primitivesuave

In this case, "bad things happening" is an exodus of the ultra wealthy, which is not so simple to reverse.

I actually read this earlier because you commented on another post about Charlie Javice's sentencing. Based on that sentencing and others, it seems like the going imprisonment rate for "losses incurred by the powers that be" is around $2-3 million/month of prison.

By that metric, you received an unduly harsh sentence. Would you happen to have some insight as to why?

Also, I sincerely appreciate you sharing this, and the means by which you shared it. I honestly think that with your situation (unlike Charlie Javice, Theranos, etc), anyone who reads your candid disclosure would interpret it in a positive way.

He also ran a series of now-defunct scam dating sites with fake profiles around 2010-2015, and has been quite consistent at only operating scam businesses in the past 15 years. I know nothing about the people who hand over money to this guy, and I hope to keep it that way.

No - you would have to file a FOIA request with the SFMTA, and then tabulate the "Total Paid" column in the response files. Below is my FOIA request template for the SFMTA [1] to assist anyone who is interested in doing this.

> Please provide all possible information on all parking citations issued between 2009 and the present day. This should include any information related to the car (make, etc), license plate, ticket, ticketer, ticket reason(s), financial information (paid, etc), court information (contested, etc), situational (eg, time, location), and photos/videos. Specifically, please provide the most recent data from the dataset I have received in past FOIA requests, with the following headers:

> Issue Year, Ticket Number, Tick Issue Date, Tick Issue Time, Agency, Tick Badge Issued, Veh Make, Veh Body, Tick VIN, Tick RP State, Tick RP Plate, Plate Exp Date, Violation, Violation Desc, Tick Meter, Tick Street No, Tick Street Name, Suspend Code, Suspend Desc, Tick Suspend Date, Tick Dispo Code, Tick Dispo Desc, Tick Dispo Date Total Paid, Total Amt Due

1. https://www.sfmta.com/public-records-requests

Just to confirm, this is what happened - he paid one twice, and therefore the other became delinquent. However, if you file a FOIA request with the SFMTA and do some basic analysis on how much is "owed per license plate", you will see that certain license plates have been allowed to accrue tens of thousands of dollars in parking fines with virtually no towing repercussions, going back as far as data is available (2013). Around 80-90% of these vehicles are rental cars, around 5-8% of them are commercial trucking companies which absorb the cost of 2 tickets per day, and this tiny 1% of license plates are effectively "lawless parkers" who drive high-end cars and accrue tens of thousands of dollars in unpaid parking fines every year.

I should have clarified - I only found his business through that list because I noticed his vanity license plate HPPYPPS, which corresponded to a business named Happy Pipes Plumbing which I subsequently found on Yelp.

Also, I found out about his van getting towed because I scraped towing records from Autoreturn (the city's main towing provider - lots of corruption around that deal). Autoreturn's website at the time had a query parameter like "?towid=1", so you could increment that to pull all towing records.

I started working on a pretty in-depth data analysis and visualization, similar to what was done here, but I got caught up with my day job and some rock climbing dreams. I handed over all my research to a few local reporters a while back - they were really excited to talk to me in person about it, but I haven't seen anything published since.

I did a fascinating analysis of SFMTA data a few years back. They posted a public list of names and license plates [1] that they refuse to take down, despite many emails from me over the years. I found a particular license plate that belonged to a plumber with an impeccable 5 star reputation on Yelp, whose business in SF was effectively ended by street sweeping fines. He accidentally paid the same ticket twice, which resulted in his work vehicle being towed for excessive delinquency on the original ticket, which culminated in him moving his plumbing business to Utah.

I mentioned his 5 star reputation because several people got on Yelp over the years and described situations where he wouldn't even charge them money if he could fix something in a few minutes. It was very sad to learn how the SFTMA ran an honest plumber out of our city, and still won't take his name down off the list below (even 8 years after the deadline to respond).

I don't mean to draw undue attention to that list - please bombard the SFTMA with emails to take it down, it is a very obvious invasion of privacy and laughably unnecessary.

1. https://www.sfmta.com/reports/escheatment-posting-october-20...

I use Aurora Postgres at work, where we pay approximately 9x more for equivalent resources to PlanetScale (according to their pricing page [1]). This is not an endorsement of PlanetScale as I've never used it, just pointing out that the premium for using Aurora Postgres is many multiples higher than virtually every other Postgres provider.

1. https://planetscale.com/pricing?architecture=x86-64&cluster=...

IMO it is a smart decision to implement this as a self-hosted system, and have the AI make PRs against the IaC configuration - for devops matters, human-in-the-loop is a high priority. I'm curious how well this would work if I'm using Pulumi or the AWS CDK (both are well-known to LLMs).

I consulted for an early stage company that was trying to do this during the GPT-3 era. Despite the founders' stellar reputation and impressive startup pedigree, it was exceedingly difficult to get customers to provide meaningful read access to their AWS infrastructure, let alone the ability to make changes.

Anyone's time is their own to decide how to spend it.

You seem to be trying to portray some kind of malicious intent by someone who in real life is a very nice and humble guy.

Just to stymy the spread of misinformation - this is in reference to an interview from 2019 or 2020, where he said he would prioritize women/minorities for mentorship. While I personally don't agree with that notion (IMO mentorship should go to the most qualified person, regardless of race/background), I definitely didn't hear it as a refusal to help white men.

Adding a link to the ZoP for the uninitiated: https://peps.python.org/pep-0020/

I suppose that for this case, an underground black market of VPN providers might emerge - average individuals setting up VPN software on a cloud service provider, and then selling monthly access to people. Aside from the obvious danger of getting ripped off (someone might put you on a slow shared VPN with many other people, or shut down the server at any time), there is also the possibility of someone monitoring all your Internet activity.

I just want to add to the chorus that your visualizations are outstanding.

I exclusively use Waymo in SF, even if it costs a bit more than Uber. You'll most often get a great human Uber driver, but there's a very real possibility that the person is a bad/unsafe driver or the ride is unpleasant for a myriad of other reasons. With a Waymo, you know exactly what you're buying.

The output quality is remarkable. You mentioned that there are 1 billion illiterate people who would benefit from this, and I would add that there are at least 1 billion additional people who would benefit because they speak a regional dialect. There are many countries across the developing world where the AI tools and translation apps only produce output in the official government dialect (e.g. the Thai spoken in Bangkok, the Hindi spoken in Delhi, or the Mandarin spoken in Beijing). It would be interesting to see how a voice model could be "fine tuned" to better serve a specific regional dialect.

First want to say that I sincerely appreciate you working on this problem. The proliferation of deepfakes is something that virtually every technology industry is dealing with right now.

Suppose that deepfake technology progressed to the point where it is still detectable by your technology, but is impossible for the naked eye. In that scenario (which many would call an eventuality), wouldn't you also be compelled to serve as an authoritative entity on the detection of deepfakes?

Imagine a future politician who is caught on video doing something scandalous, or a court case where someone is questioning the veracity of some video evidence. Are the creators of deepfake detection algorithms going to testify as expert witnesses, and how could they convince a human judge/jury that the output of their black box algorithm isn't a false positive?

Brought back memories to over a decade ago with Mathematica. It's quite easy to pattern match simple expressions like:

``` expr = foo[bar[k], baz[V]]; expr /. foo[x_, baz[y_]] :> {x, y} ```

But in real-world use cases, this paradigm quickly exposes some unhandled edge cases, or cases where the evaluation becomes prohibitively expensive. Nowadays I use TypeScript, and this has ignited some curiosity into finding out if TS does anything to optimize for this in type inference.

Fundamentally, Postgres is storing rows in 8kb pages of tuples, where each tuple has transaction metadata like `xmin` and `xmax` which are used in MVCC. This means a transactional update is usually only updating one block of memory + any TOASTable columns. Meanwhile, ClickHouse is storing data by columns, so a transactional update on multiple fields of a row necessarily involves updating multiple memory pages.

Personally, I architect systems to use Postgres for the write-heavy parts of a workload, and ClickHouse for the write-once parts (time series, analytics, logs, etc). ClickHouse is also the best tool I've ever found for compressing enormous datasets, and is very useful even as a simple data warehouse.

I completely agree that people should be able to enjoy Yosemite without needing to "rough it", but realistically this is something that is only accessible to people who can afford to spend upwards of ~$300/night (that's the room rate in El Portal, right outside the park entrance). Factoring in the park entrance fee, transportation, and food, it's easily a $1k weekend trip - not something the average American has access to.

The campsites are affordable and several have nice bathrooms/showers, but they get snatched up en masse at the beginning of the season by tour operators, so your only hope of getting one is by monitoring campflare.com.

We all collectively support an enormous commercial venture in and around Yosemite Valley, so we all collectively contribute toward increasing the price of entry and thus making the valley less accessible for everyone else.

I'm a regular in the Yosemite rock climbing scene, and Aramark is a universally despised entity for many reasons (gross negligence, outright lies, rampant neglect). One of their executives recently lost his job for hitting golf balls into Ahwahnee Meadows.

However, until people would rather camp than stay at the $600/night hotel, and cook over a campfire rather than eat overpriced food, there will always be a lucrative opportunity in Yosemite Valley that is only available to the most politically-connected companies.

The leaderboard is fascinating. Some people are clearly putting a lot of time into this, while the rest of us are trying to sneak phallic shapes past your CNN.

1. https://drawafish.com/rank.html?userId=1753510318634_cdeh6a4...

This is really interesting and I'm sure it has many useful applications - what I am most impressed by is the popovers in your dashboard, which show SDK examples. This is a brilliant UI idea (executed with great polish) and is the first time I'm seeing it.

Unfortunately I'm not a potential customer and don't have useful feedback on the market landscape - all I can add is that I really love your design. Also it seems like you all launched the new landing page while I was typing this comment a little past midnight, so kudos on the work ethic as well.

Thank you for the heads up - I made some poor React decisions when I was getting started for the sake of expediency, and now need to untangle them.

Working on https://ts.coach, a free guide to Typescript that runs entirely in the browser.

There is a US Treasury API endpoint [1] to see how much money actually comes in.

1. https://fiscaldata.treasury.gov/datasets/gift-contributions-...

If you are searching for some insight into human nature and intellect, you may find the history of the Roman Republic (and it's transition to an empire) to have shocking parallels to modern-day events. Trump is remarkably similar to Sulla, who showed the next generation of leaders how to break the rules to gain power. Caesar is coming...

For me personally, I only care about the tab autocomplete. If that's a feature I have unlimited access to, I will happily pay $20/month for 10x productivity.

These are awesome pointers. I especially agree with your first point about writing verbosely - I would do this all the time when I was teaching kids. I would copy and paste things and write all kinds of hairy logic, so they can see how useful it is to use different syntax and apply some refactoring.

I taught kids how to code for several years, both in person and through a website that taught Python to ~500k kids. My eternal advice with teaching CS is to delay abstractions as long as possible. Teach them how to draw a circle on a canvas, then get them to move the circle to different parts of the canvas, then get them to draw multiple overlapping circles/rectangles/etc. Once you've got that down, make a bouncing ball animation, simple games, and so on. Most kids don't really care for leetcode problems these days, so it's best to stick to things that are visual, creative, and instant-feedback (i.e. games).

This was a really useful realization when I was teaching kids to code in person for ~5 years. I started off teaching Java with an abstraction layer (https://www.bluej.org/), but I found that many students would be intimidated by anything that ventured outside of the abstraction (e.g. just run `javac` on your computer with a `.java` file). My most successful students intuitively understood how to turn a text file on their computer into a useful piece of compiled code, or into a JAR bundle that they could share with others. They also used the various abstractions that were thoughtfully introduced over time - e.g. an IDE that gives you a run button, libraries like Firebase which enable some exciting multiplayer game options, and so on.