Hacker Newsnew | past | comments | ask | show | jobs | submit | geon's commentslogin

The "below average" engineers are largely the juniors and the programming non-programmers.

The juniors can't be replaced because all senior engineer were once junior.

The non-programmers won't be replaced because they are not really programmers to begin with, so there is nothing to replace.


No. Natural language is vague, ambiguous and indirect.

Watch these poor children struggle with writing instructions for making a sandwich:

https://youtu.be/FN2RM-CHkuI


> The code that LLMs make is much worse than what I can write: almost certainly; but the same could be said about your assembler

Has this been true since the 90s?

I pretty much only hear people saying modern compilers are unbeatable.


repl.wasm is 1.6 MiB

Seems a bit much. How large would the todo example be?


This is not the smallest REPL binary out there but it's far from the largest I've seen. My anecdotal evidence suggests it's somewhere in the middle or maybe smaller than average. We haven't configured our web server to gzip these assets (we should). Gzipped repl.wasm is 339K. This binary hasn't been run through wasm-opt, either.

Hoot is primarily an ahead-of-time compiler, not an interpreter, so this REPL demo has a lot of stuff that a production binary wouldn't have like a macro expander, runtime module system, and an interpreter. A significant chunk of extra code. We have a todo list example [0] in a separate repo, btw. On my machine, todo.wasm is 566K uncompressed, 143K gzipped. And that includes a virtual DOM diffing algorithm implemented in Scheme.

That said, there is quite some room to optimize binary size. For example, we know we're generating too many local variables in functions [1], many of which could be optimized away with an additional compilation pass. We expect that to have a significant impact on binary size. There's also certain workarounds we're doing due to features missing from Wasm at the time of implementation that add to total binary size. Lack of native stack switching for continuations is one example. Adopting the stack switching proposal should shave off some bytes.

[0] https://codeberg.org/spritely/hoot-ffi-demo/

[1] https://codeberg.org/spritely/hoot/issues/193



Your reasoning makes no sense.

It is the output that could be copyrighted, not potential future output. Otherwise anyone could allready claim copyright on everything.


Obviously, I wasn’t serious :)


And since the developers who have to do the last 90 % were not involved in the first 90 %, they will have no clue how to do it.


So now the last 90% is the last 99%.


I moved because Ghostty feels just like the native terminal but allows me to set the color scheme. I have it set to match the vscode Monokai theme.

No, macos Terminal will not let you use whatever colors you like. It will helpfully adjust the colors you select to increase contrast. And it can't be disabled. It bugged me for years.


Reads more like it makes fun of trans people to me.


Regular compliance would be be to stop tracking users.

A ton of websites don’t even track users but have the cookie popup because they think that’s what you’re supposed to do.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: