I’m sure it’s happening but if the fediverse gets big you can bet your car that they will deanonymize you in the fedi.
I think the advantage that federation provides is that there is no central trove of data you can easily access. With commercial platforms like Twitter or Reddit, all the data happens to be in one place under a single jurisdiction. However, the nature of the Fediverse is such that it’s a bunch of different servers scattered across the globe. Collecting data effectively means having to make deals with each server operator on case by case basis, and governments can only coerce servers that happen to be in their jurisdiction.
You can collect a lot of information by crawling the activitypub endpoints. Its all nicely formatted as json, so you can parse it easily. However, the lack of advertising or other commercial aspects means that companies wont waste any thought on it. And for governments its likely still too small to care.
Yeah, and a lot of the information is public by its very nature. I think this would mostly apply to stuff like private messages or locked accounts as opposed to public feeds.
Collecting data effectively means having to make deals with each server operator on case by case basis, and governments can only coerce servers that happen to be in their jurisdiction.
Federated public content in fact makes it easier to gather this data. It enables for many more entrypoints to the data stream. Of course federated content doesn’t have to be public (eg. XMPP) but so far that’s the direction most ActivityPub apps have taken.
The nature of social media is that the content is typically meant to be public. The reason people use social media is to share their thoughts with the world. However, there is plenty of content that’s not public such as private messages people send to one another, or private accounts that don’t publish their content publicly. This kind of content is much more difficult to mine in a federated environment.
I feel like we’re too small to care about. A lot of us use the Fediverse but also use traditional social media so they capture from there (and though their partners).
What makes you think they do not do this already, CIA honeypots the Tor network same like other attackers. So what makes Fediverse any better right, nothing. Trump community same like n#zis will abuse the network for their evil purpose, same like they do with Tor. See what the so-called truth platform did, they took the Mastodon source code, created their own copy and pasta version and people believed it was new and unique.
You can collect data here, same like on every other network or forum. If you want a fully anonymized platform then in theory there should be no upvotes, everyone gets the same username so everyone can hide and opinions would render pretty much useless in a sense that you cannot connect the dots from a to b over c.
You are already deanonymized the moment you write more than 3-4 comments because your style, your arguing etc. is normally pretty unique on its own, same like your political motivation, or what you post and advertise in such a network.
If your true question is if it will end up like Facebook, the answer is no but can it be abused to extract or infiltrate the network and communities, sure thing. It depends on who hosts the instance, if it is connected to other networks, and who runs the show. A pedo network will sure as hell just use the code run it and never connect to other servers, what I want to say is that you can abuse everything if you want to. The question is how much effort it takes.
I know people do not like to here that but a solid platform depends on strong moderation. Even if someone would abuse the network, the moment you report them the moment his node or instance could be banned or called out, this is why it is so important that people constantly check the network, platform and that they are allowed to report things.
For sure, they do. Just think about the Fediverse instances, which use Cloudflare and/or Google Analytics? They make it for the big player very easy to track you through the whole internet. Or Administrators who use a Gmail or Microsoft email address as contact, forcing users to send mails to tracking companies, when they have a question.
The easier it becomes to contribute a node to Fediverse, the higher the chance of idiots setting up their own instance, adding tracking from unnecessary providers. Either because they’re stupid and think they know what they’re doing, or because they listen to cranks who praise GAFAM like it’s the best the web has ever had to offer.I don’t think so. This sounds like very hard work (expensive) for very little added value to them.
I would think that intelligence agencies have a small group of people thinking about the potential implications of a developed fediverse, but I don’t think that they have invested too much into it. Many of these services are not encrypted and these institutions can likely figure out the physical locations of servers hosting the instances, so from a “national security” perspective, I don’t think that they consider the fediverse a threat.
not everyone is only trying to sell you something
for example gallup pollsData is being collected at a higher level than social media sites. Content delivery networks are capturing the data. Mastodon uses a CDN. I think it’s safe to say that data is being collected across all the sites you visit, including the Fediverse.
to what end?
Maybe but I can’t really see a profitable reason to do so.
To be frank, the OP seems to have no understanding of the underlying technology at play here.
I suggest to the OP to use umatrix webbrowser extension and examining what 3rd-party websites are contacted when you visit say… reddit, vs say… a mastodon instance.
You will see that visiting a mastodon instance does not contact any 3rd-party trackers or ads.
If this is not proof enough, then OP, you should realize that the whole point of the fediverse is to shed ourselves of the shitty ad-tech environment that arose during the 2010’s, there is not a centralized controller of a federated site like mastodon or peertube, so you can always spin up your own server that is completely free of tracking and just use that.
I hope that makes sense.
What I imagine when I think of a scenario like OP is the companies actively scraping Fedi profiles and posts and then using machine learning(?) to map the data to known users. Trackers wouldn’t be necessary.
Of course, that’s a lot of work and I doubt it’s happening soon (if ever). Still interesting to think about.
Yes this
