Yes, with a major caveat. An instance will search only communities that at least one user on the instance is subscribed to and only as far back as the time the first user on the instance subscribed to the community.
Lemmynet’s design structure has some weird choices in it, motivated either by laziness or to keep garage servers from being overwhelmed, and that’s the biggest and weirdest one. I’d like to see federation = full and complete synchonization from server launch to present, but I doubt the motivation is there to implement it. Maybe things will be different when kbin eventually surpasses Lemmy
Interesting, can you explain a little more? I’m very curious.
If that was mandatory you’d kill of smaller/personal servers, which is bad.
Should be a host option to decide what to sync.
Oh and you’d also need to solve an issue with a new instance joining and completely flooding everyone else with requests for past content.
It’s all solvable but the reasons to do it this way aren’t just random.
Oh and you also face issue with duplicate content (between instances) which search engines generally don’t like.
Ideally there would be native integration for search engines where they’d just connect with the fediverse like another instance and have special handling for the content. They could either display the content in their own UI or allow the user to pick their default instance that’d open the given content
I suspect that’d be the best way but we won’t see it unless federated apps become mainstream.
deleted by creator
That’s interesting that lemmy doesn’t generate canonicals. I would have thought that the original instance something is posted on would set the canonical, and other instances can point back to that - it really seems like this sort of problem is exactly what canonicals are made for. Does anyone know if there’s a reason for not using them (other than dev time, which is 100% a good reason)?
Thank you for this. I also use Kagi so im totally gonna set the lens up.
deleted by creator
What’s Kagi?
It’s a subscription-based search engine. From what I’ve seen, the results seem pretty good compared to everything else at this point, but IDK if I’d want to pay monthly for search (though honestly I’m seriously considering it at this point).
As someone who’s been a paid user for a few months now, it is totally worth it. It has its moments where I need to try another engine, but the vast majority of the time it’s way better. The ability to rank sites is a game changer, and as a bonus I can block Pinterest and the usual fluff from results and never see them again.
I bet it still respects search modifiers? Both DDG and Google frequently ignore mine.
Oh snap, + 1 for Kagi usage, I’ll try your implementation
I’d say the short answer is no and the long answer is yes.
Searching across instances is difficult, for the reason RandomBit mentioned in this thread. But you don’t go to reddit and expect your search results to include Hacker News, Twitter, etc. When you search on reddit, it also only searches the local instance, it just is that there is only one instance. So the search is exactly the same as reddit.
With that said, there is probably room for a service that provides cross-instance search by subscribing and indexing communities like a crawler, rather than relying on users to create the federation.
FediSearch I guess is similar to your idea, though I think the goal would be to make a new and open search index specifically containing fediverse websites instead of just using Google. I also feel like the formatting should be more like Lemmy, with the particular post title and short description showing instead of the generic search UI.
The idea of a fediverse search is really cool though. If things like news and academic papers ever got their own fediverse-connected service, I could see a FediSearch being a great alternative to the AI sludge of Google.
Not yet, I don’t think
I think the different domains and levels of federation makes it less convenient. Personally, I think the Fediverse should collaborate on a shared wiki for community knowledge and whatnot.
Any web search seaches Lemmy as well. Just seach for your user ID or display name to find your own content.
I guess what I want to learn is that if I were to type into Google “what are the best iPhone games Reddit” I would get a bunch of Reddit threads does the same thing work if I were to end the question and “lemmy”
That would be the question. You can certainly do a site search. At least duckduckgo can do that. It may not be so easy to just search lemmy, the threadiverse, or the fediverse for example. Do not know.
All search engines do this. You can append or prepend site:<domain name>.<tld> and it will filter by site.
They even do wildcards, so to search any lemmy instance with “lemmy” in the domain name, you could do site:lemmy.*
To include others, you can us an OR operator.
So to include kbin.social you could use the following filter: site:lemmy.* OR site:kbin.social
Unfortunately, this won’t work for instances that don’t have lemmy in the name.
Probably if the instances have lemmy in their domains like mine, or lemmy.ml, lemmy.ca, or in their name (mine is “pe1uca’s lemmy” so if the domain was different maybe search engines could also work with that), but not for ones like, lemmit.online, sh.itjust.works which don’t have lemmy neither in their names nor domain.
It’d be like searching for content posted in sites which use wordpress.Give it a try and let us know.
Kbin is… using the search button.
Maybe using brave search or something?
I think using Lemmy’s own search engine provides all the results that you need and the stuff you’re looking for in all instances rather than using a different engine in my experience.
Whoever is in charge of discuss.tchncs.de has converted the community search into a lemmy content search
oh, so that isn’t how it is supposed to be? I was wondering how to search for communities, this is the only Lemmy instance I have ever used, maybe I should pick a different one?
I don’t know if you need to pick a different one per se. It’s all likely a work in progress. I did notice the community subscribe buttons are also often screwed up and are plain text instead of a clickable button. Not sure what that’s all about. I’ve just been enjoying version 0.18 and the owner has the extra filters for top of the last 12, 6, and 1 hour
