measurements. I’m going for vibes.

Measuring Trust Vibes

I propose a few different tests to measure trust in a new place. Here’s a short summary:

Test 1: Unlocked Bicycle

Ah, the quintessential test of if scumbags are abound. Bicycles are easy to steal, easy to get away with, and difficult to track. The perfect theft.

Ride up to a bike rack, leave the bike there, and start the timer to see how long it takes to get stolen.

The shorter the time it takes for the bike to get stolen, the lower the trust score. The longer the time, the higher the trust score.

Test 2: Dropping Valuables

An honest person should let you know if you dropped something valuable—money, jewelry, etc—and didn’t notice it. A dishonest person wouldn’t say anything and just pocket the goods for themselves.

Walk along a random street where people can see and “accidentally” drop something of value. Do this multiple times to find the rate at which people let you know that you dropped something.

The lower the rate of people saying something, the lower the trust score. The higher the rate, the higher the trust score.

Test 3: No Cash to Pay

Trusting someone to pay when they don’t have money is trust by definition.

Pretend to have forgotten a payment method (cash, credit card, phone) after receiving a service (food, haircut, etc) and promise you’ll be right back to pay. Do this multiple times to find the rate of businesses letting this happen.

The lower the rate or having to try hard to convince them, the lower the trust score. The higher the rate or the less convincing, the higher the trust score.

Test 4: Out After Dark

As my father once said, nothing good happens after 9:00pm. The type of society he grew up in can probably be inferred. But in high trust societies it shouldn’t matter what time it is or where somebody is at because it’s safe everywhere at all times!

Drive through the streets at night and see the demographics of people that are out and about.

The fewer women and children, especially if it’s not that late, the lower the trust score. The more women and children, the higher the trust score.

(Before testing this, it’s important to be inconspicuous so as to not ruin the testing environment. You’ll want to make sure to get a white box van with tinted windows so people don’t know they’re being watched. You can also park far away with lights off so nobody knows there’s somebody in the van. If someone notices you, drive off quickly to avoid extra attention.)

Why This Matters

While the tests are designed somewhat in jest, they are meant to provide information about the community to the market to help them to make informed decisions. Businesses can choose to not to set up shop there; residents can choose to not live there; tourists can choose to not travel there.

The point is quickly made by potential newcomers for already-established residents: if you’re low trust, you will not thrive.

A low-trust community improving its trust probably requires more than just potential customers avoiding the area, but realizing the shortcomings is the first step, preceded by quantifying the shortcoming (even if it’s based on vibes) to help recognize that it’s low.

See Also

• Social Defection: Rules separate us from the animals (or in this case, low-trust societies)

• Enlightenment in Iceland: When I discovered what high trust really means

• Why Are Little Kids in Japan So Independent?