lexh

don't think it's that deep mate. just sloppy wording on GP's part.

Well... since o1 (Sept 2024), most models generate a "hidden" thinking trace before the visible answer... surely you have seen this by now: "Wait, that's wrong...", "Ah, now I have the full picture...". The model prunes dead ends when it composes the final answer.

Demanding sources for this is odd. It's literally been a headline feature of every frontier model for two years.

I guess you are "technically correct" that no model can "un-emit" tokens... but I don't think that is what anyone was saying or an interesting point to make.

Edit: see also this recent post, which details another place where revisions can happen, upstream of the reasoning token emission:

https://www.anthropic.com/research/global-workspace

Similar idea here that may be of interest: SQLite DB on static host via HTTP range + WASM.

https://news.ycombinator.com/item?id=27016630

https://www.researchgate.net/publication/334693268_Portia's_...

Consider adding that snazzy gif in the README to the docs landing page. I went straight to the docs and then hunted for a screenshot to no avail.

But... this particular project does have such automation in place? It isn’t hard to find:

https://codeberg.org/ziglang/zig/src/branch/master/.forgejo/...

https://copilot.microsoft.com/shares/RviQKF1om6oxZY2pENL6c

Sample size is 2 now!

Isn’t part of Vercel’s value proposition a robust global CDN in front? Seems quite a bit different than one sweaty VM in Helsinki.

This is a great write up.

I wonder if any of these techniques turn up in whatever the magic sauce is in D2’s TALA layout engine, which is in a league of its own IMO.

https://d2lang.com/examples/tala/

The current SOTA closed model providers are also all rolling out access to their latest models with better pricing (e.g. GPT-5 this week), which seems like a confounding factor unique to this moment in the cycle. An API consumer would need to have a very specific reason to choose GPT-4o over GPT-5, given the latter costs less, benchmarks better and is roughly the same speed.

To be a little more fair... that example is tidily slotted into the EXAMPLES section, under the heading "You can extract images from a video, or create a video from many images".

I don't think most people read the man pages top to bottom. And even if they did, then for as much grief as you're giving ffmpeg, llm has an even larger burden... no man page and the docs weigh in at over 8k lines.

I get the general point that ffmpeg is a powerful, complex tool... but this is a weird fight to pick.

Use the debatably intelligent machines for this sort of question not Google.

It seems “Yud” here is a shorthand for Yudkowsky. Hinted by the capitalization.

Why is this flagged? Seems to fit the submission criterion of "anything that good hackers would find interesting" and "gratifies intellectual curiosity" pretty cleanly... e.g. government technology contracts and data analysis, privacy and surveillance, a major tech company and its business practices, the intersection of technology and public policy, etc.

It is Postgres specific, but I’ve gotten a lot of mileage out of the advice in this article:

https://supabase.com/blog/postgres-audit

I think some of it is shown on this page:

https://docs.dagger.io/features/visualization/

Gee whiz is this person is in for a treat when they discover the joys of OpAMP https://github.com/open-telemetry/opamp-spec/blob/main/speci...

Turtles all the way down.

A bit over my head, but I enjoyed the way the writing brings us along for the ride.

This can’t be the first pass someone has made at something like this, right? There must be literal dozens of SIMD thirsty Gophers around. Would a more common pattern be to use CGO?

What line of work is full of former programmers?

The impression I was left with is that this tool would write things to disk. It would be helpful to know how that data is stored. I wouldn’t want my password manager OCR’d and then sitting in plain text on disk for example.

“pretty scary stuff” indeed!

This would inevitably end up ingesting secrets, right? Like say from my password manager? Or API keys in my terminal?

Lots of ways for this to go sideways even if the data stays local.

What’s the plan there?

All the staff at my local coffee shop ask “for stay or for takeaway?”

They never deviate from this exact script and I find it grating every single time. Coffee is decent though so I let it slide.

Edit: “local” meaning Southern California.

Oh, data collection. And subscriptions.

This makes sense (and myQ’s privacy policy is a nightmare: https://www.myq.com/privacy-notice) but I’ve never understood how this particular bit of data is valuable to anyone. Any ideas?

  > Perhaps it is the article On Sense and Reference by Gottlob Frege?
  >
  > http://www.scu.edu.tw/philos/98class/Peng/05.pdf

You're telling me i provided $14/m in ad revenue? Of course not.

Do you really think Google just set the price arbitrarily? That would be surprising.

My impression has always been that ad free tiers are actually less profitable for video providers.

Fantastic. Love the functionality on public GH and always find myself missing it at work.

Will this make it to GH enterprise eventually?

I was delighted when I discovered that you can also pipe stdin into Wireshark (GUI). I've not seen a lot of GUI applications that support this flow.

e.g.

  > wireshark -k -i <(ssh -l root remote-host "dumpcap -P -w - -f 'not tcp port 22'")  

This just came across my desk this morning, it discusses the benefits of hot water bottles as an alternative means of keeping warm.

https://www.lowtechmagazine.com/2022/01/the-revenge-of-the-h...

(HN discussion from 2022 https://news.ycombinator.com/item?id=30023681)

I assume you would feel differently if the market had crashed and your purchase price was $100k less though… avoiding PMI while also having a lower principal. Hindsight.

I too appreciate the humor. That said, homelessness often means limited or no restroom access. I’ve had someone turn down food on the basis that it would cause them GI distress. What an awful balancing act.