Lightnews — Scholar-powered news

LightNews

Joe Fabisevich

@mergesort.me

3.9K followers 71 following 4.3K posts

“Apps genius” - Colleen. I used to work on making Twitter a bit healthier, then that all went to hell. Now I make puns and indie apps @plinky.app. Born and raised New Yorker, trying to do a little good and be the friend you made along the way. 🍕🐱⚾️

Posts Media Videos Starter Packs

Pinned

Joe Fabisevich @mergesort.me · Nov 9

Hi, I’m Joe! I used to work @ Twitter on all the hard things like misinformation, disinformation, harassment, and civic integrity. I also spent a ton of effort on little improvements to make people’s lives better. These days I’m an indie developer making it easy to save links for later, with Plinky.

Plinky helps you easily save links from any device, so you can enjoy them later. Save on your iPhone, iPad, Chrome, Safari, Firefox, Zapier, and so much more.

Joe Fabisevich @mergesort.me · 11h

You got my vote. 🌟

Joe Fabisevich @mergesort.me · 13h

Nm what’s hypha witchu?

Joe Fabisevich @mergesort.me · 13h

No true HATEOScotsman

Joe Fabisevich @mergesort.me · 14h

Isn’t the answer cryptocurrencies?

Joe Fabisevich @mergesort.me · 14h

Fuck yeah.

Joe Fabisevich @mergesort.me · 16h

Now that we have a name 50% of the work is done. I’m a little mad that I have absolutely no time right now, otherwise I’d quickly whip something up with Codex.

Joe Fabisevich @mergesort.me · 16h

Lorem atsum

Joe Fabisevich @mergesort.me · 1d

I'm kinda done debating because you're making it clear you didn't even read the research you cite. (I actually have read the papers and you can't extrapolate from 16 developers in a randomized trial, 8 of whom had never used Cursor — a few of didn't even know the features Cursor offers.)

Ruben Bloom (Ruby) on X: "I was one of the developers in the @METR_Evals study. Thoughts: 1. This is much less true of my participation in the study where I was more conceintious, but I feel like historically a lot of my AI speed-up gains were eaten by the fact that while a prompt was running, I'd look" / X

I was one of the developers in the @METR_Evals study. Thoughts: 1. This is much less true of my participation in the study where I was more conceintious, but I feel like historically a lot of my AI speed-up gains were eaten by the fact that while a prompt was running, I'd look

Joe Fabisevich @mergesort.me · 1d

I’m well aware of this subject matter but what you’re describing is a different problem. For example workslop is a matter of laziness and incentives, mediocre results from people who don’t use AI in a novel way but instead regurgitate whatever it says verbatim. That’s not who I’m talking about.

Joe Fabisevich @mergesort.me · 1d

Current models simply do not hallucinate 30% of the time. If you believe that than odds are you haven’t tried any model that was released in the last six months, because the amount of progress they’ve made on the hallucination front is wildly good, especially for software development subjects.

Joe Fabisevich @mergesort.me · 1d

Nothing was stopping them, but people learn differently. Have you ever met a developer who learns better from video than reading? Or by doing than sitting in a lecture? There are a lot of people who will learn better from AI (and many worse or far worse), but you can’t disqualify them if they learn.

Joe Fabisevich @mergesort.me · 1d

A lot of people see X posts about vibecoding from $0-100K and underestimate how many thousands of people there are out there who are using AI as an on-ramp to learning more about the technical subjects you say they will discover and cry about. Far from everyone of course, but I keep an open mind.

Joe Fabisevich @mergesort.me · 1d

See this is the perfect example of what I don’t think is universally true, but is viewed derisively as if it is. I work with people who over the last six months have gone from Uber drivers or teachers to building real software — only — because they’re curious and interested in these exact things.

Joe Fabisevich @mergesort.me · 1d

I don’t think it’s my default assumption based on what I’ve seen working with people who have no technical experience because AI does a good job of defaulting to best practices *but* I definitely think it’s plausible or likely that we’ll be annoyed when non-technical people contribute to codebases.

Joe Fabisevich @mergesort.me · 1d

Wouldn’t a similar argument apply I front end developers today vs. 15 years ago when many people chided them as “fancy designers”? Or constructing a scenario, is there a future where someone who has product instincts can deliver “meaningful” code/results in a team despite less technical training?

Joe Fabisevich @mergesort.me · 1d

This is definitely a question of where do you draw the line of “a software developer”. It’s allowing more people to build meaningful software, but I recognize that those are not the same. On the other hand most developers today know zero about infra that was required to do software 15-20 years ago.

Joe Fabisevich @mergesort.me · 1d

And sea lions. 🐈‍⬛

Joe Fabisevich @mergesort.me · 1d

When you come back home to see your cat has transformed into a sea lion. #caturday

A black cat sitting on a railing

A black cat stretching out on a railing and looking up

Joe Fabisevich @mergesort.me · 2d

Researchers can have a lil loop, if they behave well.

Joe Fabisevich @mergesort.me · 2d

The irony of why you’re going viral is not lost on me.

Joe Fabisevich @mergesort.me · 2d

I could be convinced. 🙂

Joe Fabisevich @mergesort.me · 3d

This is exciting as long as Elastic doesn’t ruin Jina. 😭

Joe Fabisevich @mergesort.me · 3d

Worked great, zero complaints. 😄

Joe Fabisevich @mergesort.me · 4d

Swift uses `keyword` which feels like a reasonable trade off and not too ugly. Admittedly TypeScript is a hobby and isn’t my lingua Franca so I’m not sure that would fly with template literals.

Joe Fabisevich @mergesort.me · 4d

I mean, it was good to *me* so I’m content without another Codex. 🙂