ses1984

Very large models that aren’t large language models.

A script will last forever, and doesn’t cost tokens to run.

That statistic doesn’t mean much.

The question is how long does it take for all the air in the plane to be replaced.

I am sorry. I should not have said that. Your reply was very generous. I was a jerk for no reason. No one should ever act that way here or anywhere.

People may converge on a new node like forgejo, so from that point of view things are not decentralizing, the dependency graph stays roughly the same shape, but it’s a step in the right direction. It’s marginally better than github hosting everything.

That’s what I should have said originally instead of being an asshole.

It's software, which can be infinitely and freely copied, people are going to copy it, and they should, because not everyone should write their own service from scratch (on top of an scm they wrote from scratch (on top of a language and operating system they wrote from scratch too)).

If only you bothered to read the first line of the article, directly under the title:

>I moved my code from GitHub to a self-hosted Forgejo

Why would a gymnastics gym get robbed? It’s just a bunch of smelly equipment that’s hard to sell and probably very little cash.

Pushing buggy software that could result in some expected nonzero number of incidents per year can be done as an intentional tradeoff, any time the cost of incidents is lower than the opportunity cost of moving fast.

Dare I say that most software engineers literally plan to hit Murphy's Law?

If you build websites, and you never get hit by Murphy's Law, it could mean you are being too conservative.

If you build bridges, your job is to make sure you never get hit by Murphy's Law.

Engineering controls basically mean making it impossible to do something in a way that results in catastrophe.

You can get an over-the-cup pour over for ~$10-20, basic blade grinder for the same, and a pack of filters. That’s all you really need.

You may be completely right, but these examples are pretty meaningless without context, like what is the rate of lose/loose confusion per x words over time.

I haven’t noticed the same trend.

I’m guessing most cases of loose/lose switch happen when English isn’t someone’s first language.

And yet with all these techniques easily available to them, modern filmmakers still churn out a lot of trash.

It’s a matter of taste and style, not technique or quality of gear.

But I don’t agree that old==good and new==bad. There was plenty of trash back in the day, and plenty of great stuff now.

Breaking bad had fantastic cinematography and it didn’t really use advanced techniques like dolly zoom, crane, drone, etc. It had buckets of style though. Compare it to one of its contemporaries, Dexter, which had completely unremarkable, boring, functional cinematography.

The single most important advancement for modern cinematography is to be able to instantly see what your shot is going to look like without having to develop film. This allowed filmmakers to shoot at night, use natural or diegetic lighting, etc. it used to be risky and require a lot of testing to do anything other than bathe your scene in bright light and then use even brighter stage lights for highlights, to give cameras enough light, and to ensure that performances and stunts are visible in the final product. This is way way more impactful than drones and dollies.

No one wants their whereabouts subject to constant surveillance, except everyone who carries a “normal” cell phone, in other words not a burner.

Maybe some of us are frustrated that the political party of your ideology has gone far beyond civil discourse.

How about conservatives have some civil discourse with the nazis in your political parties.

I strongly appreciate being able to see the thread without logging in (or ever visiting that site).

Neither teachers nor administrators take continuing education seriously, in my local experience.

I asked copilot how developers would react if AI agents put ads in their PRs.

>Developers would react extremely negatively. This would be seen as 1. A massive breach of trust. 2. Unprofessional and disruptive. 3. A security/integrity concern. 4. Career-ending for the product. The backlash would likely be swift and severe.

Sometimes AI can be right.

How was Cambridge analytica debunked?

Why does knowing the odds make it not gambling?

The odds are known in a game of roulette, is that not gambling?

You can calculate the odds in games like blackjack to attempt to gain an edge, is that not gambling?

What constraints and how do you enforce them?

No actually it was feb 28th

https://x.com/i/status/2027651077865157033

And yet, some people find themselves compelled to continue gambling long after they’re drowning in debt.

If you don’t want to call that addiction, fine, but you can’t deny that it happens.

Donald trump addressed the Iranian people in a video message and told them to rise up when the war began.

There is actually an em dash in there,

it isn't a disaster—it's the violent…

Is your consciousness now the same as the version of you that wakes up tomorrow?

You didn’t but in the context of the comments immediately above yours,

>Eg Nazi Germany still benefited from a high-ish level of social trust, despite numerous atrocities.

>What's the benefit of social trust when a society can commit such atrocities?

Then you come in with

> It's not like atrocities started with Nazis

Your comment pretty much directly says the nazis weren’t that bad and in context looks like you’re agreeing with previous poster they are high trust.

The original context of this thread is the validity of china’s data in this trust survey. Interjecting with positive example of nazi germany is not even correct. There’s no way anyone can argue in good faith that nazi germany was high trust.

The nazis don’t need you to come to their defense.

Nazism is basically ignoring the will of their own people and actively working against them.