News for hackers. Stuff that matters.
Re:Terence Tao on GPT-4 by endisneigh on Wednesday April 12, 2023 @09:39AM (#35539469)
it's plain text. it can be cached. it's embarrassing.
Re:Terence Tao on GPT-4 by blowski on Wednesday April 12, 2023 @09:46AM (#35539571)
Perhaps they've optimised for the normal levels of traffic they receive, not unanticpated spikes. That's hardly an embarrassing choice.
Re:Terence Tao on GPT-4 by endisneigh on Wednesday April 12, 2023 @10:03AM (#35539827)
I'm not talking about the administrator of this instance, I'm talking about how mastodon is fundamentally designed. Even an iPhone could probably serve the (static) text of Tao's post to a million people on a 5G connection. Computers are fast.
Re:Terence Tao on GPT-4 by blowski on Wednesday April 12, 2023 @10:10AM (#35539947)
I run Mastodon on a $5 VM. If I hit HN homepage, my site will probably be slow as well. I don’t see how that reflects on Mastodon though.
Re:Terence Tao on GPT-4 by endisneigh on Wednesday April 12, 2023 @10:25AM (#35540239)
what am I saying is that there's a way to design a site on a $5 VM that serves static text that wouldn't be slow even if it was #1 on HN.

if you disagree with that then we'll have to agree to disagree.

Re:Terence Tao on GPT-4 by shakow on Wednesday April 12, 2023 @09:48AM (#35539599)
Feel free to share your know-how https://github.com/mastodon/mastodon
Re:Terence Tao on GPT-4 by endisneigh on Wednesday April 12, 2023 @09:54AM (#35539702)
Mastodon's architecture inherently does not prioritize speed of first page load. Any PR would be rejected :D
Re:Terence Tao on GPT-4 by patates on Wednesday April 12, 2023 @09:50AM (#35539643)
I thought what and when to cache was one of the hardest problems in computer science.
Re:Terence Tao on GPT-4 by endisneigh on Wednesday April 12, 2023 @09:51AM (#35539654)
Something like Mastodon is likely 90% reads, or even more. How often is the OP editing their post? When they post, cache it. Kind of like what I did by including it in my original post. Caching is a hard problem because sometimes invalidation results in serious consequences, therefore knowing when to invalidate becomes potentially intractable. In this case there would be few, if any.
Re:Terence Tao on GPT-4 by jibe on Wednesday April 12, 2023 @09:57AM (#35539749)
it was cache invalidation, and naming things. Just throwing stuff in cache is easy.
Re:Terence Tao on GPT-4 by CyberDildonics on Wednesday April 12, 2023 @10:52AM (#35540734)
Why would it be difficult to cache static text?