panstromek

4 ms·

31.
The key insight for programming with LLMs effectively(yoyo-code.com)
2 points by panstromek 7mo ago|0 comments
32.
by panstromek 7mo ago
> It feels a little tricky to square these up sometimes. In my experience, this heavily depends on the task, and there's a massive chasm between tasks where it's a good and bad fit. I can definitely imagine people working only
33.
by panstromek 8mo ago
> There must be a really good reason for this, such as Rust doesn’t interop well with C++ Yea, I'd bet it's that. Ideally, you'd want to stop writing C++ and continue with Rust on all new code, but Rust has stricter semant
34.
by panstromek 8mo ago
Lot of DOM APIs are like that. You have methods like element.[parent|children]() which implies circular structure, and then you have APIs like element.click(), which emits a click event that bubbles through the DOM - which means that elemen
35.
by panstromek 8mo ago
> Modern C++ pretty much solves the safety issues. I always wonder how can one come to such a conclusion. Modern C++ has no way to enforce relationship between two objects in memory and the shared xor mutable rule, which means it can&#x
36.
by panstromek 8mo ago
> besides the obvious Well, what else is there besides the obvious? It's a browser.
37.
by panstromek 8mo ago
I don't see the idea is visual tools, I never even heard somebody to talk about it like that. The plan is to target enterprise customers with advanced features. I feel like you should just go and watch some interviews or something wher
38.
by panstromek 8mo ago
> what if the parsing step was performed immediately on stop symbols rather than later and merged with the program graph immediately rather than during a separate build step? I think this is the way to go, kinda like on Github, where you
39.
by panstromek 8mo ago
Interesting. It looks to me this was more about the portability of the resulting binary, IIUC. Dion project was more about user interface to the programming language and unifying tools to use AST (or Typed AST?) as a source of truth instead
40.
by panstromek 8mo ago
I think neither Unison nor Smalltalk use structural editing, though. [edit] on the level of a code in a function at least.
41.
by panstromek 8mo ago
IntelliJ diffs are also really good, they are somewhat semi-structural I'd say. Not going as far as difftastic it seems (but I haven't use that one).
42.
by panstromek 8mo ago
I'm referring to UI interfaces that allow you to do structural editing only and usually only store the structural shape of the program (e.g. no whitespace or indentation). I think at this point nobody uses them for programming, it&#x27
43.
by panstromek 8mo ago
I think agree (but I think I think about this maybe a one level higher). I wrote about this a while ago in https://yoyo-code.com/programming-breakthroughs-we-need/#edi... . One interesting thing I got in replies is Uni
44.
by panstromek 8mo ago
I have to respond to your point, though. Whether 30% cut is excessive depends on whether devs feel like they are getting a good deal. As far as I can tell, game developers don't seem to complain about Steam cut very much, it seems like
45.
by panstromek 8mo ago
I'm not judging that, it just seems to contradict the "But Steam shows us another model..." sentence, so I'm trying to make sense of that.
46.
by panstromek 8mo ago
Steam takes 30% cut, though?
47.
by panstromek 8mo ago
> Note: This image has been edited to include a pile of cash. I giggled
48.
by panstromek 9mo ago
So, when operating system gives you invalid file, it magically becomes valid, because your UI code is in a different file. Sure, that sounds plausible.
49.
by panstromek 9mo ago
There's so many ways this benchmark can go wrong that there's pretty much no way I can trust this conclusion. > All the loops call a dummy function DATA.doSomethingWithValue() for each array element to make sure V8 doesn't
50.
by panstromek 9mo ago
Our website codebase contains a workaround for a bug in native Android file picker in Samsung One UI. How are you supposed to solve this by "deploying your own platform?"
51.
by panstromek 9mo ago
No, you can't always do that. We have workarounds for platform bugs that were even fixed, because we get users with old devices that can't upgrade. You cannot fork a phone of a random person on the other side of the world. Once a
52.
by panstromek 9mo ago
> Maybe that's me, but I rarely saw teams which over-document, under-documenting is usually the case. This is a good point, although this recently changed with LLMs, which often spit out a ton of redundant comments by default.
53.
by panstromek 9mo ago
Sure, but you can't always fix the bug if it's not in your system.
54.
by panstromek 9mo ago
I also find that phrase super misleading. I've been using a different heuristic that seems to work better for me - "comments should add relevant information that is missing." This works against redundant comments but also isn
55.
by panstromek 9mo ago
It might look ok from user's point of view, but lot of the problems fall on web developers who have to work around a bunch of these issues to make their pages work in Safari
56.
by panstromek 9mo ago
Late on a lot of standards, quirky in many ways and just a lot of bugs, especially around images and videos. Also positioning issues. They recently broke even position fixed, which broke a ton of web pages on iOS, including apple.com
57.
by panstromek 9mo ago
I like this, especially because it focuses on the actual problem these contributioms cause, not the AI tools themselves. I especially like the term "extractive contribution." That captures the issue very well and covers even non-A
58.
by panstromek 10mo ago
I have bumped into this myself, too. It's really annoying. The biggest footgun isn't even discussed explicitly and it might be how the error got introduced - it's when the struct goes from POD to non-POD or vice-versa, the ru
59.
by panstromek 10mo ago
If you deploy on phones, CPU and memory is a major problem. Pick a median Android and lots of websites consisently fail to deliver good experience on it and it's very common to see them bottlenecked on CPU. JSON is massively innefficie
60.
by panstromek 10mo ago
I think he has said this in some stream and the majority of the time was spent on the game. He also said many times that the game is way more difficult to make than the compiler.

More ›