jerjerjer

The joy of nondeterministic systems.

Wish they'd do a Switch 2 edition without a screen. I use mine (Switch 1) 100% docked.

They didn't even start on the dump part yet. Only initial 5% of the shares are trading so far.

If the history of email had gone somewhat differently, the last email you sent could have been rescinded or superseded by a newer version when you accidentally wrote the wrong thing. It could have auto-destructed if not read by midnight.

Immutability is one of the best things about email.

We had replaceable batteries in phones for years. There's no reason battery replacement has to involve 20 steps and require ungluing the screen.

Programmer adjacent data can already be consumed from git repos. Atlassian has PM data.

A cube with an edge of 2.1 meters.

I’d be surprised if crypto scams haven’t already seen their high watermark in terms of actual victims.

NFT was the peak.

Seems to be a strange setup for a marketing photo.

Why wouldn't it be Chromium/V8-based? New browser engine is multiple developer-years of effort.

Do they even have any legal presence in UK to fine?

When I want it to be displayed in the exact same way everywhere.

Also checked the PDF compilation and it is a surprisingly effective way to explain algorithms considering there are no words at all.

include

Interesting, although to me it looks more like a way to split one file into several (which is rather useful).

extends

What's the difference with anchors? Looks the same, except works with include (and doesn't work with any other yaml tool).

dynamic pipeline generation

Which is even harder to reason about compared to anchors, although certainly powerful.

I don't think anchors' primary function is to allow global definitions (of variables or whatever), rather it's more like arbitrary templates/snippets to be reused through the YAML file.

In GitLab, where YAML anchors have been supported for years, I personally find them very useful —it's the only way of "code" reuse, really. In GitLab there's a special edtor just for .gitlab-ci.yml, which shows the original view and the combined read-only view (with all anchors expanded).

I agree that it's hard to point to the specific line of the source code, but it's enough — in case of an error — to output an action name, action property name, and actual property value that caused an error. Based on these three things, a developer can easily find the correct line.

So is there a UI available to browse MCPs or is it API only?

Tried to find it but docs point to GitHub, GitHub points to docs and none seem to point to MCPs listing.

Your car, no. An issue there could hurt other people.

Which is already covered by existing laws. Same as other categories given by dangus.

It's incredibly easy to remember your hand. Just ask any of your friends that play poker -- they can surely remember a bad beat. And it's pretty much impossible for an attacker to guess.

There's an implication here that users would pick a random hand. I'm sure a set containing all flushes, straights, full houses and four of a kind would account for most of the used passwords.

That kind of thing is surprisingly hard to implement.

If response contains prompt text verbatim (or it is below some distance metric) replace the response text.

Not saying it's trivial to implement (and probably it is hard to do in a pure LLM way), but I don't think it's too hard.

More like it's not really a big secret.

I wonder if this "speed up the audio" trick would save you even more.

At this point you'll need to at least check how much running ffmpeg costs. Probably less than $0.01 per hour of audio (approximate savings) but still.

I feel he may have seen value in military technology.

LOTR has fairly pronounced anti-industrial themes.

Bullet points are taboo now.

Is there a benchmark to measure real effective context length?

Sure, gpt-4o has a context window of 128k, but it loses a lot from the beginning/middle.

ads (sponsored sections) in nearly every video.

SponsorBlock for YouTube resolves the issue.

Wonder what Ted Kaczynski would make of LLMs.

I want to see people who used LLMs all their conscious life enter the workforce. It's going to be amazing. I mean soon we'll see people who are essentially raised by LLMs.

Agreed, the takeover would be peaceful and even celebrated.

It took television about 50 years to hit stride and go beyond film, but it got there. Shows like The Wire, The Sopranos, and Mad Men achieved something not replaceable by the movie or the novel.

Can you elaborate on the differences between television and film? Especially considering the examples you cite. I'd agree that live broadcasting is a considerable departure from a film as a medium. Still, the shows you reference are very cinematic - longer, sure, but for me, they are as close to a film experience as possible.

I mean, they also self report here: https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_articles_w...

I want to see people who used LLMs all their conscious life enter the workforce. It's going to be amazing. I mean soon we'll see people who are essentially raised by LLMs.

What it essentially does is a debugging/optimization loop where you change one thing, eval, repeat it again and compare results.

Previously we needed to have a human in the loop to do the change. Of course we have automated hyperparameter tuning (and similar things), but that only works only in a rigidly defined search space.

Will we see LLMs generating new improved LLM architectures, now fully incomprehensible to humans?