Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter's Recommendation Algorithm

blog.twitter.com

Post reply on HN
Print view View original

831–840 of 1001 posts

Re: Twitter's Recommendation Algorithm

#831

Post by blitzar »

drstewart wrote: post #687
yurodivuie wrote: post #613

Earlier quoted context omitted.

It's healthy to have a normal amount of cynicism. They released it for a reason. "The goal of our open source endeavor is to provide full transparency to you, our users, about how our systems work." Why be transparent (or try to appear transparent)? To convince people to trust your platform (or to recruit - which seems to be another goal of the post). Why would Twitter want or need to do this now? Well, there is a bi…

[flagged]

I agree, which is why I wonder what your motivation is to defend Twitter. You're posting about this for a reason. If I were a social media company, I'd probably have paid agitators to defend them.

Re: Twitter's Recommendation Algorithm

#833

Post by blitzar »

eyelidlessness wrote: post #796
koolba wrote: post #90

It's reassuring to know that billion dollar tech companies write CI exactly like I do: https://github.com/twitter/the-algorithm/blob/main/ci/ci.sh Permalink: https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92...

Well for such a flawless CI setup I would expect a much longer commit history of “trying it again…” and “descending into madness…” and “might as well summon the eldritch…” and “oh no what have I done, …” and “there this should probably work, nothing unspeakable to see here!” and “oh no not again…”

"Removed test for 100% CI pass"

Re: Twitter's Recommendation Algorithm

#836

Post by fnands »

nelox wrote: post #655
tric wrote: post #11

From https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92... ( "author_is_elon", candidate => candidate .getOrElse(AuthorIdFeature, None).contains(candidate.getOrElse(DDGStatsElonFeature, 0L))), ( "author_is_power_user", candidate => candidate .getOrElse(AuthorIdFeature, None) .exists(candidate.getOrElse(DDGStatsVitsFeature, Set.empty[Long]).contains)), ( "author_is_democrat", candidate => candidate .getOrEls…

It’s April Fools’ Day where I live.

Legit was my first thought when I saw the headline. But apparently it's real

Re: Twitter's Recommendation Algorithm

#837

Post by ivalm »

helsinkiandrew wrote: post #825
jonathanmayer wrote: post #45

Context: I teach at Princeton and study social media and recommendation systems. From a very quick skim of the repositories, this appears to be quite limited transparency. The documentation gives a decent high-level overview of how Tweet recommendation works—no surprises—and the code tracks that roadmap. Those are meaningful positive steps. But the underlying policies and models are almost entirely missing (there are…

> But the underlying policies and models are almost entirely missing... Without those, we can't evaluate the behavior and possible effects of "the algorithm And neither can spammers find and test the cracks and edge cases that would allow them to break the system, that does sound reasonable to me. If they were public there would be an arms race between spammers/those wishing to game the system and Twitter engineers.

Then don’t pretend to release “the algorithm.”

Re: Twitter's Recommendation Algorithm

#838

Post by that_guy_iain »

evolve2k wrote: post #639
sillysaurusx wrote: post #450

Earlier quoted context omitted.

Elon seems embarrassed: https://twitter.com/elonmusk/status/1641908130274525187?s=61... It’ll be interesting to see what gets cut. Maybe just the Elon flag, but maybe others too.

Elons reply “I only learned about it now!”. What a crock of shit. We literally went through reporting a few months back where he was clearly instructing the team to make sure his tweets always come up, for everyone.

If I remember correctly, he noticed that his account with 100,000,000 plus followers was only getting 40,000 or so views. This lead to a Twitter engineer saying maybe his content wasn't interesting enough to get views and getting fired. And later on Musk said that they found a bug in the fanout system because his account was so big it basically broke things and they fixed it.

I remember everyone agreeing with the Twitter engineer who said maybe his content wasn't interesting enough, while to me that seemed odd. If people follow an account and they tweet and they're online they should have a high chance of seeing that tweet. That's the entire point of following someone. If someone I've followed tweets something I would like to see it.

Re: Twitter's Recommendation Algorithm

#840

Post by mort96 »

kaba0 wrote: post #815
mort96 wrote: post #788

Earlier quoted context omitted.

I wouldn't want to use a build system written in Java for non-Java code. Adding the whole JVM as a dependency just for the build system isn't worth it,

Surely that 5 megabytes will break every computer out there.. While we are at it, why not just strip out libc as well? It’s just bloat, right?

It's not a disk space thing.
Post reply on HN