In the last few weeks, I have been trying to organise the #Mastodon development work, so our (very small!) resources are used efficiently. This led us to a tentative roadmap for Mastodon 4.3, with a (very optimistic) release date before Christmas.
I was in a discussion today with @TechConnectify and @feditips about reply toxicity here, and I've tried to make all of that into a coherent post.
I want to go to #MastoDev with a version of this, but need to let it sit overnight and probably rewrite it. But I also think it's something anyone interested in non-corporate-controlled social media needs to think about, because a fair number of people go "influencers, fuck 'em" and that's... not how I think this needs to be understood. If you don't care about them, fine, but care about their followers who will follow them away.
Because like it or not, if you want to go against corporate-owned social media, you need these people too.
Is there anyone working on generalizing the Mastodon API, or documentation to follow for apps that are using it for things other then Mastodon? #fedidev#mastodev#mastodonapi#APIdev
That seems like a pretty big change. I've got questions, like:
Why not put them under a "show tags" thing?
On Tumblr they're visible, but they show you the first line or so by default and then you can click to show the rest. Should we have that?
At the moment, when something appears in your Home feed because you follow a hashtag, there is no indication about why. If hashtags are hidden, people are going to be even more confused.
I get that people want to not see spammy hashtags at the end of posts, but to make them totally hidden to the reader seems a bit extreme?
Anyway, if you feel similarly, leave a comment on the Github pull request: https://github.com/mastodon/mastodon/pull/26260 (You might need to make a Github account, but that's pretty easy.)
This is probably too hacky but can't help trying it out. Phanpy's shortcuts settings are saved as (private) notes on my own profile, which will allow export/import in the "cloud".
It's possible to use the API to add notes for other profiles, and they're hidden from (self, currently-logged-in) own profile. I'm using it to store data 🤪
NOT available on the dev site yet, still on my local machine.
this enables full-text search for posts you haven't interacted with, as well as full-text search for accounts, and includes several advanced filtering operators and parser fixes.
#MastoDev and #FediDev: In a few threads and forums there have been tons of chats about fixing the "remote follow" and related issues on the Mastodon/Fediverse WebUI experience.
Plugins are no fix. And what I'm about to share is not a "fix." But maybe a bandaid, while the protocol itself is worked out. What do you think?
It's only 55 lines of Javascript and on any page it asks you your home server ONCE, and then remembers and adapts all links from then on:
Hello fedi. i am trying to solve the "fetch all replies" problem once and for all that makes the fedi feel a lot more desolate and with a lot more reply guys in it than it should be. this is take two, where before i had it triggered by a button, but now i think it should happen on the server-side whenever you expand a post. can anyone help me out figure out how to make this more efficient by only fetching posts that the server doesn't already have? i am not sure what the best strategy would be, and if anyone with experience doing efficient rails and SQL stuff could give me some pointers that would be gr8. the patch is actually extremely simple it just needs a few nice things to make it not DDoS everyone.
I'm just a humble country chicken admin, but I've been thinking a lot about the cost of the fediverse and how we can improve it for all of the instances... in a coop kind of way
The cost of handing all of the media is THE major problem for me atm.
Hi Fediverse admins / devs! I've got a question about instance software.
Up until now, I've been running a fork of glitch-soc that has served me well. Unfortunately, however, I'm starting to run up against some of Mastodon's limitations and the effort to maintain my own fork is just too much. (seriously, fuck Rails and especially Webpacker :blobfoxangry:) I'd like to switch instance software to something with more features out-of-the-box. I'm considering Calckey, but I don't have any experience with it or any other Misskey derivative. The feature set looks great on paper but I'd like to hear from someone who has actually run it (or at least seriously evaluated the software). I'd greatly appreciate any and all input! I'm especially curious about these questions:
The "Fediverse Software Comparison" table has a few mistakes in the Mastodon and glitch-soc columns. They're minor, but it makes me wonder if that table is really trustworthy? The table is based on personal knowledge
Alright here's a draft of adding post titles and slugified URLs to masto. Needs tests and a little tidying, but this is a first small step towards a couple things that i think will be cool to do halfway between microblogging and forums :) https://github.com/NeuromatchAcademy/mastodon/pull/35
If anyone who knows rails and redux better than me wants to laugh at my code (but secretly i will be learning) then feel free to take a whap at it
(edit: yes i know other fedi things and in fact many normal websites already do this) #MastoDev
For #FediFetcher I'd like to use the ActivityPub API, rather than the Mastodon API to pull replies.
For example: At the moment I'm using the api/v1/timelines/home endpoint to get all statuses in my timeline. I then attempt to get replies from the remote servers.
But the uri that mastodon returns in that response is to the Mastodon API endpoint. This is annoying, because obviously Mastodon != Fediverse, and I'd love to just go to ActivityPub, rather than implement each software separately.
Is it possible to somehow get ActivityPub endpoints for those statuses?
On the Mastodon Public Roadmap this feature isn't even in the "Planned" section: https://joinmastodon.org/roadmap 😐 (MAS-37 Restrict who can reply to a post)
If you too think this very basic safety feature, one of the most thumbsed-up on the Mastodon issue list, should be a higher priority, go to the Github issue and add your 👍 and a comment to bump it into the devs' awareness: https://github.com/mastodon/mastodon/issues/8565
I haven't taken a look yet, but if we implemented a hashtag syntax for nonstandard characters to be able to use things like DOIs and ISBNs and etc. like this:
the current state of play is that Mastodon has a hardcoded list of supported language codes, sending an unsupported one will result in it being silently ignored and replaced with the server default, and API clients have to just kinda guess what's available, so this is a nice upgrade. would love to see other Mastodon-compatible APIs implement this method too.
Due to a pretty serious bug on Phanpy; on Mobile Safari where alert and confirm dialogs will stop working after going 'back' (https://github.com/cheeaun/phanpy/issues/178), I took this chance to rethink the confirmation dialog UX, in this case, for boosting/reposting posts.
Had a shower thought / question about #ActivityPub. Would it be possible to develop a federated server without a media cache? I believe that it should be as long as the server is configured to proxy media requests back to the original source. This would, of course, greatly increase latency and decrease reliability (media is lost if the remote instance goes poof). But in exchange, it dramatically reduces the storage space needed to run an instance. That would reduce the cost of operating an instance on cheap VPS services. I can also think of a couple use cases around the current push to simplify instance hosting.
I think one of the most powerful changes we could make to content discovery on Mastodon is by showing hashtagged posts to everyone following that tag.
Right now, you only see tagged posts from your federated feed, which is (I think): posts and boosts from your follows, from anyone on your server, and from anyone followed by anyone on your server
If you’re on a big server then yes, that is a lot of posts, but it’s no way near all of them!