this post was submitted on 05 Jul 2023
1850 points (96.8% liked)
Technology
59168 readers
2521 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Embrace, extend, extinguish. Only proven way to destroy decentralized, free, open source solutions.
First stage embrace might not even be malicious, but with corporations it will eventually lead to someone thinking: how can we monetize our position. It is just nature how business works.
https://en.m.wikipedia.org/wiki/Embrace,_extend,_and_extinguish
It's worth pointing out that the wiki article lists several examples of Microsoft using this approach but I wouldn't class many of them as successful.
Not only was it not very successful, it's an old outdated Microsoft playbook from the 90s/early 00s and was targeted at closed source competitors and freeware, not open source software where you can just fork out a separate version.
By all means block Meta instances if you want, but they have 3 billion users, they definitely don't give a shit about a "competitor" with a few hundred thousand users. If simply the presence of a corporation in the Fediverse is enough to destroy it, then it wasn't going to last long anyways. It's embarassing that "embrace, extend, extinguish" caught on around here just because it's a catchy alliteration.
Let me offer a rebuttal. The fact that this playbook even exists and is well-known is a cause for concern. Yes, Microsoft’s campaign wasn’t very successful, but that doesn’t mean Meta won’t try or learn from Microsoft’s mistakes. I ask: is the probability of this happening non-zero, and if so, is it lower than you’re comfortable with? For me, and many others here, that answer is no.
Moreover, this is a greater problem: Meta is well-known and has practically infinite marketing budget. They can spin their app as the de facto, causing many people to lose control of their data. By association, some people will blame the Fediverse and not Meta. Defederating signals that we are not willing to participate with them and tells potential Fediverse users that they will not be able to engage with us—and whatever they decide, we cannot impact more.
The crux of my argument is risk management. Defederated is a conservative measure to prevent possible issues in the future.
Honestly this is just pure paranoia because nobody has given a solid reason as to why they would give a single shit about the few hundred thousand users here. Your only argument is "well it exists, so maaaybe they'll use it but better" which has no basis. As for losing control of your data, you have no control of your data here. It's public information. Any person, corporation, computer literate cat, etc can already scrape everything you post here. Don't mistake anonymity for data privacy.
Like I said, block em, defederate, whatever measures you want to take are an option, but for the love of god let's just stop parroting nonsense at eachother because it sounds clever. I came here to get away from reddit culture.
I just wanted to say, I am by no means technical but your position is exactly what I was thinking, if an open source project can't survive when it's competitors start using it, then it's never going to survive. The whole point is for it to be interoperable, resilient, and antifragile, and there are plenty of open source projects that achieved that. Competitors switching over to open source is a natural progression of any open source project if one assumes it is successful.
If they don't give a shit then why do they add federation feature at all? It doesn't make sense.
Right now it's only supported for Instagram accounts right? So slap in ActivityPub and you've got an already working way to extend your app. It's easy, it's fast development, and it's cheap. It makes tons of sense.
Also, Meta and the rest of FAANG are a company of a bunch of nerds with a history of open sourcing software. This isn't some crazy play, this is completely normal for them.
Yeah and it’s also normal for them to act like sociopaths and shrug and say “sorry, this is just how capitalism works” when it gets exposed how cynically awful they been behaving.
There is zero evidence ethics will be followed here, Silicon Valley has spent decades building a good argument the precise opposite will happen.
What does ethics have to do with any of this? Like you said, it's all capitalism. The total amount of users in the fediverse is a rounding error on their 10-K. Why would they care about stealing the userbase?
Corporations don't act ethically unless they can monetize it or they're regulated.
Counterpoint: it’s not about capturing the current audience so much as heading a threat off at the pass.
I’m not going to argue way or other re: defederation. Just putting myself in their shoes and looking at the field they’re entering. They likely recognize there’s a brief window right now to capture twitter’s disaffected audience as they stumble while a nontrivial subset of those users are exploring open-source, non-corporate alternatives.
It makes perfect sense for them to cast the widest net they can in this moment. And it also makes sense for them to try to stifle the non-corporate side before it has a chance to gain any solid footing.
Exactly. Which is why I believe that all this fearmongering is because of Meta's reputation (rightfully so) rather than because Meta actually has a plan to destroy the fediverse. And it's not the like the fediverse can be actually destroyed, people can always start new instances at any time.
My take was that most people 1) don't want Meta/Facebook spam - low effort memes, propaganda, etc. and 2) don't want their content to be used by Meta. The former seems pretty easy - just defederate and you don't see any of their crap. The second is sort of a gray area... Whether or not you are diametrically opposed to Meta/Facebook or not, once you post your content to a public site, it's available. I haven't been here long, but defederation seems to work both ways, so FB would have to scrape content from known instances to get that content unless I'm mistaken.
FB could smoke any instance by DDOSing scrapes whether intended or otherwise, but once you post your data on a public forum, Meta could theoretically use it.
But to your comment - I don't see what starting a new instance would do for anyone for #2. Any new instance is discoverable by nature, so FB can come knocking at any time for content whether you defederate or not.
I’m just here for the beans
In Microsoft's case I agree. However Google successfully used EEE to essentially kill of XMPP where they initially added XMPP support to Google Talk, then extended it with their own features which weren't up to spec, and then later killed off XMPP support.
So when's extinguish come in? XMPP still exists, google dropping support didn't kill XMPP, it just doesn't work with their app anymore. They weren't trying to kill XMPP, they were just going what Google does and dropping projects as soon as they aren't profitable.
Google successfully did this to XMPP.
https://ploum.net/2023-06-23-how-to-kill-decentralised-networks.html
Thank you for this article. It shows exactly what's Facebook's plan. They will join in, make their own implementation that doesn't work well, pass the blame to the other platforms that use the protocol*, which in turn pressures them to debug and slow down themselves around Facebook's stuff, and then they cut them off entirely.
The correct attitude is to extinguish Facebook now. They're not welcome.
*And yes, this would work. Users are absolutely gullible about this shit, even without ever being told anything directly. Look at Apple users and their blue/green speech bubble thing. Every single flaw with the system is Apple's fault - but the dumbass cultminded users see the green speechbubble and blame the other users for the flaws, not Apple. They literally just did the stupid tribalism comic and it worked.
I’m not going to say you are wrong, but I have yet to meet a single fucking person that actually cares about bubble colors.
I hear this parroted so often, but never see it myself. Didn’t see it when all I had used was Android devices, didn’t see it when I tried an iPhone and got involved in their own communities.
XMPP still exists. Google dropped support for it, that's definitely not killing it. Google drops support for projects all the time by the way, it's kind of their thing.
Google dropping support for XMPP is what put it one foot in the grave. They abused the protocol to gain the lion's share of users for Google Talk, and then cut off any resistance that remained. It exists still, technically, but when's the last time you heard about or used it? I only know about it because EVE Online players used it for large group text communication before Discord became a thing.
XMPP still exists in the same way that critically endangered animals still exist: barely and by the adamant will of some dedicated few.
XMPP wasn't even remotely popular until Google integrated with it, I tried Jabber back in the day lol. Google brought the users it lost, you can't argue this was an attempt to kill it. At worst it's the same as before Google integrated.
That's the problem though. If XMPP had grew organically then it would fare much better. With how it happened, XMPP's growth was mostly because of Google, and that put a lot of pressure to other servers and the protocol's development to cater to them, because they had the majority of the users in their platform.
App I work on, we're replacing XMPP with messages over push/rest/websocket. XMPP is not fun to use compared to newer stuff.
enthusiast dev here, can vouch, having to make a XMPP library for myself for a bot I ran, I HATE the protocol with a burning passion, it's weird and not how you would expect it to be. I'm sure the complexity of the standard didn't help against its downfall. That being said, fully think that it will be harmful in the longrun of Activity Pub for Meta to be jumping in. but there will be some enthusiasts that still use it regardless.
It looks like articles today are saying that Meta is delaying integrating ActivityPub at launch.
That said, I'm not seeing how we get to the last E, extinguish. By its very nature, ActivityPub is decentralized to avoid total control. So even if Meta embraces the technology and wants to monetize it (because capitalism, of course), extending ActivityPub would (hypothetically) be open source - or they would fork it, diverging and making their version closed, and otherwise not function in full with other ActivityPub instances (like with kbin, Lemmy, and Mastodon). Without buying the platform from the developers in full, I don't see how ActivityPub or the greater Fediverse dies. And I could just be missing something obvious, so if you can explain how we get there, I would really like to hear and understand.
I guess the only way I could see it is if Threads got so popular that people literally stopped using the other apps - but I also don't see that happening, because anyone already using stuff like Mastodon are using it because Twitter, Facebook, etc, suck ass and they've moved away from sites like that.
EDIT: Thanks to the one person that actually replied, I saw I was on the right track at the end, but failed to see the obvious (as I assumed).
It’s hard to predict but the extinguish part would come from bigger non-Threads instances implementing compatibility with Thread-only extensions (in the interest of their users, or for money) and fragmenting the community. Threads then becomes the defacto ActivityPub standard. Maybe some instances stay true to the standard but with extremely reduced communities because now they can’t see what other instances are publishing. So now you have to decide between your ideals and your social network. At best, you’re back to square 0.
It happens in the extend part.
Large corporation will have much more resources, they will implement features and refactoring, which small open source teams do not have capability to implement. They will start pulling users because they support features that other do not.
This also means that they will start getting control.
And then finally they just cut the communication, and split the community. All the way they can claim to be working "for the community"
This is it right here.
If you need a real-world example look at the original web browsers:
NCSA Mosaic (the very first web browser) fully supported what would be later known as HTTP verison .9 . There was universal compatibility because there was only one browser supporting HTTP. Later Netscape Navigator would come on the scene and add functionality that was not supported in Mosaic (like the tag for example), but nothing hugely breaking page views between the two browsers.
Fast forward to Internet Explorer v3, v4 and v5 where MS would not only show all the pages that the prior browsers would, but they EXTENDED by letting HTML still work without following all the same standards. It was easier to write pages for IE than it was to the specification. Then EXTENDED again by MS added ActiveX to web sites meaning now ONLY MS IE could display these pages, and for a time that meant only Windows computers could. This is the Extinguish part.
The "Extend" step gets adopted because its attractive to users.
Here's a non-computer analogy:
Lets say your current car get 25MPG. Now lets says that Shell come out with a gasoline that would let your same car go 40MPG with zero changes. Just buy Shell gas now at nearly the same price as anyone else's and you get significantly more range. Most people would do it. Moreover, Shell buys Honda and starts manufacturing cars designed to work on that same new Shell gas could go 60mpg with even more power! So when you go to buy your next car 5 years later after using the gas, you don't want to turn down 60MPG with more power. That Shell/Honda looks very attractive! All this time all the other gas stations have been going out of business because few people want to pay nearly the same amount for gasoline that only gets a fraction of the range. In the end, ONLY Shell gasoline is being sold, and nearly everyone drives a Shell/Honda to get the most benefit. This is Embrace, Extend, Extinguish.
Took us a while to shake off IE monopoly, only to squander it and now we have chromium (and to lesser extend, WebKit) monopoly. It's not as horrible as the IE monopoly yet, but we're currently in the "extend" stage here with Google forcing standard that benefits them and inconveniences their competitors.
I actually witnessed IE's rise, leaving netscape navigator and opera in dust, and then open source phoenix (later firefox) rising from ashes, steadily taking back user share. Google chrome took a good chunk too and by that time IE was done and desperate enough to give in and use chromium framework.
There was a point in time I thought it's impossible, the close source monstrosity with neverending standards incompatibilities will stay on quick launchers forever but it did not. What a journey.
i.e.: The IE approach. Take an open standard (HTML), then fill in the gaps it's missing with proprietary components (ActiveX), wait until your solutions become entrenched, then start doing evil stuff (implementing HTML slightly wrong so that developers have to do extra work to support compliant browsers).
I was struggling to get all the way there initially, but that makes sense. Thanks for actually taking the time to respond!