News for hackers. Stuff that matters.
Re:On the Twitter Hack by stickfigure on Monday July 20, 2020 @11:46AM (#23898957)
How exactly do you propose to split up Twitter? Split the userbase into 100 groups? Who goes in which group? And what happens when people want to switch groups... ultimately coalescing in a single node? Because, you know, network effects.

I don't like Twitter either, so I don't use it. It's a really effective solution!

Re:On the Twitter Hack by basch on Monday July 20, 2020 @12:00PM (#23899124)
There's other ways to deal with anti-trust besides breaking the companies apart. Data portability, interoperability, ownership of your own social graph. If these platforms were forced to act more like email, where leaving google for microsoft doesnt mean i can no longer contact my friends, the network effects that give them monopoly-like power diminish.

Imagine how ridiculous email would feel if you need a gmail account to email gmail peeps, a microsoft account to email outlook peeps, and another separate account to communicate with people in every mom and pop community. Imagine not being able to export your address book in an industry standard format. Imagine not being able to use whatever third party client you want to access your own data.

Re:On the Twitter Hack by auganov on Monday July 20, 2020 @02:16PM (#23900412)
> "ownership of your own social graph"

What would that mean?

> If these platforms were forced to act more like email

Who decides who's "following" who?

Say I decide to use a competitor. Do I then get to "take" my Twitter followers and can still broadcast to them on Twitter from a different app? Under what conditions can Twitter block your activity? Either they will have to let a lot of spam in or they'll have to ban most non-Twitter activity from coming in.

How do we define interoperability? Do we make Twitter as it is right now a government mandated standard? How do we introduce new features?

I just don't see how you can do all this or how it makes sense at all.

Although email does certain things well it's not without issues. Running your own server is tough. Even many fairly mainstream services have deliverability issues. Spam is rife. And it's especially a big problem when running your own or using a small provider. Email is terrible for initiating unsolicited interactions (in some places might even be illegal).

Re:On the Twitter Hack by ssivark on Monday July 20, 2020 @06:38PM (#23902622)
All this is actually worked out in some detail (search keywords federated/indie web). Check out the PubSub model, and the ActivityPub protocol as an example, and Mastodon as a working proof-of-concept.

As a starting point, imagine email & RSS and work forward from there. I'm not saying it's trivial, but it's also not insurmountable.

Re:On the Twitter Hack by auganov on Tuesday July 21, 2020 @02:52PM (#23909672)
I did actually do a lot of reading on ActivityPub yesterday and it's exactly the mess I would have imagined it to be.

The standard doesn't define all that much. Arbitrary undocumented JSON fields are perfectly fine and used heavily. Just following the standard you wouldn't get too far, would need to follow Mastodon's implementation to be truly compatible [0].

Many instances have already banned each other, some client apps have outright hard-coded blacklists in. So again would the government forbid that? But then what do you do about unruly instances?

The whole model IMO is pretty wrong in favoring instances over individuals. Twitter is a free-for-all, this model is completely different in optimizing for small groups of like minded people.

Secure private DMs are pretty much impossible in AP.

Rate limiting, spam protections etc are left completely up to the implementer. So would the government define that under an anti-trust scheme?

Now if one considers it to be just some low-level building block for distributed social then okay, maybe it's fine. The idea is basically social-related stuff in JSON over HTTP. Okay. But it doesn't really solve much of anything in practice. It offers a strictly worse experience in name of decentralization. This is exactly why I'd be afraid of the government stepping in and forcing these kinds of decisions on companies.

[0] And then it only gets you compatible with other Mastodon-style instances (and most seem to use the exact same implementation, just different themes). There are other ActivityPub apps that operate under a completely different UX/UI scheme. Which is both nice and weird in that the standard has obviously been written with Twitter-like stuff in mind.

Re:On the Twitter Hack by stickfigure on Monday July 20, 2020 @02:04PM (#23900321)
So let's imagine Facebook/Twitter/etc was based on open protocols like SMTP...

* You'd have the same spam and abuse problem that email has.

* You'd have zero control over the information you post to it, just like email. There would be absolutely no way of stopping or even slowing down the Cambridge Analyticas of the world.

"Ownership of your own social graph" is complete and utter nonsense. As soon as you share data with other people through open protocols, they own it. Just like email.

Even email is tending towards centralization, with a few big players that can manage the spam and abuse and still get their messages delivered.

Re:On the Twitter Hack by yjftsjthsd-h on Monday July 20, 2020 @07:58PM (#23903128)
> There would be absolutely no way of stopping or even slowing down the Cambridge Analyticas of the world.

They wouldn't be as much of an issue in the first place, because there isn't a single company with a single collection of data to siphon through a single API with a single entity that has to give you access in order to collect whatever you want. A Cambridge Analytica that has to separately try to get data from O365 and GMail and FastMail and ProtonMail is much less dangerous.

Re:On the Twitter Hack by basch on Monday July 20, 2020 @02:24PM (#23900483)
>You'd have the same spam and abuse problem that email has.

Don't you have that already, and they have anti spam measures? How would a facebook post coming from facebook or twitter change what they remove as spam?

Re:On the Twitter Hack by jsnell on Monday July 20, 2020 @02:51PM (#23900732)
What you're basically proposing is a model of spam where only the content of the message is considered. I do not think that would be very effective in practice, in the modern world.

Email spam filtering is already a lot more powerful, since it can consider things such as IP reputation. (IP reputation would be a non-starter for a federated social network, since it would require passing raw IPs along with messages in the federation protocol. And that is just unthinkable from a privacy perspective).

When all entities posting messages on a social network are managed by the same identity provider, and the identity provider can share information about the account with the social network, you get yet another layer of increase in power. You can, for example, tell whether the account is newly created or not. Or whether it was bulk-created along with 10 other accounts posting similar spam. Or a dozen other account reputation signals which let you classify accounts into organic vs. abusive, and use that to inform the classification of messages from those accounts. And again, the only way this happens is if the identity provider and the social network are one and the same. If they were separate companies, it'd be inexcusable for either party to share detailed information with the other.

Re:On the Twitter Hack by basch on Monday July 20, 2020 @03:51PM (#23901243)
I dont see why the identity providers couldnt pass trust ratings between each other.
Re:On the Twitter Hack by jsnell on Monday July 20, 2020 @04:18PM (#23901455)
There's one element of the tragedy of the commons. Honest Bob's Social Media Emporium might not be entirely motivated to mark their users as untrusted, when most of the pain of the spam is felt by the other thousands of providers they federate with.

"Hey, doing SMS verification of new accounts is expensive, but required by the other nodes on the network. How about we don't, but say we did?"

Imagine a solution to email spam that replaced looking at IP address reputation with looking at the sender's claim about their own reputation. Laughable idea, right?

The other problem with your suggestion is that you can't do abuse prevention with a single signal, even if it's not Boolean. I even gave an example in the previous post, with checking whether accounts doing similar abuse were clustered together at account creation time.

Re:On the Twitter Hack by basch on Monday July 20, 2020 @05:06PM (#23901912)
Then " Bob's Social Media Emporium" as a whole would get dinged in the trust web. If you cant trust the verifier, you wouldnt trust any of their participants.

I guess nitpicking the details doesnt matter, but none of these problems seem insurmountable.

Re:On the Twitter Hack by jsnell on Monday July 20, 2020 @05:25PM (#23902085)
Oh, man. You dropped the "Honest" out of "Honest Bob". That's harsh, it's like you think he's some kind of a crook :(

Once the biggest node on this network starts not peering with smaller nodes due to them being untrustworthy, what do you think happens? I suspect that the users will move from the smaller nodes to the bigger ones. And then we're right back to where we started. People will be complaining about monopolies and will want the big nodes broken up.

Re:On the Twitter Hack by dane-pgp on Monday July 20, 2020 @07:46PM (#23903065)
This does seem like a genuine problem, and there are already signs of this happening in the Fediverse. Perhaps it needs a name, so that people can notice when it is happening, and discuss potential solutions to it. The name I propose is "Federator's Dilemma", as it seems like a game theoretic problem.

As for a solution, my intuition is that the first step might be to create a Sybil-proof voting system for determining how popular individual services are. If a service is refused federation by other services which collectively contain 90% of user accounts, then that's a sign that the service doesn't do enough to stop its users from posting objectionable content. Similarly, service A could report to service B how many abuse reports it is getting from its users about people on service B, with a clear rule about when service B would be blocked.

This isn't a perfect system, and I admit that a Sybil-proof global anonymous distributed voting system might be harder than just solving the Federator's Dilemma more directly, but I think it's worth considering, not least because such a voting system would be useful for countless other applications online.

Re:On the Twitter Hack by baddox on Monday July 20, 2020 @07:47PM (#23903073)
> Imagine how ridiculous email would feel if you need a gmail account to email gmail peeps, a microsoft account to email outlook peeps, and another separate account to communicate with people in every mom and pop community.

Maybe this is the problem you’re hinting at, but I expect it would be very rare for me to ever run into someone I couldn’t email because they’re not using Gmail (especially if we include G Suite).

Re:On the Twitter Hack by metalliqaz on Monday July 20, 2020 @11:55AM (#23899065)
How exactly do you propose to split up AT&T? ...and so on from there.

The real answer is to mandate interoperability and equal treatment of competitor content. Then competition could grow on its own.

Personally I would just regulate it into oblivion.

Re:On the Twitter Hack by stickfigure on Monday July 20, 2020 @02:11PM (#23900378)
AT&T was a monopoly of local phone services. It was split up into geographic regions, each with their own independent physical infrastructure.

There's no physical locality to Twitter. You can't put all the California users in one system and all the Washington users in another.

Re:On the Twitter Hack by Timpy on Monday July 20, 2020 @12:27PM (#23899372)
> I don't like Twitter either, so I don't use it. It's a really effective solution!

Schneier's point is the threat of a malicious actor starting a war by hijacking a politician's Twitter is non-zero. People like you and me opting out of Twitter is not an effective solution to that.

Re:On the Twitter Hack by staycoolboy on Monday July 20, 2020 @05:25PM (#23902076)
If enough people opt, it is a solution. Like littering: if I throw one cigarette butt out the window, is that really an environmental hazard? No. If a hundred million people each throw 20 out the window a day, then you have a problem.
Re:On the Twitter Hack by Timpy on Monday July 20, 2020 @08:47PM (#23903382)
Alright, well I'm avoiding twitter as hard as I can so I guess any day now the President will stop affecting public discourse via Twitter.
Re:On the Twitter Hack by staycoolboy on Tuesday July 21, 2020 @12:15PM (#23908087)
You and me both. I can't tell if you're jibing, but I do think twice before clicking on links that come from certain places, which I guess is "avoiding harder". :)
Re:On the Twitter Hack by dogfoods on Monday July 20, 2020 @08:36PM (#23903319)
We used to have forums and blogs organised by interest topic or community. An elegant mode of organisation, for a more civilised age.
Re:On the Twitter Hack by notafraudster on Monday July 20, 2020 @09:08PM (#23903467)
I mean, reddit and Discord are still this. They both graft on cross-community navigation and account systems for user convenience and increasing user retention.
Re:On the Twitter Hack by bikeshaving on Monday July 20, 2020 @06:55PM (#23902757)
Rather than splitting up Twitter, what if we just stopped the constant process of acquiring and dismantling any company which dares to threaten the big tech companies. Vine should still be around.