What does this mean to someone unfamiliar with these acronyms?
Atproto (protocol powering bluesky) will be further developed and standardized by the ieft (Internet engineering task Force), a group separate from the for-profit bluesky PBC
Neat! So just more open development?
Yep, and it’s by a group which has much less incentive to fuck the protocol up.
Oh that’s so amazing! I’ve been smitten with AT Protocol since I learned about it from their paper. I have such high hopes it gets more widespread.
What’s good about it and what’s it good at?
The thing that I love about it is that you can host your own account. So if Bluesky decides that they are huge fans of fascism, you can take your account and move to a competitor, Redsky, and not lose your posts, messages, follows, etc (assuming those people also move to the new platform)
So, your account can be the same between any number of platforms, you just have to let the platform add it to their list so their crawlers can show your activity.
And, like Lemmy, you can host your own “node” (I forget what they call it. A box that can whitelist, crawl, and display accounts that people want to be visible there, similar to an instance) but you can also just host your “account” and you can bring it to whatever platform you want and people can be confident they’ve found the same person.
Projects like https://a.roomy.space/ are also super interesting. It’s sort of like how Lemmy uses and reformats the content you might see on Mastodon into a traditional forum, except Roomy can use the AT Protocol to format it into a sort of Discord concept. It’s public, but they also are working on private, self-hosted ‘rooms’ as well. There are also other projects that reportedly have managed E2E encryption for private messaging. (Edit: on this topic, ATP is very pointed at public content. Any encrypted messaging solution isn’t likely using ATP for the messaging, just for your web identity. The major thing here is having a consistent presence and login that you and your friends can follow to various platforms without issue and can’t be controlled by another entity).
I’m super not a technical nerd, so I’d have to go reread documents to give you any specifics about it as a whole. And even then, I can only really understand them to the extent that I’m not actually a developer on the protocol, so I don’t have a first-hand understanding of how it works, but the concept of it and what it seems to enable is just really exciting to me.
this portable identity feature has existed for a number of years in hubzilla, its implemented as its own ‘zot’ protocol, offering nomadic identity as well as account cloning(redundancy).
I love your enthusiasm and that’s because I probably understood about half at best.
Sounds like a very powerful idea. I come from a time before email was free, and I can imagine the hassle today if I didn’t have access to say my Gmail that’s been going for+20 years now. So with this simple example of self hosting identify I can see massive upsides.
I don’t understand fedivwrse at all but I can see how this would take it to the next level. Ietf has a lot of weight so it’s another plus to reduce the weight of IEEE and US dependency.
Exactly! On the a.roomy.place there’s a good, non-technical breakdown on what makes the concept good and what flaws it has, but the core of it is the concept of owning your own identity. The idea of “login with Google/Facebook” significantly reduces internet freedom, this gives you a way to “login as yourself”, beyond the ownership of a company. That’s the big boon here. With the IETF lending some credence to it now, it’s a good sign that self-hosted identity for your public presence will be adopted into the mainstream and a less locked-down internet is on the horizon.
/over-enthusiastic optimism
I really like how it splits data hosting an applications, the only public facing part of the network you need to host is a pds, you can use clients that work entirely locally, such as appviewlite, or reddwarf.app.
Is this one of those groups that’s going to get stacked with large corporations who proceed to sabotage the protocol over time to make it less useful and competitive against their own products and services?
Like activitypub?
What appreciable differences are there being in a working group under the IETF vs W3C vs WHATWG?
Unsure about the IETF and WHATWG, but if at all, they’ll be better than the W3C. The W3C was, and still is, a group co-opted by GAFAM to essentially make the web as hard as possible to implement so that only big corpo can “do it correctly”, and they brought us wondrful features such as literally DRM in the HTML Standard.
To this day the W3C is one of the big reason the internet doesn’t progress.
ActivityPub, aka the Fediverse, aka the open social web, aka what powers Piefed, is created, built by and built into the web, run by the W3C.
So while I agree that DRM sucks being built into the web, I disagree W3C is controlled by GAFAM. With DRM the W3C’s point was that everyone is doing it anyway at least we can have a standard way to doing it, making it easier for ends users to not have to jump through some new hoop and guaranteeing even that on alternative browsers, software or hardware aren’t excluded.
Big Tech do plenty of awful things we can give them direct credit for. W3C isn’t really one of them.
