davely

So you can read Hacker News on your watch.

I find myself doing this randomly when I don’t have access to my phone (believe it or not, it happens for legit reasons) through something called ųBrowser.

It’s probably masochistic but it works in a pinch. I’ve been thinking of writing my own watch app to do this, as well as an app that could use the Feedbin API.

Not entirely unexpected given that these models have 2x - 5x more parameters than the 120b gpt-oss model.

It’s impressive their performance gets so close given the size!

The study that said devs were 19% less productive was flawed in a number of ways: small sample size, poorly chosen tasks, no prior training on AI tools.

The science is by no means conclusive. A joint Harvard / MIT study found developers completed tasks up to 50% faster [0]!

[0] https://arxiv.org/abs/2302.06590

All 15 comments (so far) on this are by green accounts since it was posted 3 hours ago. This makes me mildly suspicious.

Edit: All accounts have been registered since this post went live.

Excuse me! If that truly is the case, then why are you sending out 418 status codes?

Mobile phone manufacturers were telling users this long before the iPhone was ever invented.

e.g., Nokia 1600 user guide from 2005 (page 16) [0]

[0] https://www.instructionsmanuals.com/sites/default/files/2019...

I don’t know if it’s AI, but where I live in the Bay Area, Waste Management spies on the contents of our recycling bins with cameras when the truck picks them up.

I have gotten warning emails with evidence of some transgression or another about 4 times since November 2024. All it says is “we found that at least one of your carts was filled incorrectly.”

Cool!

The best thing is the evidence they provide. It’s a blurry and pixelated 320 x 320 image that shows what could literally be anything.

I lived in a semi-rural area with some foothills behind our property growing up.

I distinctly remember that our neighbor had a number of different cats over the years. They were never around that long. He always named them _C.B._

I never thought much of it, maybe he was just a fan of chatting of radios.

I was in my teenager when I finally realized what C.B. stood for in this context: Coyote Bait

My house has never been cleaner. I have time to catch up on chores that I normally do during the weekend. Dishes, laundry, walk the dog more.

It seems silly but it’s opened up a lot of extra time for some of this stuff. Heck, I even play my guitar more, something I’ve neglected for years. Noodle around while I wait for Claude to finish something and then I review it.

All in all, I dig this new world. But I also code JS web apps for a living, so just about the easiest code for an LLM to tackle.

EDIT: Though I think you are asking about work specifically. i.e., does management recognize your contributions and reward you?

For me, no. But like I said, I get more done at work and more done at home. It’s weird. And awesome.

git worktrees optimize how data is shared across multiple directories. So, you’re not cloning and duplicating a bunch of code on your machine and just referencing data from the original .git folder.

The only downsides I’ve seen:

1. For JS projects at least, you still need to npm install or yarn install for all packages. This can take a bit of time.

2. Potentially breaks if you have bespoke monorepo tooling that your infra team won’t accept updates for, why is this example so specific, I don’t know just accept my PR please. But I digress.

I work for a tech company you’ve definitely heard of.

I use the “llm” tool every single day. You may not know it, and that’s okay, but Simon’s tools are an immense help to tons of developers using LLMs.

I know it’s fun and trendy to hate on LLMs, but if you’re not productive with them at this point, you’re either:

1. Working on a novel problem or in some obscure language, or

2. Have a skill issue related to how to utilize LLMs.

I did a system design interview awhile back using Excalidraw. I really enjoyed it (though I also frequently use it for normal work stuff, so was familiar with it).

But the amount of time saved because one could vibe code it rather than search through Google’s terrible results is immense and a benefit in and of itself.

Maybe this is a blind spot for me, but my inclination is to think it doesn’t matter?

Only because the company has already made you an offer in the first place, ergo they see your value and are already making a bet on you. Has there ever been an instance where an offer has been rescinded because a prospective employee didn’t negotiate?

> say “I'm looking for $150K to $160K…”

I’ll defer to your experience on this but I am under the impression that giving a range was a bad idea because a company can always lowball you: “oh, you said you were okay with 150K, why are you trying to renegotiate with us now?!”

> Companies generally respect candidates who negotiate

Is this really the case? If a company is giving you an offer, presumably they want to hire you. At that point, does the recruiter, your hiring manager, your skip level, the CTO, the CEO even care at that point? Does this respect ever come into play in future performance reviews? I personally don’t think so, but I’ve only ever worked for larger companies.

I’ve always thought you should negotiate because you’re potentially leaving money on the table, since companies will often try to lowball your offer.

There’s also the phenomenon of intra-industry trade. One big example is natural gas, which the US exports to Canada and Mexico, while paradoxically importing natural gas from Canada.

Beef and pork are another example I can think of though, this has to do with specific cuts and quality for various markets.

I just commented about the same exact thing relating to my wife. Even down to 20+ siblings!

Like, it’s so common that I’m not even shocked to hear others with this experience.

As it turns out, there were a lot of sketchy clinics preying upon desperate parents in the early 1980s.

This exact scenario happened to my wife. Grew up thinking her dad was her dad and that she was of some certain ancestry.

I remember her and I both spitting into a tube in our SF apartment in San Francisco in 2012 and mailing them off.

A few years later, we look at the ancestry data and there is a big surprise (she was ~50% something that she was sure 0% before that).

Turns out her dad wasn’t her dad and she was donor conceived — this has potentially huge health implications if you don’t know this.

By some fortunate circumstance, she ultimately found the donor… and 27 (and counting) siblings! Turns out the clinic where this all went down at in the early 1980s was sketchy and lied to a whole bunch of people and misused samples.

They are for sure going to lock this down.

I’ve gone through $25 USD in API credits in a single afternoon with Claude Code (I love it, but that thing is thirsty for dollar bills).

I’ve been reluctant to try this sort of thing out because it’s fairly trivial for them to detect this and potentially come down with the ban hammer. I’d rather not risk it.

Anyway, I’ve found myself switching back to Aider as of late because it is much more conservative in how it uses its token budget.

To be fair, a prompt fed into a generative tool _could_ be considered an artist's creative expression.

I wonder about something like this[0]. So much awesome engineering went into it. And the guy is clearly an artist and considers himself an artist[1]. As it is his own tool, are the random splatters it generates not copyrightable?

[0] https://www.youtube.com/watch?v=9XyE41_ANrc

[1] https://engineezy.com/pages/about

Oof, this gets into all sorts of weird legal grey areas.

- All of our phones do a bunch of computational photography where AI tooling improves a photo in various ways. In that case, is any photo taken by a modern phone not copyrightable?

- If it is copyrightable, what if someone uses an Img2Img tool or inpainting with something like Stable Diffusion (or Photoshop) in order to slightly modify an image. Is that no longer copyrightable?

(FYI, my questions aren't directed at or attacking you -- just interesting hypotheticals.)

I have some VS Code extension (errr... not sure which) that faintly inlines the git blame result on each line of code you're working on. I find it kind of handy.

Sometimes you don't know which commit actually caused the problem.

e.g., you realize that something broke A/B test logic on Friday. Sure, there are Jira tickets, but that's slow and annoying to dig through. There are commit messages, but things get squashed, etc. Plus, if you work in a monorepo with about 60 PRs a day, it's hard to know if it was your code or an associated library someone touched.

That's exactly when git bisect helps. It quickly narrows down which commit introduced the bug when you don't know where to start looking. Once bisect identifies the problematic commit, you can then use git blame (if needed) to see who made those specific changes.

Edit: Cleaned up what I was saying to hopefully avoid confusion.

I like the blank slate analogy. I'm in the same boat.

I have tens of thousands of tweets (why, oh why) but personally have no desire to move them over. I realize that's important for some people. For me, it's enough to just download the archive and save it somewhere, in case I ever need to reference it again.

Pointing out this one error is reductive and greatly simplifies the profound effect that LLMs (and research around them) are having across a wide range of industries, unlike the metaverse and NFTs.

This made me curious, so I just took a look on eBay.

My 4070 Ti that I bought (new, oops) this past December has appreciated 20 - 25%! At least according to what people have them listed for... no idea if anyone is actually buying them at those prices.

Of course even if I managed to sell it, everything else has gone up in that time, so it's not like I'd get to make money on the deal. Pretty wild, nonetheless!

One thing I enjoy when working on a new feature is to schedule a team call for me to walk through the code, detailing the changes or improvements I've made, thought processes and overall just give better context into what the wall of text they are going to review is all about.

Likewise, I enjoy when colleagues take the time to do the same with their own PRs. It's something I've tried to promote, but I think everyone is just in a default mode of ship-ship-ship, so people aren't eager to have another meeting to attend.

(Ideally, we shouldn't be submitting PRs that are so huge, but sometimes this isn't always feasible.)

The example was in React but I don’t see how that prevents potentially using this with other frameworks or languages.

That said, I feel like this is learning something like CoffeeScript. You have to learn all this language specific syntax (though simplified), why not just learn the language itself at that point?

  Location: San Francisco Bay Area
  Remote: Yes
  Willing to relocate: No
  Technologies: JavaScript, TypeScript, ReactJS, NodeJS, NextJS, Styled Components, PandaCSS, React Native, Vue
  Résumé/CV: https://dave.ly/files/dschu-resume.pdf
  Web: https://daveschumaker.net
  Email: hello@daveschumaker.net

Hello! I am a full-stack engineer with 10 years of experience building map-based web applications. I enjoy working with other driven and motivated people to build fun and interesting products.

Recently, I've been pretty involved in AI related work and have given talks at my company on how engineers can better incorporate AI tools into their own work flows. Additionally, I develop an open source front-end for creating images using generative AI. More information is available on my resume.