Cost is a proxy for productivity and resources. So while it is stupid to say that the energy transition is too expensive, shouldn’t we rather invest our productivity and resources into a faster and cheaper solution? Drawing focus away from renewables is dangerous as others have mentioned, because it is too late to reach our goals with nuclear.

Well according to Raiden from Metal Gear you should know that A Soul can’t be cut. And he is an expert at cutting things!

Does GitHub support range requests for artifacts? With that you could read the file index without downloading and caching the entire zip

So - I don’t think Firefox would be generating captions for PDFs on PDF creation.

I’m not sure. The blog post is not entirely clear on that.

Between the document creator and the creation software or screen reader, only the document creator would really know the context for the image.

Agreed. Context is usually very important for images. But with an auto-generated caption embedded in the document itself, you already lose some context. Because if the automatic caption is incorrectly stored as “The ___ video game installer” you cannot decide anymore if this was written by the author with the context in mind or just generated. Which I would argue is worse than no caption, as it lowers your trust in all captions.

But I wouldn’t be opposed to ML on those that can auto-suggest things or even critique how content authors write their descriptions.

Absolutely, I think that will be by far the best solution. It could massively encourage users to write their own captions if in most cases you only need to accept the suggestion. But so far, that seems unlikely to be the way forward. Why do that when you can just throw even more “AI” at the problem?

I do agree with your point about auto-generated captions being better than no captions. But isn’t it bad to insert them automatically on creation? If we use these models to caption images shouldn’t it be done by the screen reader instead? That way people can benefit from future advancements of the tech and customize the captioning system for themselves. With the current system there is no way to tell if you got a crappy AI caption that you may want to replace with a better auto-generated caption or a human written caption.

I was also confused about this. I didn’t find an article talking about yet either. My current assumption is that it is just this game and that it is still a closed pre-release. So they maybe just wanted to remove noise from people that don’t have the required specs during their testing phase. It’s also not a lock to a specific GPU vendor as I originally expected from the title.

Maybe an unpopular opinion but why would you care about how privacy invasive GitHub is? Your code is open-source anyways so MSFT can steal it wherever you host it. And if they haven’t changed it you’re able to sign up with just an email and a pseudonym. It’s not a social network where you have to post private information for it to be useful you can and most people do use it pretty anonymously.

So I never understand the outrage about GitHub and MSFT. Git is distributed anyway, the only thing that can be lost are issues and pull request histories. If they fuck up, everyone can just move. Now GitHub Actions, that is a clever thing for binding users…

Just wanted to say that I enjoy your edits. I had never seen the comic before but predicting how Heathcliff fits into it is a fun game, before I look at the original. Do you have them collected somewhere? The old posts aren’t accessible anymore I guess