jcdavis

Modern signalling systems are basically adsb for trains. (ETCS is the modern standard in Europe, this line was using the older German LZB). Problem is that they can, and do, fail. And when that happens too frequently (as it seemingly has been in Spain due to a lack of maintenance), its easy to get sloppy on the manual handling

There is definitely some, but the #1 advice is excellent advice pretty much wherever you go IMO. Beyond just scheduling: You absolutely have to be your own advocate for all sorts of stuff that will otherwise often get dropped through the cracks.

Airport jobs are some of the worst possible jobs - poor pay, bad hours, very long effective commutes (when you consider that they don't clock in until after get to the terminal, get through security), and angry customers screaming at you for something that's not your fault.

Lazy/incompetent/annoyed employees is the inevitable outcome of such an environment

I've read a looot of physiology studies about running, and the same dynamics hold true there. This is a flaw (IMO) of of the whole "Norwegian 4x4 protocol" you hear parroted around by biohacker/longevity influencer online, of course with 0 nuance or context of the underlying study and its limitations

I'll agree that "everyone" is probably an unfair characterization. But the tax benefits of depreciation are wildly touted among real estate investors.

If they didn't claim depreciation in prior years they can still get it via Form 3115. Yes this is complicated/annoying to do (almost certainly need a CPA), which you can argue is unfair, but I'm still going to have limited sympathy for anyone DIYing in this space without talking to a professional.

Hard to sympathize with the landlord class too much on this one. Everyone knows how depreciation schedule works and gets in to it in no small part because of that deduction benefit + the hopes that via 1031 exchanges etc they can delay it until death.

Its an extremely annoying trend among a subset of the tech industry who think it makes them cool

Its a wild time to be in software development. Nobody(1) actually knows what causes LLMs to do certain things, we just pray the prompt moves the probabilities the right way enough such that it mostly does what we want. This used to be a field that prided itself on deterministic behavior and reproducibility.

Now? We have AGENTS.md files that look like a parent talking to a child with all the bold all-caps, double emphasis, just praying that's enough to be sure they run the commands you want them to be running

(1 Outside of some core ML developers at the big model companies)

Context: I thought it would be fun to use the github-style year visualization for my running, while also letting me get a more thorough detail of efforts (pace + HR). All data is coming from intervals.icu (since they are much more developer-friendly that strava), and 90% vibe-coded via cline. Source: https://github.com/jcdavis/running-wrapped

Based off of my first ever forays into node performance analysis last year, JSON.stringify was one of the biggest impediments to just about everything around performant node services. The fact that everyone uses stringify to for dict keys, the fact that apollo/express just serializes the entire response into a string instead of incrementally streaming it back (I think there are some possible workarounds for this, but they seemed very hacky)

As someone who has come from a JVM/go background, I was kinda shocked how amateur hour it felt tbh.

Full text: https://archive.is/5jgNR

Its roughly a 2x2 block area of The Mission (not a hard boundary to participating, but almost everyone lives in it). I won't get more specific than that in case the author doesn't feel comfortable since it wasn't mentioned in her post.

I saw this and did a double-take - I live in the neighborhood and am fortunate enough to be a part of this community. Patty, Tyler, and Luke have done a tremendous job of creating a communal bond that makes everyone feel valued & welcome.

I now know 50+ people who live within ~2 blocks from me, who've gone from "random strangers" to "friendly neighbors" that I run into semi-randomly.

Still $3.50 at TJs in SF last week still, which is by far the cheapest around (that I'm aware of).

Pretty surprised they are still that low given prices elsewhere.

> it screwed up the API for C++, with many compromises

The implicit presence garbage screwed up the API for many languages, not just C++

What is wild is how obviously silly it was at the time, too - no hindsight was needed.

Yes: https://www.bloomberg.com/account/newsletters/money-stuff?so...

> Spotify pays around 70% of its revenue to their artists

They pay 70% of revenue to rights holders. For an artist signed to a major label like Lily Allen, they'll get ~20% of that number after they've cleared their advance.

Fun writeup. When I went through and implement the book in rust (https://github.com/jcdavis/rulox), I just used Rc and never really solved the cycle issue.

I'll +1 and say I highly recommend going through Crafting Interpreters, particularly as a way of building non-trivial programs in languages you are less familiar with - If you just follow the java/C example you are tempted to lean into copy/pasting the samples, but figuring things out in other languages is a great experience.

I spent a longass time tracking down an issue with how the reference interpreter implementation defines tokens: https://github.com/munificent/craftinginterpreters/issues/11... which was frustrating but good debugging practice.

For an absolute wild docuseries about one of these related scams, check out Telemarketers on HBO.

username checks out

The inverted method is an even easier way to achieve this :)

Getting new prior auths every year is annoying but not exclusive to Optum, thats been the case with all the specialty pharmacies I've gotten prescriptions from.

And, crucially, last year 80% of VUSXX's interest was state tax exempt.

Unless you specifically need the features of a savings account (eg ACH), money market funds are the way to go.

> The book couldn’t have been that good if it never got across to you that an airplane hitting production in 1997 was necessarily designed YEARS BEFORE 1997 (and hence the merger).

Where did I say otherwise? I'm well aware of how old the 737 is, but this particular feature (the additional exit door in the frame which is generally plugged) was introduced in the -900

For folks who want to feel better: this particular design feature (the unused plug exit) dates back to the 737-900 which has been around since 1997

For folks who want to feel worse: Highly recommend "Flying Blind" (https://www.amazon.com/gp/product/B08P98854S) - its a good book about how Boeing's engineering culture went to shit after the MD merger and the shortcuts used to get the Max delivered.