protomikron

Thx, that often works, but that's still inaccessible to me.

> Not a X. Not a Y. Not a Z. The thing he was most proud [...]"

This construction - how is it called? It is clear "AI speak" - at least I remember Claude talking like that (even if it is code).

If you think about it, it makes sense.

Say what you want about LLM-assisted software development, the chances are high that it will stay, meaning a non-trivial part of code will be written by an LLM.

So is it better to have

  - git commit (mostly only code)
  - magic or blackblock (meaning back and forth by developer with LLM, before commit)
  - git commit (mostly only code)
  ...
  (rinse and repeat)

or

  - git commit (prompt and/or code)
  - git commit (prompt and/or code)
  ...
  (rinse and repeat)

Obviously for that to work the LLM output has to be deterministic and a commit chain has to be pinpointed to a specific weight blob.

You can use ctrl+shift+t to open the recently closed tab again.

Ah I see, thx, it's a pretty default configured "gnome-terminal", which probably captures the F10.

It's an interesting file browser, the default "blue" colors gives old-school vibes nowadays.

If you wonder how to quit (if started from a terminal):

It's `ESC 0`. Or "exit" like from a shell. Took me some time I have to admit (q, ctrl-c, ctrl-q, F10, ESC all did not work).

Valid question i think in this context. I knew about GNU multiprecision library, but thought that couldnt be it, as it's "just" a highly optimized low level bit fiddling lib (at least thats my expectation without looking into the source), so it's strange why it could be damaging Hardware ...

It's okayish. Considering 64G to 128G are available for (nerd) high-end consumers you're just off with a factor 5 (if we can squeeze out a little bit more performance).

Thas is pretty astonishing in my opinion.

I think there is some kind of misunderstanding. Sure, if you get somehow structured, machine-generated PDFs parsing them might be feasible.

But what about the "scanned" document part? How do you handle that? Your PDF rendering engine probably just says: image at pos x,y with size height,width.

So as parent says you have to OCR/AI that photo anyway and it seems that's also a feasible approach for "real" pdfs.

Do you think as an observer of Roko's basilisk ... should I share these prompt or not?

This guy footnotes.

They don't.

Although this is nice, the problems with the GIL are often blown out of proportion: people stating that you couldn't do efficient (compute-bounded) multi-processing, which was never the case as the `multiprocessing` module works just fine.

Why is it always public cloud (aws, gcp, azure) vs. "bring your own hardware and deploy it in racks".

There are multiple providers that offer VPS and ingress/egress for a fraction of the cost of public clouds and they mostly have good uptime.

There's also Lektor CMS: https://www.getlektor.com/ It's quite mice solution to the whole problem (simple hosting + simple editing).

What about CC licenses for model weights? It's common for files ("images", "video", "audio", ...) So maybe appropriate.

> That project is a bash script to tail Kubernetes logs from multiple pods at the same time.

Which is also cool! You're obviously both inspired by Unix `tail`. I wonder if we will get a POSIX/Linux layer on top of Kubernetes ... where pods/services somehow map to processes/systemd etc. ;)

An annoying thing is if CLI tools don't exit with non-zero if the tool is used incorrectly.

It's pretty standard to print the usage string if the tool is invoked with invalid arguments. However the program should then also exit with non-zero, otherwise it can be inconvenient to use in scripting.

> They use the printf "%f" float format conversion so `printf "%f" 2e3` will work just fine, whereas using something that definitively is not a float makes printf exit with an error code as it fails to format the value as float, failing this check.

You are of course correct (sorry to sound like a language model), I missed the printf.

Some nice stuff, but also some stuff, where things get easier if you just switch to a language with less caveats (like Python).

My rule of thumb is to use Bash to write simple scripts, if they fit on the command line and have just one level of loop or if-else (lines can get very long, though ...). However for more complicated stuff I use the alternative programs (find, sed, awk, ...), as they behave more predictable.

Furthermore I am not sure one should use such simplified versions, that are just wrong:

    is_float() {
        # Usage: is_float "number"

        # The test checks to see that the input contains
        # a '.'. This filters out whole numbers.
        [ -z "${1##*.*}" ] &&
            printf %f "$1" >/dev/null 2>&1
    }

I mean having a '.' in the string does make it a non-integer, but what about other floats like 2e3?

It has to be open-source (meaning (A)GPL, MIT, BSD, Apache, etc.) to be even considered by bigger organizations (especially in Europe). I worked with medium and large organizations and if you're fully open-source you can even get funding [to help distributing open source infrastructure software], but if you do the "shady open-core" model I can't use it at least where I work.

So, how's the state of Elm?

I am familiar with all the social media drama on Elm - are serious things missing, or is it possible to write competitive software (you may define that as you wish) and using the web elements FFI, if things are missing?

You probably misclicked the comment section, I guess you wanted to go here: https://news.ycombinator.com/item?id=33346091

I wouldn't say Scotland's climate is representative. European countries can have vastly different local weather conditions and just because there's so much rain in the north western part (Great Britain, Netherlands, ...) it's mostly because of their more oceanic climate, which isn't comparable to e.g. central europe (Germany, Austria, ...).

Yes, I think so too.

> Recently this has been my thought... that my future projects should be AGPL from day one.

If that means that companies can't use it, because that would mean, if they use it they would be required to open-source their managed SaaS solution ... that's a feature?

> Sites like these are a perfect response to many different sides of the web design debate – including those who preach that plain HTML with six tags and no CSS/JS is the only acceptable answer for a functional website.

For what it's worth, it breaks Dark Reader.

Oh, no I get it. It's true, the original stable diffusion repo does have filters, however there are many versions that you can run locally, usually with the filters turned off.

What do you mean, without filters? Stable diffusion does not have any filter as far as I know.

That's a cool story.

Ardour is awesome and shows that specialized niche software can compete or even out-compete commercial proprietary software (like QGIS in the GIS space).

I hope it will be around for a long time and it's a nice piece of software, thx Paul.

On-topic: It's true, basic design often correlates with better content.

Off-topic: The author is one of the orignal authors of Gforth, one of the main Forth implementations.