Hacker News new | ask | show | jobs
by hvdfhbj 2045 days ago
It seems to me that social media companies should only be allowed the legal protection of being "platforms" if they are truly neutral. No censorship, no algorithms.

Facebook should only be allowed to show you a feed of all your friends' posts, in chronological order. Twitter should only show you the tweets of people that you've followed, in chronological order.

Aside from solving any potential censorship or manipulation problems, it would also make a much more pleasant user experience than the unnavigable garbage that is Facebook's current product.

10 comments

The more I think about this, the more I find myself coming around to a hybrid of your idea (strict neutrality) and Zuckerberg's original position (moderation of free speech is not their job).

Facebook et al.'s fundamental sin was optimizing the sorting and broadcasting algorithms.

I get it drove growth and made them the giant companies they are today. But it's not really a core competency, and now it's more trouble than it's worth. In fact, it's metastasized into an existential threat to them.

They should announce they're going back to naive ranking and sharing, and that they expect all external moderation requests to come through legal channels. (And then spend their billions digging the widest moat of integrated services and infrastructure they can)

They should announce they're going back to naive ranking and sharing, and that they expect all external moderation requests to come through legal channels. (And then spend their billions digging the widest moat of integrated services and infrastructure they can)

The trouble is that algorithmic feeds work really well. If they remove algorithmic feeds in an environment where competitors are still able to have such feeds, then they will be outcompeted in the long run.

This is one area where legislation and strict enforcement barring algorithmic feeds would work, though execution would be difficult.

Or at least give the user a preference as to the algorithm/sorting method they want (or the ability to implement their own! Will never happen, sadly.) I think it’s the lack of transparency into the curation. and lack of control over it, that is the issue, not the particular queries being run.
Most people have no idea how to change preferences, so whatever is default might as well be the only option. This is why Google pays billions to be the default search engine, not even the forced/only option.

It matters so little that it’s kind of odd that these sites don’t give the option, so at least they could use it in their arguments to show they have it. Probably only 0.01% of users would ever actually use it.

Make a secondary market for algorithms and split part of the revenue with twitter. I'd love that.
So all your twitter data and interactions like clicks and time spent will be sent to whoever gets you to install their algorithm? I can see why Cambridge Analytica seemed like a good idea to some people now...
Almost all of that stuff is public data which can already be scraped.
Section 230, is explicitly no censorship. The problem is that twitter has gotten big enough that you can't seem to punish it by removing its Section 230 protections. As that would essentially be the death penalty for the platform.

Hence why supposedly there's an exodus to Parler.

Section 230 of the Communications Decency Act makes no designation or distinction of “platforms”. You have deeply misunderstood the law.

What it does is to make clear that people or organizations who run internet sites that allow 3rd party submissions will not for the purpose of the law be considered the “speaker” of the content in those submissions. It also states that this hold even if they moderate including editing the content to conform to internal standards.

The law specifically allows censorship, algorithms, banning, and editing or anything users submit. The whole purpose is to clarify that content hosts are allowed to do this without being held liable for the content of the speech they are moderating.

Parent and link speak to should, not must. I'd assume that they start from a position that current laws are unsatisfactory.
The parent seemed to imply that there should be no permissible moderation at all, which strikes me as undesirable. Also the use of “provider” is common to factually incorrect descriptions of the law.

I would also add that no one prevents the original poster, or anyone else from using other services. I mean, here we are on a narrow social media site discussing the topic.

I think moderation has become far too strict on the net and I have seen countless examples of censorship due to bipartisanship. I don't think all people want that. If people prefer it that way, there need to be separate spaces.

But I believe this is at the core of the issue when people complain about moderation. I think it is very true, even sites like Wikipedia are negatively affected on political topics while paid external editors are free to roam to push up articles about themselves or their companies. This is the worst kind of moderation I can imagine.

If we talk about moderation against spammers and scammers, a vast majority would agree on the other hand. But these are two separate issues in my opinion.

> I would also add that no one prevents the original poster, or anyone else from using other services.

I think there is concerted effort to purge alternatives in the current market situation and I think this nothing else than an excuse to justify certain content moderation. I think platforms have this freedom, but there are also consequences to that.

Parent said that presentation of originally-submitted content should be 1:1 and neutrally ranked (e.g. by timestamp).

That seems a fairly modest proposal.

It's a great proposal for a product roadmap, or building a competitor but an awful proposal for legislation. Why should the government dabble in deciding how social media builds its product? You'll just crate a mat that entrenches FB and the like.

I think people VASTLY overestimate the influence of things like the FB feed on the real world. The best evidence they ever seem to have for any claimed effect is the number of people FB say saw/interacted with a post. They have a long history of inflating those numbers, yet the critics are all too happy to use them. The activity on social media that seems to actually result in real world outcomes seems to generally be the boring work of community building.

Also, how do you deal with spam or noisy posters in a "1:1 and neutrally ranked (e.g. by timestamp)" feed?

Who said anything about Section 230? The GP was describing what he thought the law should be, not what it is.
I barely use twitter, I follow like 2 people. But I have still noticed how offensively useless the timeline is. The timeline has never been anything more than an out-of-context jumble of nothing, even with how curated my following list is. It is actually shocking to me how utterly incompetent it is. It takes what I want to see and actually somehow strips it of value. There is no reason for me to use the timeline in any respect ever, but it's still the landing page.
Agreed. Tweetdeck is able to have a normal timeline, but the damage is done, all the people are like us, we've all stopped using it. It's on a downward spiral.
Most people strongly prefer when you don't show them a chronologically ordered feed.

I don't think we should design our legal frameworks around arbitrary user-hostile restrictions.

^ this. Even among people who claim to prefer chonological feeds, double-blind tests show a strong preference for algorithmic feeds.

If grandparent is saying “algorithmic feeds are so popular-yet-dangerous that we should regulate them like alcohol” then ok, that’s a valid starting point - but we should at least acknowledge that this is user-hostile in the short term.

I am so tired of this line of argument. Those are totally arbitrary suggestions without any consideration for what is practical or legally plausible.

On what basis should website owners lose their right to determine what goes up on their own website? What about the creative control one's own work? What about free speech for the site creator? What about the property rights owned to those paying cash to keep the site operational? The law should not attempt to dictate the specific ways in which it is legal for computer programs to sort information for display on a website... this is absurd.

I fully agree that website owners shouldn't lose the right to determine what goes up on their own website, but with the caveat that this right also comes with legal liability for anything that's on there.
So in effect you're saying moderated message boards should be illegal. Why should that be the case?
No algorithms?!
I agree. Isn't that what we get if section 230 is repealed?
I don't think so, section 230 is only about the platform being on the legal hook for something that's posted.
To quote the law

“ No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider”

It also provides protection for all good faith moderation even if constitutionally protected speech.

However, when Facebook “fact checks” a post, they become a publisher.

And the moderation exemption is very specific to types of content. It doesn’t, for example, allow blanket moderation for political reasons. It specifically mentions lewd/excessively violent content. It isn’t a blanket “moderation” exemption.

Also, “good faith” is explicit. Moderating one political viewpoint while allowing another isn’t good faith. Stop the Steal is censored. Many ANTIFA groups aren’t.

From the law:

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 (B) any action taken to enable or make available to information content providers or others the technical means to restrict access to material described in paragraph (1).

> However, when Facebook “fact checks” a post, they become a publisher.

Says who? I hear this over and over, but there's no legal basis for any of this. It's horse-dung.

> It doesn’t, for example, allow blanket moderation for political reasons

Who decides what's political?

> Moderating one political viewpoint while allowing another isn’t good faith.

Define "good faith". Also, you missed the "otherwise objectionable" part of that law. What's "objectionable" to one platform may not be to another. If I run a message board for adherents of a religion that say, opposes gay marriage, should I not have the option to censor posts that support gay marriage? If I run a message board for environmentalists, do I have to let climate change deniers write whatever they like?

Replying to your first point, I think it's when Facebook adds their own original content. The bit of law quoted above mentions:

> shall be treated as the publisher or speaker of any information provided by another information content provider

Emphasis mine. I think that implies any information provided by themselves they're on the hook for.

The phrase “otherwise objectionable” is so vague that courts have basically interpreted it to mean violations of community standards, or basically whatever. I’m no fan of any kind of political censorship, but they do own the servers. They’re under no obligation to host any content, and we’re all free to go elsewhere.

That said, I think most of their moderation is awful and their fact checking is ridiculous at best.

Social media are amplifiers for pay. What a ridiculous idea that we should leave such devices scattered about without moderation.
Half of this proposal would neuter the amplifier aspect of social media.