Gab, Mastodon And The Challenges Of Content Moderation On A More Distributed Social Network

from the experimentation-through-federation dept

While so many of the discussions and debates about content moderation focus on a few giant platforms -- namely Facebook, YouTube and Twitter -- it's fascinating to see how they play out in other arenas. Indeed, one of the reasons why we're so concerned about efforts to "regulate" content moderation practices on social media is that focusing on the manner in which those big, centralized platforms work could serve to stifle newer, more innovative platforms, whose very set up may inherently deal with the "problems" in the first place (see my protocols, not platforms discussion for one example).

There are a few interesting platforms out there trying to take a different approach to nearly everything -- and one of the more well known is Mastodon, an open source "federated" system that is sort of somewhat like Twitter. If you somehow have missed the Mastodon boat, I'd recommend the long piece Sarah Jeong wrote about it two years ago, which is a pretty good intro to the topic. The really short version, though, is that anyone can set up their own Mastodon community and, if others so choose, they may "federate" with other Mastodon communities. You could build a Mastodon instance that is totally isolated from others, or you could build one that connects to others and allows "toots" to go from one instance of Mastodon to others. And, of course, the federating can change over time. It's kind of neat in that it allows for multiple communities, who can set different rules, norms and standards, and thus you get much more widespread experimentation. And, unlike a fully centralized system, like Twitter, the ability for different instances to just "go there own way" if they disagree, allows for much greater flexibility, without a centralized content moderation impossibility.

I'm still more interested in much more fully decentralized protocol-based systems, but a federated system like Mastodon, that allows for a distributed set of mini-centralized instances that can join together or separate as needed, is still pretty fascinating.

However, it got more fascinating and interesting earlier this month when the social network Gab moved to Mastodon. If you haven't followed this space at all, Gab likes to call itself the "free speech alternative" to Twitter, but in practice that has meant that it's the place that many trolls, racists and other general assholes have gathered after being kicked off of Twitter. Gab announced, back in May, that it was planning to shift its platform to Mastodon, setting up its own instance. In theory, this solved some "problems" that Gab had been facing -- starting with the fact that Apple and Google had removed Gab's mobile app from their app stores (something Gab sued over, in a strategy that was not very successful). Since there are a bunch of Mastodon apps that allow users to log into any particular Mastodon instance, Gab itself made it clear that this was a key reason for the move:

Of course, building on top of someone else's better tested open source code probably also helps Gab with the long list of technical issues the site was having. And then there's the pure troll factor. Besides harboring social media trolls, Gab, as a company has always sort of gleefully taken on a trollish roll in the way it works as a company as well. And, considering that part of the very reason that Mastodon's creator, Eugen "Gargron" Rochko, set up Mastodon in the first place was to build an alternative to Twitter that was free of Nazis, assholes and trolls... it was a truly trollish move to jump onto that platform and at least imply to many a plan to "invade" (or, perhaps we should say brigading) the wider "fediverse" of Mastodon.

The switch over happened earlier this month and it's been fascinating to watch how it's all played out. The shortest summary might be that the federated model has shown to be somewhat resilient so far. Mastodon itself put out a statement urging various Mastodon instances not to federate with Gab and also suggesting that the various Mastdodon app developers choose to blacklist Gab's domains from their apps (meaning that Gab's plan to use this to get back into the app store might not work as well as planned).

The Verge has a long, in-depth article about how all of this is playing out, and it seems like, as a federated system is designed to do, different parts of the system are experimenting and figuring out what makes sense. Most of the other instances have decided they don't want to federate with Gab.

If you join a major Mastodon instance right now, chances are you won’t be connected to Gab. “All the admins that I know, that I interact with myself, have already blocked Gab,” says Rochko — including Mastodon.Social. “Essentially, they’re isolated.”

As for the various app makers, they're figuring out what they want to do:

This has turned app access into a battlefield. Developers can lock Gab out by disabling login options to the instance or completely blocking content from its servers. And several have done just that. Mastodon lists six major mobile apps on its homepage. Four of them — the Android client Tusky and the iOS apps Toot!, Mast, and Amaroq — block Gab in some fashion.

Amaroq developer John Gabelmann banned Gab to avoid potential problems with the App Store. “My core objective is to keep Amaroq publicly available and to abide by all Apple policies, which keep unmoderated extremist/hateful content off the store,” he tells The Verge. “If your network is large enough and unmoderated enough to get the negative attention of Apple, Amaroq will follow Apple’s policies.”

Mast’s creator Shihab Mehboob, by contrast, blocked Gab after users requested it. He’s gotten one-star reviews from angry Gab users, but “if hate speech is masquerading as free speech on an app I’ve built, it’s upon myself to somehow moderate that and reduce it where possible,” he says. “I understand that the Fediverse is intended to be open and entirely at the user’s discretion as to what they want to see/use/partake in, but that shouldn’t cover Nazi-based ideologies. There has to be a line drawn somewhere.”

Other app developers maintain that this blocking doesn’t fit Mastodon’s mission. The Android-based Fedilab app’s free version initially blocked Gab because of Play Store content policy fears. But the ban has since been lifted. “I will simply not block instances with the app,” wrote Fedilab’s developer. “I clearly think that’s not my role … If you want a strong block, it’s in the hands of social network developers or your admins.”

And the developer of Subway Tooter, who goes by Tateisu, is skeptical that stores will censure apps for supporting Gab. “They can run their web app on a web browser,” Tateisu points out. “If Google wants to ban it, they should start from their Chrome web browser.”

This is all quite interesting. It's also the kind of experimentation and more distributed decision-making we'd like to see more of online. This is not a truly distributed system where the power is moved all the way out to the ends, but it is a federated system where the power is moved to various nodes -- leading to more competition and variety. The fact that most of the major Mastodon instances have said they don't want to federate or deal with Gab is an expression of preferences, and in many ways a better overall system than one in which a single company is making the decisions. And it's much better than politicians telling companies what they need to do. Of course, in a world without Section 230 -- or one with a nonsensical requirement for "neutrality" in platform moderation -- would these options even be available?

Filed Under: content moderation, distributed systems, federation, mastodon, moderation, platforms, protocols, social media
Companies: apple, gab, google, mastodon, twitter

Why A Blockchain-based DRM Has Always Been A Terrible Idea

from the and-why-copyright-maximalists-gravitate-towards-it dept

I've threatened in the past to write up a post explaining why a blockchain-based DRM is a terrible idea that will flop -- and it appears I finally need to do so, with the sort of announcement that Sony is preparing to use the blockchain for "next-gen DRM." I should note that, unlike some people, I'm actually not a blockchain skeptic. I think that it does have a few potentially revolutionary and disruptive uses. But... I also think that nearly every use of the blockchain that has been championed so far is incredibly silly and pointless. In most cases, what people claim they're using a blockchain for would better be served with... a database. If you're just replacing a database with a blockchain-based system, all you're really doing is adding unnecessary inefficiency and complexity.

So while a blockchain does have efficiency and complexity weaknesses compared to a database, it does have two potential advantages -- but only if those advantages are necessary to the service being built. The first advantage is that the blockchain can be truly distributed, rather than centralized. For years, we've discussed the problems of too many centralized systems, whether it's the siloing of information, the weird incentives it creates for the central database controller, or simply the fact that a centralized system creates a single point of failure and/or point of attack for a would-be assailant. A blockchain can help limit (though not eliminate) some of those problems -- and that can open up some incredible new services. The second big thing that a blockchain does better than a database is that it creates a more trustworthy way to prevent the "double spending" problem.

The issue there is that with anything digital, it can always be copied and/or manipulated in some way. If you are trying to construct something that requires scarcity -- such as a digital currency or a specific ledger of asset ownership -- then you want to be sure that the system really has a 100% accurate record, and won't allow the same bit of digital currency to be held by multiple people (or allowing it to be spent multiple times by the same person) or, that the same asset is listed as being owned by different entities. One of the cool features of the Blockchain is that it is designed such that people can be fairly cryptographically certain that we don't have that sort of "double spending" problem. You do have to trust the math and the code, but the code is open and people are constantly checking it. Now, you can claim a centralized database can prevent these kinds of things too, but you have to totally trust whoever is in control over that centralized database. And you might. Most of the money you probably have is really in a centralized database at your bank. But, there are some advantages to have that record be on a publicly distributed ledger a la the blockchain.

The issue, of course, is finding services and applications that can really take advantage of these benefits of the blockchain, and so far, they are few and far between, though there are plenty of future possibilities where they could be super useful.

For years now, we've heard some people arguing for a blockchain-based DRM. This idea is at least marginally better than simply replacing a database with an unnecessary blockchain, because at the very least, it is an attempt (a weak one, but an attempt) to leverage one of the advantages I discuss above: the double spending issue. Obviously, as lots of people will tell you, the legacy copyright industries have decried the fact that the internet makes content super easy to copy, making their legal monopoly over the distribution of that content less than monopolistic. Thus, the folks who wish to go back to a world in which content is locked up, hear about how a blockchain "solves" the double spending problem, and they get excited: why couldn't we use that as DRM? After all, isn't the point of that aspect of the blockchain that it stops copying of digital assets?

But, that's about as far as the thought process goes. Because from there, it completely breaks down. There are few actual details about Sony's blockchain-based DRM idea, but it's not difficult to understand why it will fail. First, it's important to understand something that copyright supporters frequently forget: the copyright on something is different from the content itself. Copyright system supporters like to conflate the content itself and the "intellectual property." But as we've explained in the past while a "copyright" may have property-like elements, the underlying content does not.

The blockchain-based system for solving the double spend problem is a useful solution when it's the record or ledger entry that you don't want copied. But that's not what any DRM system would be. Because the "record" is the copyright information -- not the music/movie/book/etc. And who cares whether or not you can copyright the copyright information? You're solving the wrong problem? The content itself can still get copied. There's no way to stop that, because even if you were to somehow encode the actual content in the blockchain (a pointless idea), you'd still have the analog hole to deal with, as the content would inevitably escape the blockchain.

The other reason why a blockchain-based DRM solution is so dumb is because it actively goes against what the public wants. The reason a disruptive or transformational technology works is because it provides the public with something much better than they had before. A blockchain-based DRM solution provides a worse solution. There is no demand for such a thing. Sure, there may be "demand" on the label side, but that doesn't translate to usage.

The final reason why a blockchain-based DRM solution is utterly stupid is conveyed quite nicely by Cory Doctorow in his BoingBoing post about this announcement, in which he notes that what Sony is proposing appears to be a privacy nightmare:

Sony also implies that every transaction in which someone buys a creative work will end up in the ledger. This has extremely grave privacy implications, but it also has nothing to do with preventing copyright infringement. People who lawfully acquire copyrighted works have the right to sell them, lend them, and give them away -- and they are not liable if (for example) their data (including copyrighted works) is stolen and released online.

The fact that Sony publishes a list of the reading, viewing and listening habits of every one of its customers does not give it any basis for seeking damages from those customers if works they purchased show up in someone else's hands.

The whole idea of a blockchain-based DRM is based on a faulty understanding of either the blockchain, copyright, or (most likely) both. It involves people who wish to return to a world of artificial scarcity around content not thinking past the fact that a blockchain lets you solve the double-spend problem, without understanding why that doesn't make any sense at all in the world of copyright, where it's the content that's the issue, and not the copyright.

Filed Under: blockchain, copyright, cryptocurrency, distributed systems, double spending, drm, ledger
Companies: sony

Arresting People Associated With Anonymous Unlikely To Have The Impact The Feds Expect

from the things-don't-work-that-way dept

One of the big stories of yesterday was the wide array of raids and arrests by the FBI in order to arrest people they claimed were members of "Anonymous," who took part in various denial of service attacks. All day yesterday the number of people arrested kept growing. I first saw three, then 12 and the final tally was apparently 16. Apparently the arrests are specific to the efforts to take down Paypal after Paypal decided to stop letting payment transfers go to Wikileaks. The specific charges are "intentional damage to a protected computer" and conspiracy.

Now, I've been very clear since Anonymous started this effort -- shutting down various websites using what is effectively crowdsourced distributed denial of service attacks -- that I think the strategy is really dumb. Does it get attention? Yes. But it turns parties doing questionable things into victims. It doesn't open any new eyes to the problems Anonymous should be trying to highlight. It just draws attention to the attacks themselves. It just seemed really likely to backfire -- especially as law enforcement and politicians focused in on the attacks, rather than the reasons for the attacks, and we're seeing some of that now.

But I can't deny that their efforts, combined with the slightly more sophisticated hacking efforts both from Anonymous and the spinoff group LulzSec and others, have actually had a much greater impact than I expected, especially with things like the hacking of HBGary and the release of ACS:Law files. As I noted last month when a bunch of people were arrested in Europe with claims that they were members of Anonymous, it's not clear to me that these arrests will have much of an impact, really. Will it scare off some random kid from becoming a scriptkiddie? Maybe at the margin there will be some. But the thing is, the types of folks who get involved with these things tend to overestimate their own abilities, and dramatically underestimate the likelihood of getting tracked down or caught.

And given the very distributed nature of the group (i.e., that it's not actually a group at all), it kind of makes you wonder if the arrests will only serve to get more folks jumping into the effort, perhaps for increasingly misguided reasons. As we've stated, governments and law enforcement seem to be taking a top-down approach to this, as if they were rounding up a criminal gang, not recognizing the distributed nature of this effort and how the focus is not criminal, but ideological. Arresting people just drives home their general fear of a world in which certain entities have too much power, leading more people to hit back.

I still don't think their strategy is smart. And I don't think it'll really create lasting positive change (in fact, the backlash could do the opposite). I also worry quite a bit about what happens when they suddenly rage against an innocent party or a group or an individual who really doesn't deserve their wrath. But, at the same time, I can't see how a big FBI crackdown does anything positive, either. It just serves to reinforce their general point. And, with something like the DDoS on Paypal, it seems a bit ridiculous to suggest that it really created that much "harm." It was, as many noted, a modern version of the sit-in. Yes, it probably was a nuisance and cost some people money, but it lasted for a short while and it's difficult to argue there was any lasting damage.

Defenders of law and order will insist "something" needed to be done, and will believe that these arrests will scare off people from the next round of attacks. I think those people are greatly underestimating how people who feel disenfranchised by the world, but sense power through their internet connection, react in such situations. Punishment for the sake of punishment may make sense to some people, but I prefer that the focus be on actually getting to the root of the problem, rather than trying to attack the symptoms in a way that makes the cause grow bigger.

Filed Under: anonymous, arrests, distributed systems, fbi, hacking

Will Arresting 'Anonymous' Members Help Or Hurt Anonymous?

from the just-wondering dept

I'm on record as saying that I think the activist hacking by groups like Anonymous, designed to take down websites in protest, are not particularly smart or useful. I will admit, however, that I'm surprised at how effective they've been in drawing additional media attention to certain stories, and how they really have helped drive two particular stories forward (the ACS:Law situation in the UK, and the HBGary Federal story in the US). And while I still don't think it's a particularly effective overall strategy, I must say that I find the whole effort (and similar efforts from "groups" like Lulz Security) fascinating. I've also pointed out that what may be even more interesting is how traditional law enforcement and governments can't really grasp what they're dealing with.

I think that point may be driven home with two separate governments claiming success in "arresting" Anonymous members. Spain got some attention for supposedly arresting three "Anonymous" members late last week, and then Turkey also got some attention for arresting "32 members."

If these individuals were involved in actual criminal activities, then the arrests are perfectly reasonable. But if these governments actually understood what was going on, and actually understood what they were dealing with, they wouldn't have said they were arresting members of "Anonymous," which is hardly a group anyway, and the word "member" is misleading. Taking a traditional top down approach, these governments think that by announcing that they've "arrested Anonymous members," they're likely to scare people off from being a part of Anonymous. It's very much speculation on my part, but knowing the sort of people involved, my guess is that it's having exactly the opposite response. By presenting the people arrested as being a part of Anonymous, these governments are glorifying Anonymous, and adding to the allure. And the arrests are unlikely to scare off too many actual participants, since they quite likely think that they are better at covering their tracks... and as things move along, that will likely be true.

These governments could have easily arrested people and charged them with hacking without making the connection back to Anonymous. Making such statements is like responding to trolls on internet forums. It's tough to resist sometimes, but it only encourages more such activity. And, of course, just as with trolls, this sort of thing really only plays into exactly what Anonymous wants. It builds up the group's own profile, increases the very necessary mythology, and likely improves the efforts to do more such activities (while protecting participants even more).

In a world with increasingly distributed power, Anonymous and others are really a precursor to what's coming down the road, demonstrating how certain forms of activism really don't work the way people in power expect them to. I don't agree with their specific tactics, but I'm fascinated by their ability (intended or not) to get companies and governments to play right into their strategy.

Filed Under: anonymous, arrests, distributed systems, hacking, spain, turkey