This makes me wonder if you could make a super frictionless path from a thread on lemmy (or similar fediverse software) to some form of a wiki page that presented the same information but in a more natural form better suited to a longer term repository of knowledge rather than an evolving conversation. About sidebars and pinned threads for subreddits/lemmy communities are an extremely important part of the structure of a reddit-like, but why limit our vision of a reddit-like to only being able to create those two narrow types of persistent, documentation style information?
In practice this obviously can just be a lemmy community linking to ibis wiki pages maintained by members of that lemmy community, but I wonder if there isn't an exciting space here to explore what that process could look like if the integration was way tighter and more direct.
I think it is worth considering the argument for splitting a reddit-like from an associated wiki in the first place, why not have them just be two different types of posts, with different associated rules of editing, and two different home pages one that looks like a reddit-like and one that looks like a wiki? Same accounts, same website, same markdown conventions and text/media formatting.
Assuming a bit of careful edit permission handling for a lemmy communities associated wiki, wouldn't the end result be WAY more powerful of a community resource than a lemmy community and wiki taped together?
What if a lemmy post could be turned into a wiki post (on that same lemmy/wiki instance) with a click of button, only requiring a small amount of tweaking to restructure the information in a wiki fashion? The wiki post would of course reference the original thread it was made from and only certain accounts on a lemmy community would have permission to do this.
This capability would give a small lemmy community the ability to warp ahead of clunky, obtuse discord communities in constructing genuinely useful repositories of expert information with far less effort or friction.
The moderators of Wikipedia are largely a small, insular community that don't always care about accuracy when it comes from anyone outside their clique.
Looks like a federated wiki, which is great. And not a Wikipedia alternative. What makes wikipedia wikipedia is not the tech. Social and knowledge problems can't be solved with tech ;)
As much as Wikipedia has issues, as the ibis announcement states, it also works in many places. And federating it won't help with the issues of bad moderation, quite the contrary. And as much as I like nutomic (thanks for syncthing-android ;) ), I don't hear many good things about the lemmy moderation story. So I have my doubts. Lets hope I am wrong. Plus anyway, federated wikis is a great thing to have, ignoring the whole Wikipedia aspect.
Actually its too complicated to change the title in Ibis due to federation. Plus I would have to rewrite the whole intro. But I will keep it in mind for the future.
Honest question out of interest: Are you doing moderation on lemmy?
I just remember reading about admins/mods complaining about the lack of tooling, sometimes plain functionality (removal of certain things) for effective moderation. I am not doing any myself so that's very 3rd-party-ish knowledge (if you even want to call it that).
This is super exciting. I think one of the things a lot of people are missing here is the potential for wikis to augment existing fediverse communities. Reddit's killer feature has always been the massive treasure trove of information for hobbyists and niche interests. There is huge potential in the fediverse to take advantage of that sort of natural collaborative knowledge building process.
What are the articles written in? Wiki lang (or whatever it's called) is horrendous, IMO. Hopefully this is markdown? I couldn't find after a quick browse through codebase and I don't think it's mentioned in the blog post.
I'm not sure I see the benefit of this. The point that Wikipedia might eventually become corrupted is made moot by the permissive licensing of the information there. The main challenge of the Wiki format is with fact checking and ensuring quality, which is only made more complicated by having a federated platform.
ActivityPub is great for creating the social web. The added benefit of ActivityPub for non-social services is not obvious to me at all.
That said, it's a cool proof of concept, and I'm sure it can be useful for certain types of federated content management - I just don't see how it could ever make sense as a Wikipedia alternative.
Then again, why would a fan page want to open for contributions from outside of that fan page? Why would the Star Wars wiki federate edits with the Startrek wiki? On which page of the wiki would this make sense?
I guess you don't have to get it. I just mentioned that site as an example because it is kind of garbage, but it's useful for fans.
And the federation between fandoms would be like how different articles are connected on Wikipedia. For example, there are actors that had roles both in the Star Trek universe and the Star Wars universe.
Lots of those fan wikis just link to other websites. It's entirely possible to do that.
If you're on a Star Trek wiki, why would you want to go to a page that's almost exclusively talking about Star Wars information in relation to some actor other?
I was pointing out that the two fandoms are actually connected by some actors because the person I was responding to seemed to be unaware.
Maybe no one does this, but I've looked at character pages and clicked on actors' pages if I liked them and wanted to see what other work they've done.
The main reason people use Fandom in the first place is the free hosting. Whether you use MediaWiki or any other wiki software, paying for the server resources to host your own instance and taking the time to manage it is still a tall hurdle for many communities. There already are plenty of MediaWiki instances for specific interests that aren't affected by Fandom's problems.
Even so, federation tends to foster a culture of more self-hosting and less centralization, encouraging more people who have the means to host to do so, though I'm not sure how applicable that effect would be to wikis.
This is a cool idea, but I highly encourage you to target mobile first. Reference works will get a LOT of mobile traffic. More than 80% of Wikipedia’s traffic is mobile.
Man I just don't get it. There's a kind of wilful ignorance here or something? It's jarring. All due respect for what's been made but this attitude... I'm not offended or have disdain, just dumbfounded at the messaging.
I kind of feel like I’m walking into a construction site where someone didn’t some amazingly complex joinery, and I trip over the cuttings on the shop floor.
Then I say “you might want to clear a path here,” and I’m told by the woodworker that they never learned to use a dustpan, but it’s over there. “Go sweep for me.”
Reminds me of a place I used to work at. Small place; 10 people. I started as a sysadmin but later started programming. They encouraged me; "yes we suck at this we need help!" so I kept going. But as the work became more involved and I needed a bit of co-operation from their side, it was torture. They didn't "suck" at it, they just didn't respect or bother themselves with that kind of work.
Opensource is not some public service you just request something and the developers create it for you. If anything attitude like yours are what's burning out opensource developers.
It's their time, work and effort and I am glad they're upfront about it rather than pacify everyone's request beyond what they already do.
This is not about software licensing nor the spirit of FOSS.
There's some inconsistent messaging that's genuinely confusing me. I've shared an anecdote below (from a time when I was developing open source software) in the interest of generating discussion to clear it up for me and perhaps others, too. I don't mean to imply I know what is happening right here.
If you paste your suggested CSS improvements with example code into a new GitHub issue, that may lead to another git-savy contributor creating a pull request.
Rather than starting from scratch, would it make more sense to make an ActivityPub plugin for the open-source MediaWiki software Wikipedia runs on? MediaWiki already has some “interwiki” functionality that such a plugin could expand on, and you’d have the advantage of being able to fork content from WP and other MW projects without having to re-format it. Plus you’d be able to leverage other MW plugins—Semantic MediaWiki in particular could add a lot of useful functionality to federated wikis, like articles that could query and aggregate information from other federated articles rather than just linking to the text.
Mediawiki is an extremely complicated project with 1.2 million lines of PHP. For me it was much easier to implement this project with technology Im already familiar with. But of someone wants to create a Mediawiki plugin I would be happy to see that.
The link is virtually unreadable, it formats really strangely on mobile. The text is in a 1cm wide column on the right side, allowing only ~3 letters per row.
Very interesting concept. Since the original wikipedia is in most parts published under licenses that permit copying & adaptations, are you planning to integrate their articles as a snapshot for the sake of having a solid foundation?