Plus people apparently don’t know what „algorithm“ means. Sorting by average rating is an algorithm. Filtering by genre is an algorithm. Anything that takes an input (a database of books), performs a discrete set of steps and produces an output (an ordered list of books) is an algorithm. Even if it’s not performed by a computer but yourself standing in front of your bookshelf.

Then block them?

Alright, out of all the comments in this thread, this is the one I’ve reported to the mods. I will accept your arguments (even though the same ones have been made over and over) but what’s up with all this hate? Who hurt you so much that you have to throw paragraphs full of insults at someone?

I guess there are different opinions about what downvotes are for. Personally I think they shouldn’t reflect if something matches my opinion but if it’s worth reading. For me, a downvote says “this is badly written”, “this is rude” or in general “this shouldn’t be on people’s front page”. I will gladly upvote a post/comment that contradicts my personal belief if the author put effort into it.

They can siphon your data no matter what you do. As I’ve said in other comments, everything on the internet has been crawled and scraped for literal decades. This post is already indexed by a bunch of different search engines and most likely by some other scrapers that harvest our data for AI or ad profiles. And you can do nothing about it without hurting your legitimate audience. Nothing at all. There’s robots.txt as a mechanism to tell a crawler what it should or shouldn’t index but that’s just asking nicely (mostly to prevent search engines from indexing pages that don’t contain actual content). You could in theory block certain IP ranges or user agents but those change faster than you can identify them. This dilemma is the whole reason why Twitter implemented rate limiting. They wanted to protect their stuff from scrapers. See where it got them.

Most important rule of the internet: if you don’t want something archived forever, don’t post it!

Exactly the point I wanted to make. Hopefully that came across.

Not my original intention but glad you liked it.

I don’t know if gatekeeping is the right word but some people treat you like traitors if you even suggest that federating with Meta might be a valid option. Just look at the upvote/downvote ratio of this post and some comments I got. Some people are very entrenched in their opinion and I wouldn’t be surprised to soon see posts with “We must defederate from everyone who federates with Meta”.

That might actually be a feasible compromise.

See, that’s the nice thing about a federated solution. I can use my personal instance that federates with Mata and you can use one that doesn’t and we can both be happy and still talk to each other. Being able to pick an instance that suits your preferences is the biggest selling point of the fediverse.

And how many of those Google Talk users would have created an XMPP account if there hadn’t been any federation at all? If someone was on XMPP and asked a Talk user to come over, the odds they would be convinced were very small. Why would they when they can ask you to create a Google Talk account instead?

As I said elsewhere: this would still have been the case if Google hadn’t used XMPP in the first place. All those people you have lost when Google defederated either wouldn’t have been on XMPP at all or care enough about privacy and open source to have both. It’s not like you have lost anyone who had been there before Google started using XMPP. Same with Threads. Every single user who is on the fediverse today will still be here if they defederate (unless they leave for other, unrelated reasons of course). Considering our existing userbase’s interests, I don’t see many people giving up their existing accounts in favor of a Threads account.

I’ll tell you a secret: they care enough to scrape everything. Not only the fediverse, every single website that’s accessible. And that’s not a thing for the future, that has been a reality at least since google became popular. Do yourself a favor and look into the server logs of an average webhost and you will find a whole bunch of crawlers. Some are for search engines, some are for other purposes.

I wrote my M. Sc. thesis on specialized crawlers (back in 2015) and you wouldn’t believe how much research has gone into that and how effective modern crawlers are at finding every single thing that ever got uploaded to the net. The only thing needed is enough hardware to throw at the problem and that’s exactly what Meta, Google, Microsoft, Amazon and all the others have. As a rule of thumb, if archive.org or your favorite search engine has indexed it, everyone else has it as well or has access to someone they can buy it from. There is no such thing as unscraped content on the internet (unless you lock it behind access restrictions and those would apply just the same to federation).

Edit: I don’t have access logs enabled on my instance and obviously can’t see what happens on other instances but I would bet that this very thread will be picked up by at least five different crawlers before the day is over.

When they removed XMPP, they disconnected the Google users from the rest of XMPP. And since it was the largest instance, XMPP as a whole basically died. People couldn’t use XMPP to stay in touch anymore with people using Google’s thing, and vice-versa, which XMPP as a whole kind of came to rely on.

And how is this different from not using XMPP (or being defederated) in the first place? Their platform would still have become huge and pulled over users. People don’t use a platform because they like the technology. They use it because it has people they want to talk to. Even google themselves had to accept that with their failed Google Wave which didn’t even survive the closed beta because people who got an invite couldn’t talk to their friends who didn’t get one.

And I’m explicitly not assuming Facebook will play by the rules in the long run. I’m saying while they do, let’s talk to them, even if it’s just for a month. When they eventually start breaking the rules, we can still defederate and I assure you that we won’t lose significantly more users than we would have if we had defederated earlier. Because why would we? We’re on the side of the open-closed divide that values privacy and open source software. Our only reason to ever switch to Threads is to talk to people that aren’t on Mastodon/Lemmy/Whatever. If that’s a requirement for someone they have to do that either way. But for someone who starts out on Threads, we might suddenly create an incentive to create a Mastodon account to keep talking to their friends. They probably won’t leave Threads but they will use both, pulling the fediverse into the public conciousnes.

I don’t see who this gesture is aimed at. Meta execs won’t care. They lose nothing. The result will be the exact same as building a closed platform from the start. And the users won’t even notice because the mainstream doesn’t even know that the fediverse exists.

I agree that getting people away from Meta should be our priority. But we don’t do that by cutting them off from the fediverse and then begrudgingly making a Threads account to talk to the people we haven’t won over yet. We do that by showing them that there alternatives that they can use without losing access to the content they already follow.

I don’t see how Google killed XMPP. They removed it from their own application which is the exact same effect as not speaking an open protocol in the first place. Nobody forced other XMPP-based applications to change. You know what made me stop using XMPP and not keep my Jabber instance when I moved to a new server? Clunky applications and the fact that everyone I had in my contact list was also on another more comfortable platform (at first ICQ and MSN, later Discord) and prefered to contact me there.

If we federate with Threads, they may use their power to push changes to ActivityPub but nobody forces any of us to implement them, making them effectively irrelevant. If they do something that’s incompatible with the rest of the fediverse, they effectively defederate from everyone else. So either way the end result is two separate sub-fediverses and the more popular one will win out. Hint: it’s gonna be the one with billions of dollars of funding. If we let them in as long as they play by the rules, we have a chance to educate people on what’s beyond Threads, it gives Meta something to lose if they defederate (their users’ ability to talk to people in their friend lists). At least we don’t lose anything that we wouldn’t also lose by not talking to them in the first place.

Have a look at the upvote to downvote ratio on the top level post. Currently 22 up / 9 down. I can fully respect people not agreeing with me, it is a controversial opinion after all. Fine, ignore it or respectfully post an on-topic reply and move on. But actively marking it as “this is bad” or leaving replies that just repeat arguments that I’ve explicitly addressed… I don’t get it. Especially when my last two paragraphs are literally:

I hope this gives you a bit of a different perspective on the whole discussion. I’ll still respect anyone who sees things differently and will actively follow the discussion. I’m just annoyed by all the identical knee jerk reactions that have been floating around.

If I see new arguments or the discussion quality on Threads is bad enough that I have to block half their users I might still defederate for them but for now I’ll keep calm and wait for the situation to develop.

Could you please be a bit less aggressive? The post is titled “Why I probably won’t defederate from Threads” and I’ve given my personal arguments. I’ve also addressed the points you’re making, both in my post and in other comments over here.

It isn’t titled “Why you shouldn’t defederate from Threads” or “Why we shouldn’t defederate from Threads”. It’s not even titled “I will never defederate from Threads”. There are many valid and well-written arguments for defederating. I posted my counter-arguments for a bit of balance yet I have no intention to force or even convince anyone to do as I do.

So let’s please be civil around here and not go around brigading people who have different opinions with arguments they have already heard multiple times. If you want to set a good example, then do that for a respectful discussion just as much as you try to do it for the ethical aspects of federation.

Mostly answered in my other comment but for visibility and completeness, I’ll try to answer it here again.

The content may look different on the surface but that’s mostly because the different applications present them differently. Under the hood, it’s almost identical. Users create posts, other users can reply or like. What differs is the details. Mastodon posts are by default limited to 500 characters, Lemmy/kbin posts can be a few thousand, Plume posts can be even longer. Mastodon and Plume only have likes, Lemmy and kbin add dislikes (downvotes). Posts and comments that come from a different application may look a bit weird in whatever you use but at least they show up in your feed and you have the option to click a link to see them in their original form. In facht, @eatyourglory@mastodon.uno’s reply to your comment came from a Mastodon instance and they will see my reply in their Mastodon feed.

As for the difference between kbin and Lemmy, they’re two different pieces of software that interact with the Fediverse in a very similar way. When someone wants to setup a Reddit-like fediverse instance, they can freely choose between them based on personal preference (I chose Lemmy for my instance because kbin is harder to install and update). Imagine them like being able to choose between a phone by Apple, Samsung, Huawei or Nokia. All those phones have their own specific pros and cons but because they communicate through the same protocols, they can still talk to each other. An instance would be analogous to your Samsung phone or my Apple phone or my partner’s Nokia phone. There are many fediverse instances that use Lemmy as their software same as there are many Samsung phones in the world. The fact that most of them have very unimaginative names (looking at you, https://lemmy.ml/ and https://lemmy.world/) doesn’t help but as positive examples like https://beehaw.org and https://feddit.de. Both use Lemmy and can talk to any other Lemmy instance but their names make it clear that they are their own thing.

Your argument is fully valid and defederating might be the right thing to do for you.

In the end, I have to weigh their cost to scrape my data against my cost to access content that I need. As someone who has built scrapers for scientific purposes before, I can tell you that building something that scrapes Mastodon and Lemmy instances is not a single cent more expensive than getting my data through federation. It’s also probably a lot more reliable because they can get everything, not just what their users subscribe to. On the other hand, my cost for accessing my friends’ and family’s posts as well as corporate social media accounts if I don’t do it through federation is creating an account in their proprietary app. And then they will be able to get a lot more of my data than they could ever scrape from my Mastodon profile.