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

Complaints mount after GitHub launches new algorithmic feed

theregister.com

Post reply on HN
Print view View original

101–110 of 155 posts

Re: Complaints mount after GitHub launches new algorithmic feed

#101

Post by lamontcg »

epicide wrote: post #31
MichaelBurge wrote: post #12

Earlier quoted context omitted.

What exactly does "optimizing for Github recommendations" mean? I'm not writing and debugging 10,000 lines of C just to farm Github stars. It seems crazy to think anyone would change how they work for that. Whereas, people on Youtube/Twitter/Facebook/etc. really do change their videos or writing to spread better.

> I'm not writing and debugging 10,000 lines of C just to farm Github stars. And neither would anybody else. That's the problem. (i.e. they'll do something else besides write and debug code to farm stars) Let's say I determine (or at least believe ) that the GitHub algorithm prefers projects with a README with lots of images and emoji, MIT-licensed, and lots of forks. Obviously, there are already some really great pr…

Create 100 different npm projects that should be one bit of functionality but ship every single method as a different github repo and update all of them continuously with new content in their README.md and such in order to suggest to the AI that you're super busy in maintenance. Then abandon all of them once you land the FAANG job you're looking for.

Re: Complaints mount after GitHub launches new algorithmic feed

#102

Post by dgritsko »

eloisius wrote: post #6

I look forward to the quasi-spam slurry of infiniscroll feed crap once growth hackers start trying to game the algorithm to get more traction for their project. I don’t think you can introduce a recommendation algorithm without it having a negative effect on the content it’s supposed to aid discovery of. Pre-recsys, the content is made for human consumption, but once you add the recsys, the AI itself becomes part of…

> I don’t think you can introduce a recommendation algorithm without it having a negative effect on the content it’s supposed to aid discovery of.

This feels like it ought to be a corollary to Goodhart's Law ("When a measure becomes a target, it ceases to be a good measure"), or perhaps a specialized application thereof.

https://en.wikipedia.org/wiki/Goodhart%27s_law

Re: Complaints mount after GitHub launches new algorithmic feed

#103

Post by mostlysimilar »

ticviking wrote: post #14
eloisius wrote: post #6

I look forward to the quasi-spam slurry of infiniscroll feed crap once growth hackers start trying to game the algorithm to get more traction for their project. I don’t think you can introduce a recommendation algorithm without it having a negative effect on the content it’s supposed to aid discovery of. Pre-recsys, the content is made for human consumption, but once you add the recsys, the AI itself becomes part of…

I’m convinced the ai recommendation audience is the origin of the “death of the internet”

I agree. How do we fight back?

Re: Complaints mount after GitHub launches new algorithmic feed

#104

Post by hedora »

VWWHFSfQ wrote: post #37
eloisius wrote: post #6

I look forward to the quasi-spam slurry of infiniscroll feed crap once growth hackers start trying to game the algorithm to get more traction for their project. I don’t think you can introduce a recommendation algorithm without it having a negative effect on the content it’s supposed to aid discovery of. Pre-recsys, the content is made for human consumption, but once you add the recsys, the AI itself becomes part of…

I don't think it will be targeted at growth of a project. It will be optimizing for recruiter spam in the feed.

No, it's not like they're LinkedIn or something!

Oh wait...

Re: Complaints mount after GitHub launches new algorithmic feed

#105

Post by MrStonedOne »

Destiner wrote: post #91
u2077 wrote: post #85

Nobody asked GitHub to become a social media platform. Why not improve search instead? When I want to find something, I will look for it, and know what I’m looking for. There’s no point in pushing a bunch of random repos in my face with an algorithm that can and will be abused.

Their latest code search thing [1] is pretty powerful already imo. [1] https://cs.github.com

No post body was provided.

Re: Complaints mount after GitHub launches new algorithmic feed

#107

Post by dividuum »

Jaruzel wrote: post #81
smordistan wrote: post #75

Earlier quoted context omitted.

It certainly seems like a watershed moment where the primary purpose of our machinery has shifted to serve the machines instead of the people.

In tens years time, the internet will be purely services and content created by machines, for consumption by other machines.

There is a conspiracy theory named "dead-internet theory" that postulates that this has already happened.

Re: Complaints mount after GitHub launches new algorithmic feed

#108

Post by u2077 »

Destiner wrote: post #91
u2077 wrote: post #85

Nobody asked GitHub to become a social media platform. Why not improve search instead? When I want to find something, I will look for it, and know what I’m looking for. There’s no point in pushing a bunch of random repos in my face with an algorithm that can and will be abused.

Their latest code search thing [1] is pretty powerful already imo. [1] https://cs.github.com

Sweet! Joined the waitlist.

Re: Complaints mount after GitHub launches new algorithmic feed

#109

Post by hedora »

epicide wrote: post #31
MichaelBurge wrote: post #12

Earlier quoted context omitted.

What exactly does "optimizing for Github recommendations" mean? I'm not writing and debugging 10,000 lines of C just to farm Github stars. It seems crazy to think anyone would change how they work for that. Whereas, people on Youtube/Twitter/Facebook/etc. really do change their videos or writing to spread better.

> I'm not writing and debugging 10,000 lines of C just to farm Github stars. And neither would anybody else. That's the problem. (i.e. they'll do something else besides write and debug code to farm stars) Let's say I determine (or at least believe ) that the GitHub algorithm prefers projects with a README with lots of images and emoji, MIT-licensed, and lots of forks. Obviously, there are already some really great pr…

When I was a young child wandering the streets of $EXOTIC_LOCATION, I came across a street stall full of $OBSOLETE_COMPUTER.

I've searched far and wide for a $TRIVIAL_ALGORITHM_IMPLEMENTATION that brings back the nostalgia of that day.

Before running this code, it is important to have a proper computer to do it full justice. I recommend the $AMAZON_AFFILIATE_LINK

... 20 more screens, and 200MiB of JavaScript trackers later:

One line of code that does not compile, copy pasted from stack overflow.

Re: Complaints mount after GitHub launches new algorithmic feed

#110

Post by hn_throwaway_99 »

eloisius wrote: post #6

I look forward to the quasi-spam slurry of infiniscroll feed crap once growth hackers start trying to game the algorithm to get more traction for their project. I don’t think you can introduce a recommendation algorithm without it having a negative effect on the content it’s supposed to aid discovery of. Pre-recsys, the content is made for human consumption, but once you add the recsys, the AI itself becomes part of…

> I hate it because GitHub was one product I really loved, but since MS bought them it seems like they are doing all this random stuff to solve problems that aren’t even a problem. How about getting back to the basics, like reliable uptime, and less glitchy JS soup?

I'm not a fan of this recommendation engine either, but I also don't think that's a fair assessment of what has happened since MS bought them. I've seen a ton of valuable features added since they acquired GitHub, primarily things like GitHub Actions and GitHub Packages.

That said, I agree with your last sentence. Even before this week's nasty database issues that have caused outages every day, GitHub was struggling more and more with reliability.

I know in a large company that it doesn't work to just "throw more bodies" at reliability/quality issues, but GitHub must do some org retooling to get their reliability problems under control. Despite being a big fan of GitHub features, I can't have a critical service go down with the frequency that it does.

Post reply on HN