Woweee. Why am I only finding out about this now?
https://wedistribute.org/2024/04/emissary-social-web/
Thanks, WeDistribute, for a better write-up than I could have done. I can't wait to deliver more!
Woweee. Why am I only finding out about this now?
https://wedistribute.org/2024/04/emissary-social-web/
Thanks, WeDistribute, for a better write-up than I could have done. I can't wait to deliver more!
@jenniferplusplus @hrefna I love this thread and the fact that you’re all thinking about these things. Not to muddy the waters, but...
This sounds a bit like FEP-5624 in that the original poster may want to to choose which parts of a thread to sync with others.
I think the original poster could be considered the primary source of truth, which might also minimize the amount of “gossip protocol” required to sync this distributed db.
@madeindex Well, I only claim to be a UX designer in my posts.. my black turtleneck and infinity scarf still haven’t arrived from the online hemp shop, so it’s not official yet.
@madeindex Yeah, in the follow up discussions about this article, I heard that a fork is being discussed (dunno by whom) for *political* reasons (don’t know more details on the politics)
I’d love to learn more about the specific politics that would make someone say that Mastodon wasn’t the right fit for them, but on the surface it seems like a bad move.
That effort would certainly be better spent on making Mastodon itself better, or helping out smaller projects instead.
@madeindex Absolutely. This should be standard/default behavior on every app.
@[email protected] @[email protected] Yes! There are several pieces. FFmpeg manipulations are in https://github.com/benpate/mediaserver
#FFmpeg is amazing but the API is very terse, so it takes a little getting used to. Media Server wraps this into a (slightly) nicer API, and manages cached file uploads, too.
The app I'm building is called #Emissary - https://github.com/EmissarySocial/emissary -- it uses #htmx and the media server above to manipulate images and audio files (video coming soon)
I'm happy to DM or chat more if I can help you out more
Yes. Stack Overflow isn't magic. I think they succeeded because they focused heavily on SEO, which brought both ask-ers and answer-ers to the site.
Personally, the "accepted answer" is the killer feature. Dunno if NodeBB, Kbin, or others already support this.
There's potential in SO's gamification aspects, too. I'd love to let third-party sites to award badges or "endorsements" and display them on my profile page. This could work in all kinds of trust/credibility situations.
I should really do more research before I say anything too definitive.
I guess I'm too much of an engineer to get why you'd need a whole new fork for political reasons.
Do you have a link to the new project, and their reasons for discussing a fork? If someone wants Mastodon + extras, I'm curious about what would keep them from just contributing back to the core.
@utzer @dansup Yeah. We could always just build a new network with special features that would require an account on a Q&A -style server.
But an emerging feature of "fedi" is that you can take your identity everywhere. I'd really like to support this somehow, too.
Perhaps it just means using main Mastodon (or whatever) account as a "universal inbox" for notifications. But then we link you back to a site with SSO, so you can interact more richly there.
Dunno.. TBD.
I'll take any/all advice.
@dansup Yes. After Bandcamp.
I had an online discussion about a Federated-Yelp that raised some interesting points that might apply to SO as well -- How do you “federate” the features that DON’T fit into the standard social media formulae? Things like “accepted answers” might only work on a centralized server.
Also, this might fall under the threaded discussion WG. I’d love to talk in more detail if you’re ever interested.
@jjude The specific features to record audio would certainly need to be a bunch of custom JS, not htmx. But all of the stuff around it that makes it a viable app? Yes. Htmx is fantastic for that.
The key is putting the right features in the right layer of the app. I just integrated FFmpeg into the back-end of an htmx-powered app, to transcode various kinds of audio on the fly. It works great. 10/10 would recommend.
@maybeanerd Great point. And I have to say I’m not totally up to speed with the state of whatever Mastodon forks that are already out there.
I guess this is my question: if there are already thriving forks of Mastodon, what benefit would another one provide?
The most likely outcome I see is a bunch of time spent with little to show for it.
I’d rather see that energy go into improving the core of existing apps out there already (Mastodon or otherwise)
Well written article with many good reasons to NOT fork Mastodon.
https://wedistribute.org/2024/05/forking-mastodon/
My favorite part was about a previously attempted fork called the Florence project. According to the article, they were apparently better at having meetings than producing actual code. I laughed. I cried. I relived so many years of software development trauma.
@Sarahp Absolutely, will do
I have a few more “T’s” to dot, and “I’s” to cross, then I’ll probably open up a limited beta test. I’d *love* to chat once there’s something more tangible than my silly videos.
@rizzi This looks absolutely gorgeous. Kudos!
I really struggled with this, because there’s so much that’s NOT covered in the specs.
I made this list of resources that have helped me. Hopefully it helps you: https://emissary.dev/fediverse-resources
@paytonrules I hope I don’t sound wishy-washy here, but the idea keeps evolving with every person who adds their comments. I’d call it “stone soup development” if I ever wrote a book.
Fedi could AT LEAST provide identity and sign-in.. and maybe allow people to post/boost reviews from their profiles.
The biggest value might be in using the social graph to determine “trustworthy” reviews and filter out the bots.
@it Despite being absolutely crushed 2:1 in the voting, the “no” faction has made several convincing points... this among them. “Not polluting my timeline” will be an important factor to keep in mind.
I wouldn’t want my timeline filled with random reviews either, but I might want to use my friends list to narrow down reviews I can trust vs garbage bot posts.
And I might want to rally my friends to spend money at GoodCo instead of NaziCo, if they both carry the same kinds of stuff.
@steve Yes. That would be reasonable of them. It would certainly be a good model for something on the Fediverse.
I’m kicking around the idea of a “seven degrees of separation” algorithm, where posts/opinions/reviews have a 1x multiplier for people I know/trust/like, 0.5x for second degree “friends of friends”, 0.25x for third degree connections, and on into oblivion.
No idea of it would work, or if we even have the data to do it, but the math would be fun.
@steve This is a really interesting take. Thank you!
It seems like the issues here are: small group of friends => not enough reviews, vs. large group of unknowns => potential abuse by bots.
Could the answer be found in threading the needle between these two?