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

711–720 of 1001 posts

Re: Twitter's Recommendation Algorithm

#711

Post by jillesvangurp »

The irony is that I prefer Mastodon's sort by time and don't try to be clever approach to this expensive and futile attempt to feed me an endless stream of click bait. I objectively spend more time on Mastodon than on Twitter at this point. It's more engaging for me. It's how Twitter used to work when it was still nice to use.

If Twitter wants to put a stop to the user exodus and save lots of money in the process, here's what they could do:

1) Add an off switch to the for you feed. I'll click it right away and never turn it on again. Stop wasting minutes of CPU time on my behalf. I never asked for it. It doesn't do anything for me that I need or want.

2) Sort by time, filter by hashtag. Twitter used to be about real time information. I don't care about things that happened days or weeks ago. I don't need to see all of it. This is the core feature that made Twitter popular. Mastodon has it and it is absorbing users from Twitter by the millions. It still works. Restore this feature and make it the default.

3) Join the fediverse. That's where a lot of the former hard core users went. They still exist. They still post messages. They still engage with each other. They just don't use Twitter anymore. Allow people to follow mastodon users. Allow mastodon users to follow Twitter users. Not that hard to implement and probably would do wonders for user engagement.

Re: Twitter's Recommendation Algorithm

#712

Post by underdown »

jefftk wrote: post #563
db48x wrote: post #551

Earlier quoted context omitted.

Do you think the code looked like that prior to Elon's purchase? I suspect that there was another name there before. Separately, which of these groups do you think that they use as a control?

> which of these groups do you think that they use as a control? When you run an A/B test you randomly divide your users into groups, one (treatment) getting the new behavior and one (control) getting the current production behavior. So your question doesn't make much sense?

You’re describing a multivariate test. Two groups under two different algorithm variants. Multivariate tests still require control groups.

Re: Twitter's Recommendation Algorithm

#713

Post by ISL »

jillesvangurp wrote: post #711

The irony is that I prefer Mastodon's sort by time and don't try to be clever approach to this expensive and futile attempt to feed me an endless stream of click bait. I objectively spend more time on Mastodon than on Twitter at this point. It's more engaging for me. It's how Twitter used to work when it was still nice to use. If Twitter wants to put a stop to the user exodus and save lots of money in the process, he…

As near as I can tell, Mastodon doesn't really have #2 in the list above. Last I heard, the social architecture was hostile to comprehensive indexing of the entire fediverse for search.

That's probably one of the biggest reasons that I have remained on Twitter even after setting up a Mastodon persona.

Re: Twitter's Recommendation Algorithm

#714

Post by lenkite »

slg wrote: post #658
panarky wrote: post #512

Earlier quoted context omitted.

The fallacy of false equivalence systematized in code. Now one side can spew as much disinfo and incitement to violence as it likes, and any algorithm change that prevents this shit from getting amplified will be rejected as bias. BSaaS = Both Sides as a Service

This shouldn't really be a surprise to anyone. It was reported years ago that Twitter was unable to cut down on hate speech because the automated systems they developed triggered too many [debatably false] positives on Republican politicians and that was bad for the company's reputation. If Twitter wanted to prevent future code changes from impacting that approach, there needed to be something like this in the code o…

[flagged]

Re: Twitter's Recommendation Algorithm

#715

Post by jasonhansel »

CathalMullan wrote: post #55
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…

Only used for metrics, apparently. [0] /** * These author ID lists are used purely for metrics collection. We track how often we are * serving Tweets from these authors and how often their tweets are being impressed by users. * This helps us validate in our A/B experimentation platform that we do not ship changes * that negatively impacts one group over others. */ [0]: https://github.com/twitter/the-algorithm/blob/7f…

Interesting which "groups" they care about (e.g. mainstream political parties).

Re: Twitter's Recommendation Algorithm

#716

Post by lenkite »

panarky wrote: post #512
roughly wrote: post #153

Earlier quoted context omitted.

I expect they're tracking the red team/blue team metrics because of the political shitstorm that's been the GOP's assertions they're being silenced by The Algorithm.

The fallacy of false equivalence systematized in code. Now one side can spew as much disinfo and incitement to violence as it likes, and any algorithm change that prevents this shit from getting amplified will be rejected as bias. BSaaS = Both Sides as a Service

[flagged]

Re: Twitter's Recommendation Algorithm

#718

Post by throwaway202303 »

threeseed wrote: post #478
mochomocha wrote: post #192

Earlier quoted context omitted.

... Metrics tracked in AB test. So even if it's not explicitly encoded in the algo (or implicitly through some of the features plugged in), they'll pick the winning cell as long as it doesn't hurt Elon's metrics (I'm just parroting the comment you quoted). It doesn't have to be in the algorithm for the systems to be tweaked to please Elon vanity metrics. [I've been running lots of ML AB tests over the years, some in…

That lines up with reporting from Casey Newton a few days ago where a handful of VIPs e.g. Musk, LeBron James, AOC were being used as weather vanes to understand what the algorithm was doing. It definitely isn't just metrics. Any algorithm change that negatively affected Musk was clearly not going live.

[flagged]

Re: Twitter's Recommendation Algorithm

#719

Post by mardifoufs »

fanagra32 wrote: post #530
corbulo wrote: post #201

It's disappointing the comments are so obsessed with the political angle to this that there's a total lack of appreciation (or discussion) of opening up the most influential social media platform in the world.

"Opening up"? You must be kidding. Nothing is open there. It's just open-washing. A few nice diagrams, but how the services _actually_ work is still hidden.

If you ignore the hundreds of thousands of LoC... then yeah I guess it's just diagrams? Are you sure you actually looked at the main code repo?

Re: Twitter's Recommendation Algorithm

#720

Post by J_tt »

ISL wrote: post #713
jillesvangurp wrote: post #711

The irony is that I prefer Mastodon's sort by time and don't try to be clever approach to this expensive and futile attempt to feed me an endless stream of click bait. I objectively spend more time on Mastodon than on Twitter at this point. It's more engaging for me. It's how Twitter used to work when it was still nice to use. If Twitter wants to put a stop to the user exodus and save lots of money in the process, he…

As near as I can tell, Mastodon doesn't really have #2 in the list above. Last I heard, the social architecture was hostile to comprehensive indexing of the entire fediverse for search. That's probably one of the biggest reasons that I have remained on Twitter even after setting up a Mastodon persona.

Mastodon searching is a lot more focused on hashtags rather than the contents of peoples posts. Consider it more "opt-in" discovery.

The existing user base is strongly in favour of a more organic social graph from exploring tags for shared interests.

Browsing tags is a very normal thing to do on the platform.

Post reply on HN