Alternatively, imagine a world where the US government passed a “privacy bill of rights” and also required online platforms to be freely interchangeable via open protocols like ActivityPub.
Won’t happen any time soon, and if you ask why, go read !news@beehaw.org for a little bit and come back.
I think among other issues would be the Gmail-ification and iMessage-ification of the fediverse. What I mean by that is open standards like email are dominated today by many people using Gmail accounts as it is popular, “free”, and comes with a ton of features. Then google started “walling off their garden” by adding features that only work between gmail accounts. Similarly, apple also took the open standard SMS and started adding on features only available between other iPhones.
What we might see is some of the coolest features the fediverse has ever seen, but it will come at the cost of most users ignoring or dealing less with “irrelevant” things not on meta ran instances.
Hope we can resist such a change, but that is what I am concerned about.
Even though email is supposedly “open”, and federated, is no longer is really the case. Big services like Gmail are suspicious of non-big-name servers, and often flag email coming from them as spam.
About a year ago I came across an article from a guy who’d been running his own email server since the 90s, and finally gave up. I couldn’t find that article in my quick search, but I did find this:
https://twitter.com/greg_1_anderson/status/1425113874722820100
“I run my own email server. It’s no longer a good idea, because the anti-spam arms race makes delivery from small independent servers very difficult, even when you keep yourself off the block lists, so it’s a continuous struggle. Would switch, but I have too many domains/addresses”
This is very true, I have hosted my own email before and if you are doing it yourself and not going through a big player like google to host it then your stuff sometimes gets treated as suspect by filters. Used to beg people with Gmail accounts to flag my emails as “not spam” whenever it showed up in the spam folder.
Plus, if they do create cool features, why would we not also add them?
Limited developer time.
@CanadaPlus this is referring to far in the future. In the long scale of things, developer time is not so limited. Fedi doesn’t necessarily have a time limit after all, it’s just going to go stronger over time. I don’t see a stopping point.
In the Fediverse you are still 100% under the control of whoever runs the server. Your user accounts can’t move between servers. There is no easy way to export communities and import them on other hosts. On top of that, all the federated features are completely optional and can be switched off.
Fediverse really doesn’t offer any securities beyond what a plain old Web forum does, all the federation aspects depend on everybody playing nice with each other.
At the moment even basic GDPR conformity isn’t given, as there is no way to export all your data from an instance, a deletion request for your data also doesn’t seem to be guaranteed to make it to other instances.
If Facebook builds something with ActivityPub and it gets popular they can play the whole embrace, extend, and extinguish game from start to finish.
If there are some big players (like in email), i think the biggest risk is that the big players would end up only talking to each other.
Similar to email, where a random host is likely to be spamming, that might happen here too. (Although I’m not that familiar with the protocols here)
We have the power over ActivityPub
Who is ‘we’? And who doesn’t say that there’s something on top of activitypub?
Plus, if they do create cool features, why would we not also add them?
Because we don’t have multiple thousands of paid developers.
One of the “powers” of OSS is that the license usually required changes to be fed back upstream.
If Meta were not to do that the authors of Lemmy could ask someone like EFF to take legal proceeding against them.
Well think of the iMessage example for a second, other phone manufactures wanted to extend upon SMS with RCS to enable cross-platform read-receipts, better image quality on messages, and more… and you can use RCS between various android phones, but apple has not yet adopted RCS. Then because of the pre-existing market share of iPhones being so high, if you want read-receipts, high quality image messages, and more you with most of your contacts will either have to convince all of your friends and loved ones to use a third party app or cave and get an iPhone.
The features don’t have to be revolutionary, they just have to find ways to flex their market share with their features. And their market share is almost destine to be huge if they put any meaningful effort or money behind it.
@emi @shipp I think an open standard converted to a walled garden is still better than a garden walled from the beginning.
I can still send emails to GMail accounts.
I can still send SMS to my friend’s iPhone.
I wish everything was fully open, but at least I get to chose my email provider or my SMS app. (Although SMS is completely irrelevant in Europe these days, due to providers still charging money per message.)
That’s an interesting example, but note that in Europe, at least, WhatsApp is king. I only mention it because the walled-garden approach Apple favours isn’t necessarily a guaranteed outcome, and third-party apps can happily become the norm among non-tech people.
What license is it open sourced under? I think AGPL (or one of the GPLs) would be the only one that could sorta force that issue.
The community would have to enforce the rules of the license. It wouldn’t stop them from attempting to commercialize but the code would all have to be 100% free and public. All of it. Free to review, change, copy, etc.
I still see three possible ways for Embrace, Extend, Extinguish.
The first is the Android way: while Android is FLOSS on paper, Google makes sure to cram as much important functionality as possible into their proprietary and closed Play Services blob. I have no intimate knowledge of ActivityPub, but I reckon it would be relatively easy to contribute a tailored modules wrapper under any required GPL licence, and then use that wrapper as a gateway to closed-source extensions.
The second is the Gmail way as described by Emi here: build a portal based on established standards that becomes hugely popular because of its ease of use and feature set, then start sneaking in cool non-standard features that only work inside your walled garden.
The third is the Microsoft Java way: build your closed-source clone from scratch and make it just incompatible enough for non-technical users to think that the original implementation is broken.
I sure hope there’s a large group of servers that refuse to federate with servers run for profit. I didn’t come to be a product and be manipulated with algorithms.
I don’t see anything inherently wrong with servers that try to generate some kind of income (servers don’t pay for themselves after all) but it’s absolutely the right of every server to choose whether or not to federate with them.
I’d take issue with free labour (e.g. unpaid mods) on a profit-making server.
In fact, I hope we sort out a fair and simple method to support servers in a way that makes people feel liket hey are also getting something.
One easy option is a server can have their own emojis like Twitch & Discord. A simple method is for Gold/Silver that goes to whatever server the comment was made to.
Please no. I don’t want this place to be emoji ridden. This is where people go to look for useful information and discussion, not a colour soup comment section.
I worry that through federation Meta will be able to track users of non-meta instances. Then you won’t even know you’re being traced
There’s a difference between generating income naturally through a platform and whatever the hell public companies are trying to do.
For instance sports teams would naturally have their own instance. They can generate more income naturally from their fans that way. Because their fans want to interact with them. They have a product that people want to pay money for.
