matsemann

That's how the earlier image generation models worked. You had one ai to generate images, and one to understand if it's a fake generated image or a real one (like, you have a real image of a dog, use the ai to generate an image of a dog, and pass both to the ai trying to see if it knows which is which). They both start out bad, but evolve together.

https://en.wikipedia.org/wiki/Generative_adversarial_network

Needing to have all keys present to program a new one yourself is often a security feature. Otherwise someone borrowing your car or getting ahold of one of your keys could possibly make their own.

AI makes it possible to fix bugs and solves issue one wouldn't do before. But just because it's easier now doesn't mean it's automatically worth it. There are tradeoffs (like code complexity or breaking stuff), or it really just wasn't that important to fix if it was in the backlog forever. I've seen devs go down this rabbit hole, and perhaps overdo it. It feels productive, but at some point it might stop to provide value, and other things should've been prioritized. It's just easier to say to the AI "fix bugs" instead of solving real issues. At some point boasting about running 10x agents in parallel to fix bugs becomes a sign that there's a lack of value provided to boast about.

When Norway (unfortunately) privatized and tenderized* the railway, they at least made a provision that every ticket should be purchaseable through a government owned entity. Which means you can use the government app ("Entur") to travel plan and purchase, even between operators and even modalities (trains, busses, trams, etc).

* (using the word wrongly), and it was a hack job. The existing operator is still owned by the government (NSB, renamed to Vy), but then had to bid on contracts to win their existing lines back. Which they of course were predetermined to lose most of, otherwise all that work from the right side government would have been useless.

Of course, NSB then also couldn't have any advantages having been the sole provider from before. So men in suits decided which IT systems they were allowed to keep, and which should be the basis for the Entur app/system.

I came in as a consultant at Vy to reimplement what got severed, without having any prior knowledge/advantage.

So much weirdness, and so much waste

You have multiple comments here in my thread where you wrongly describe how SO works. It's quite meaningless to discuss with you when you lie about points being awarded for stuff it was not etc.

What matters is however the active users answering questions.

Because the people providing the value here, those answering, saw them day after day. That the freeloaders didn't holds no weight.

Most of the badges was for reviewing, a "this question looks fine to me" gave the same progress as "close the question".

Why did it matter if the question you visited was closed as a duplicate? Still you would find your answer by following it. People wrongly think a duplicate was something wrong. They probably could've made it clearer.

Don't hold your breath, people here always claim these wild experiences but never back them up.

Do you have a link?

It's only absurd if you think the question being closed as a duplicate was calling you out for doing something wrong. This was how it intentionally worked: group everything, and provide a canonical answer. Duplicates were good, the idea being that more ways to ask the same question could lead more people to find the answer. The other question just happened to have more people find it (hence more points), but someone finding yours could then be redirected to the more popular variation of the question.

> Also the policing of questions based on geeky criteria and not what users wanted.

This is wrong. People don't get that SO is moderated almost entirely by users. Users are those that down vote. Users are those who vote to close a question, and when a question reaches consensus it's closed. Users are those watching the review queue. Mods mainly did bigger stuff with additional tools.

So users wanted it how it was. And for good reason. If you actually participated on the site, like watching new questions arriving in a constant stream, it was obvious that most of it was low quality. People here will bemoan they got a question closed or no help, but that's to me mostly entitlement. The rules are clear, it was to create a Q&A site, not solve your specific question for free. To me that's not hostile.

They didn't claim they've read ai pharma textbooks, just used a pharma textbook (which I guess is heavy to read) as a comparison to being easier to read than ai written text.

Just skip the powerpoint if you can't be bothered to create it. It's painfully obvious for anyone listening that the presenter didn't come up with or think through what they're talking about.

Agreed. We're still bottlenecked when it comes to feature work because we need to think of the ideas, user test, talk with stakeholders etc. But somehow inventing loads of tasks to feel busy anyways. Those tasks are the side quests, things we normally wouldn't take on. It feels great to finally squash that bug or do that rewrite. But you're adding stuff onto everyone else's plate as well. Discussions, reviews etc, a constant barrage of new things to consider.

Not if you turn off your brain. Never actually stop to question the path you're going down, because the speed is making you blind. Which is what I feel I see more and more. Or how do you even know it's the wrong answer if you didn't think it through before starting?

One example: setting the agent ablaze on some task, keeping it going, and after a few days of working with the agent you have a feature. But it's not what you were supposed to make. The agent didn't have all the context and made something slightly off, and you never internalized the problem enough to even spot it. If you actually stopped to think a bit on how the feature would have to work to tie it in with everything else, like you would have done before, you would've spotted it.

Second example: something wasn't feasible to do before, due to not being worth the time investment. Like a feature that would need a bigger architectural change to implement. You and the agent take a crack at it, and rewrite huge parts of the code, and it seems to work. But now no one understands how, and it might have been the wrong way to solve it, you just went with it. Before, you would've sat down and discussed and thought about the architectural tradeoffs. Now one decision was just made with little thought.

Only because anthropic&co will price themselves just below the cap where people contemplated switching. In so, the Chinese models still set a sort of cap on the prices they can charge, even if not directly.

I think that's an orthogonal issue. A webpage can be crazy with loads of JS stuff, but still print nicely or behave well defined when it comes to semantics, screen readers etc.

At around 10 years old, I remember spending more time on the word art for my titles than the rest of the paper for school projects, heh.

Also remember that I set up some kind of templating, where I could print the same document multiple times, but it would go down a list of values. So I could print personalized birthday invitations, I felt so cool 8)

It's increasingly clear Europe can't rely on the US as an ally.

I see people claiming that others are "closing it wrong". Something about you have to snap it using the edges, absolutely not push into the center?

Until uv, I used docker to run python projects locally. Got so fed up with venvs and versions and slowness and lack of wheels and spending so much time mucking around with it when cloning or updating a repo.

Uv hasn't fixed all issues, but made it far less painful.

Perhaps if you know how. Watching my dad use his iPhone.. Every time have to teach him something new, and hope he remembers it.

So stupid interface. No discoverability.

What a shitty way to dismiss anyone with any concerns whatsoever.

Can you explain a bit what it is one is buying? A prompt? I don't get why it can't be the component, isn't it better to use that than a prompt?

By random chance you probably wouldn't get that many zeros. Of course, one could say it's just as random as any other string, but..

I think your definition is off. Js was never for "code". The code itself is just a means to a goal. That certain things better can be achieved without js is good.

> This also helps the model maintain a more consistent playing style. Sometimes there’s a sacrifice available on the board, through many turns. A human would usually either decide he wants to do it or not, and stick with that decision until something concrete changes. But when I played this bot before adding recency, the bot would be ignoring the tactic for many moves then after an unrelated quiet move, go for it. It felt very alien and off-putting.

From the position history in the previous post. Good point. Same with bishops "sniping". If a bishop is moved to b2, and remains there or 20 moves, it can be "forgotten" and opponent suddenly has a pin or something underprotected. But if I saw the exact same board, but the previous move was moving the bishop to b2, I would be aware of it. That kind of local thinking is of course something I should learn to not do, but at my level (1700-1800 FIDE) it's how humans play.

Well, it's also impossible to link to the official one (here https://www.jwz.org/xscreensaver/ )

without seeing eggballs in a cup due to JWZ's ego. So I say there's no winners with two big egos like these.