aeturnum

As someone who only ever asked or answered a couple of questions, Stack Overflow has been on the way out for close to a decade by now. The website never (from my inattentive perspective) did anything about the key problems of a question site:

- If something was marked as a dupe by mods, that was it. There was no way to suggest that a dupe was marked in error on your question or others.

- The site had no standard mechanisms for separating questions in time or by version. So many answers for "python 3" that answered for, like, 3.8 or whatever. Same for operating system and basically any other contextual category.

- Following answers was incredibly inelegant. Would you like to know if the chosen answer switches? What about if there is a new answer for the next version of the software? Trick question we don't track software versions!

- The community had an extremely high learning curve, I think in part because the site had so few affordances. It was very easy to be rude on accident and it left a bad taste in ones mouth.

- The site decided to entirely ignore meta organization. Do you want a category of questions that are kinda silly? What about separating highly complex questions or debugging particular errors from general questions about how to use the tools? Nope, we're not going to do any of that.

I am shocked it lasted as long as it did.

I mean, these things are multimodal. For most people it's easier to cut enough calories to lose weight than it is to exercise enough to lose weight (though both are physically possible). But, of course, if you are losing weight you should be exercising! It's the best way to retain muscle mass, which is vital for health.

I am sure that when I've lost weight it's more the portion control than the exercise - but the exercise is a critical component. When you go from being sedentary to doing even casual exercise you feel better and get more energy - which you can then use to be more active. Without some kind of virtuous lifestyle-change cycle you are likely not going to shift your body mass, kind of no matter what ultimately puts you over the edge.

I agree the client is pretty scuffed, but I don't understand your problems:

> I have frikkin 2 different topbars that seemingly have similar options.

I am not sure exactly what you mean by "top bars" but my two top sets of menus are "Steam View Friends Help" and "STORE LIBRARY COMMUNITY <USERNAME>" which seem quite different.

> The fact that the store page often is a better look at the game than the one in the library.

You can go to the store page from the library! Why would you assume you should re-explain the game to a player? I think reminding you of when you did things in the game makes a lot of sense. I wish it listed playtime and not just achievements.

> Impossible to see what dlcs that is enabled unless you happen to borrow the game from someone in your family.

I do think their DLC display is bad, but it's far from impossible. Properties -> DLC shows you all your DLC and which is enabled. Also clicking on the DLC tab does show you which you own (and all the ones you own will be enabled by default).

Like I said, I think steam has a lot of problems, but I don't understand your complaints at all. I would complain about the awful performance of a webview, the annoying behavior of store pages, the foolishness of the "All" tab on communities that mixes different kinds of content in one place but it will never change because people can't find it otherwise, etc.

It's "sort of" a dark forest problem because admitting you're using AI is socially risky. Indeed, this entire thread is built on the idea that Epic's lazy use of AI has been bested by the savvy & righteous programmers at Valve[1].

So people who think AI is actually a really useful tool and that using it to make your project better will imagine that everyone who makes things better today uses AI at least a little, and assumes they don't talk about it because they don't want to get yelled at. Meanwhile, if you think AI does bad work (or is often used badly) you can ascribe bad software to bad uses of the tool. We can never know for sure because people don't talk about it.

[1] I happen to agree with this dichotomy but that belief precedes AI

Yah - people get fooled by the drop in the relative percentage of the population when the absolute numbers are not actually changing that much. What is real is that the structure of overall society does change - food production becomes less of a concern and people shift their focus to producing other things - but it's wrong to imagine that food production is less important. It just becomes relatively well optimized and less economically productive (i.e. lower ROI for capital). You still need the food just as much.

I am pretty sure their first axiom is "this is only interesting to people who know TA already" and, as you suggest, that's probably correct. There's no reason to annoy your primary audience in trying to make your hobby project accessible to people who are not interested in it.

> Some can be simplified too if this can be done without loss of fidelity.

If you don't understand why the paper is in its current form, how do you know what changes would damage the fidelity for the AI? I think your example of alien knowledge is misleading because it suggests the aliens don't need to understand the info - but in this case we intend to keep using AIs to expand this work. So if you transform it into something that humans think makes more sense without understanding why it's in a form that we hate now - you risk losing something in the form that you don't understand.

I think what I am saying is quite different from that, but you're free to do whatever you like.

I am not representing Anthropic's position I am representing mine. I couldn't care less about their policy. As I said be mean to LLMs if you want, I just don't think it's good for you.

I think you might be interested to look into studies on implicit bias and how being exposed to representations and behavior patterns will get into your subconscious. No matter your intent. Also - it's not like you're going to be hypnotized - I just think it's preferrable to build your ability to write directly and politely.

No, but you are building neural pathways. It's not that it's immoral to be mean to matrix multiplication. It's that getting into the habit of communicating rudely and without considering how your words may impact someone will get you in the habit of communicating that way in general. Ofc it's not "wrong" to talk that way to a machine, but why build the habit?

I can't provide any code samples, but this only strikes me as medium-cursed as far as python code goes. That stuff can get nasty! Once a team realizes they can modify anything at runtime things can go bad quite quickly.

Nice hack for maintaining old behavior tho!