Ask HN: Should social network algorithms be publicly available? | Hacker News Reader

Ask HN: Should social network algorithms be publicly available?

36 pointsnnurmanov27 comments

Regarding Elon Musk's latest twits on creating a new social media platform as he claims Twitter is “failing to adhere to free speech principle”, IMO, we may not to create a new social network, we need to force the existing social networks to make their algorithms publicly available. And we should have some way to influence how algorithms work. I don't know how but voting comes to my mind.

View on HN

Recommendation providers should be decoupled from content providers. Not unlike movie studios and movie theaters. Open-sourcing current algos won't help, they are flawed, but their flaw isn't really in the lack of openness. It is in the lack of competition. If you make the best recommendation algorithm in the world that is miles ahead of everything in existence, you wouldn't be able to make a product out of it, because you need the whole package in order to compete, so your only options are either to try to caught up on features to companies that have vast profitable businesses or sell your creation to said companies. And, by the way, voting on a changes to a single algorithm that makes choices for everyone won't help since a lot of people are going to be pissed of by any decision that is made, which, in turn, means that everyone is going to be pissed off by that algorithm since virtually everyone is going to see a decision they don't like being passed(and so we are back to square one).

Not just publicly available. They should be reproducible.

I image a scenario of a journalist who wanted to investigate whether some post on a social network's wall was pushed by a group of certain interests or maybe analyze the trends and understand who's behind them. There should be a way to receive a set of inputs and a reproducible way to get the same output.

It could start with a user interface where a user could click on a post and they would see the full context of what has led to this post being shown to them at a given moment.

I think it would help eliminate the current trend of shoving whatever machine learning models deem the most addictive and would make the web more transparent and user friendly because then the developers would have to optimize differently.

So to summarize, I think the algorithms AND datasets should be public. In addition to that, even if they were not public, their inputs and outputs should be reproducible and accessible.

If they publish dataset and code, but all decisions are made by trillion parameter NN, which costs millions to train, is it really reproducible?

The point is that the social networks should be mandated to make them reproducible. If they can't do it with their ML models (and they can't), then they should come up with some other way. As another person in this thread suggested, they could just display chronologically ordered posts by the accounts users follow or maybe some other useful way.

As I said, that would make social networks more transparent and user friendly, just by eliminating the black box ML models.

The thing is that it will make situation neither more transparent nor more user-friendly. You will get expert-systems that are not only so complicated that no one really understand the whole system, but they will provide even worse results(here goes user-friendliness).

And as I said in the reply to the person that proposed that everything should be in chronological order, there are inherent reasons why it isn't being used that much anymore(it requires discipline, it doesn't suit short sessions).

Problem with social networks is multi-faceted and won't be solved just by mandated openness, reproducibility or direct democracy. Nor it will be by the three of them. And badly-designed regulation is certainly going to make situation even worse.

Give me a choice of algorithm from a list of algorithms. Promote your choice by making it the default on your platform.

Wouldn't most people just stick with the default?

It really depends on how good and distinct competitors are. On one hand most people use default browser(since they are not really different). On the other not that many people use default messenger.

This would be a good idea

This would be great, but it would have it's own set of problems. Like: monetization problem, thing that existing social networks aren't really made for multiple algorithms, the question of who is custodian of the data, problem of how and where to host these algos. They ought to be solved before we have these nice things.

FOSS social network (like Mastodon) with FOSS algorithm for content recommendation would be the ideal place to be: therefore you would have competition on all aspects: clients, algorithms, and all in the open.

Yes. Honestly the “algorithm” should just be your followers’ latest status updates chronologically - but there’s no accidental or hyped up ad clicks in that.

Ads are lifeblood of social networks, if no ads then they have to find some other ways to make money. Taking into account that a few people are ready to pay for using social networks, this may not be a good venture model. Although, I might consider paying for not seeing ads:)

Paying for YouTube premium is easily one of my best expenses. I use FB and IG less and less because they’re increasingly a dumpster fire of annoying ads.

There are reasons why it is not used that much anymore. And "hyped up ad clicks" isn't even the biggest one.

Personally I think it would be interesting if each state/province or whatever would host it's own people's speech. The public ought own an interoperating, internetworked systems of public speech.

I dont think Elon's complaint mirrors yours. I dont think his offence is chiefly targetted at algorithms. The need to sign in, evermore people being banned, the ongoing elaboration of what is allowed not allowed- these are real barriers. Personally algorithmic curation has not denied me access to my friends- it simply mixes in other things.

I'm fine with whatever algorithms social networks want to use. What I'm not fine with is them being forced on me to show me things I didn't subscribe to. I just want to know what my friends are up to.

In my view, social networking plays an enormous part in today's social ills. I'd like to go back to a simpler time when we were in greater control over the media we consumed -- not by forcing the algorithm to change, but just by opting out of it altogether.

You vote with your feet and don’t use social media. There’s no “free speech principle.” Private companies can allow or censor whatever they want to on their own platforms.

Countries can influence their decisions as well. There were cases when some posts were removed per some country officials' request. I have another thing to bring to the table. I heard numerous cases when FB support completely ignored cases. There is even a post on Reddit of a guy, who managed to resolve his case by buying an Oculus gadget. Just recently I heard a case with Upwork. I am trying to understand here, what is this? A corporate policy? A resource shortage? When you buy a car and it is broken, you can go to court and seek resolution there, but when you have a social account with a large user base or an Upwork account that brings you money, where do you go? I am not saying that the social network are going to change something, but at least we deserve some explanation.

What good would an explanation do if it’s bullshit or not actionable? If you don’t use the social media platforms then it doesn’t matter what they do or why. If you choose to use them you are always playing by their rules and they don’t have to explain anything. You are not their customer, you are the product.

Just the algorithm being publicly available won’t help much, unless all the data that is input to the algorithm is also publicly available.

I think social media and recommendation algoritms will get regulated. Since social media algorithms might be built to cause addiction. Same principle as tobaco industry, eventually it will be regulated since it affects health.

I hope EU is cooking something. At least they should regulate support, so people could receive answers to their support issues. Currently it is not.

These are companies with trade secrets and algorithms are trade secrets. You don’t disclose such as that is your competitive advantage.

any algorithm will be gamed so it doesnt make a lot of sense. Why do we have algorithms again? My feed is chronological, which forces me to only include the right amount of people, and the right people. When twitter randomly resets it to algorithmic, it is noticed, and it's always a downgrade

Sure, why not. However, will they be? No.

I will say no. The algorithm isn't the problem at all. There is 1 thing and 1 thing alone that is the problem.

Twitter and others exist entirely because of Section 230. That as a platform they will not be held liable for the content they host.

But there's a caveat. They may only censor content within very specific categories which most people would agree with.

>No provider or user of an interactive computer service shall be held liable on account of-

>(A) any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected; or

That right there is the extent twitter may censor. They cannot go beyond this but they can go less. You are not obligated to remove all lewd things from their website. Twitter clearly does not given how much porn and gore they host. Twitter actively allows death threats. Good faith is super important in this rule.

So lets roll this up into what exactly is the problem in the context of current events. Rachel Levine and mass banning of conservatives.

Rachel Levine was unimpressive achieving nothing of note. She got appointed to a 4 star admiral position solely because of being trans. Then got labelled as woman of the year? Where is the achievement? Just immediately right to 4 star general and woman of the year?

That's not even the controversy. The controversy is that Twitter is banning people for saying "Richard levine spent 54 years of his life as a man" or "men aren't women" not even naming anyone or even linking trans issue. It is a fact that men aren't women. That's why they are separate words.

The conservatives literally wrote what is factual and got banned. They intentionally crafted it this way to show the bad faith.

Now let's roll back to section 230. Twitter is acting in bad faith to restrict specific viewpoints. In so doing creating a protected class of people. Not unlike how twitter actively allows Putin or various other tinpot dictators but doesn't allow trump? It's all bad faith by twitter.

The problem is obvious. Twitter is in clear violation of section 230 but the government refuses to enforce this rule. So Twitter gets all of the liability protections and none of the requirements of free speech.

Now go back to conservatives who just got disenfranchised. Do you think they are going to take the ban and say, you know what maybe I was wrong about trans people. No. They are going to be radicalized more.

Twitter is creating hatred for trans people. As a trans person, twitter is one of the key players in why trans people are hated, and worse the coming consequences for this will not be good for us trans people.

Now imagine twitter got a legal letter saying section 230 good faith will now be enforced. That a single ideological ban like "men aren't women" will be an immediate suspension of their section 230 protection and they will be held as a publisher.

This immediately solves the problem. Twitter doesn't exist without section 230, they would be forced to immediately solve this huge problem.

Reply on news.ycombinator.com