Is Lemmy searchable from search engines such as Google?
Is Lemmy searchable from search engines such as Google?
Part of the utility of Reddit to me was the ability to append "Reddit" to searches and be served relevant threads to my search query since Reddit was one of the last places to get information from real people. This has proven to be invaluable in troubleshooting. Is this possible with Lemmy or the greater Fediverse? As we pick up more users and content, it's my hope I could do the same with Lemmy.
Or is it more like Discord where information isn't indexed by search engines? One thing I've always despised about Discord is the inability to find information if you don't belong to the communities that produce it and I'm hoping that's not the case here.
Hopefully this isn't too silly of a question I'm still trying to wrap my head around how all this works.
Until it is easily accessed you can use these to find posts:
https://fedi-search.com/
https://www.search-lemmy.com/Hope this helps.
A little hack, add
(intext:"modlog" & "instances" & "docs" & "code" & "join lemmy")to your Google search, it'll search across instances.yes, there is nothing stopping google and ive already tested my own instance, content shows up fine.
i expect the software likely needs an "seo pass" so that it can be optimized for search engines. this will come in time.
Form a technical standpoint lemmy, kbin and co are just as searchable as reddit. Google and co can index them, because everything can be accessed and interpreted using normal web technologies (http, HTML, CSS, ...) without needing a login or an invitation to a server and everything is discoverable via links. Perfect for search engines. How good the search results on Google and co are depends on how often and how thorough the search sites will crawl the fediverse. This in turn depends on how important the fediverse sites are deemed by the search engines. This depends on how often they are linked from from the rest of the web and on the amount and qualiry of content and tons of other factors.
I would imagine once everything is federated, we’ll be getting a lot of duplicate results on Google. “How do I do X?” will return the same post across all federated instances unless Google figures out that it’s all the same post and only retrieves a single copy of it.
There's already a lot of duplicate content on the web. Search engines already have a system to rank duplicate results and not respond with only identical answers. I'm pretty sure this won't be a big problem. And if it turns out to be a problem, they'll adjust their algorithms 🙂
This morning I got a Google alert for my username (yes, I set a Google alert for my username because I'm full of myself) and it sent me to my Lemmy profile. I'd interpret that as my home instance being indexed?
How do you do this?
Set a Google alert? Here: https://www.google.com/alerts :)
Yes but it will take at least a few years of organic use before you can do what we do with Reddit and add ‘Lemmy’ at the end of your search query.
A cool google hack; you can use “site:<domain> <search query>” on google to look for something on a specific website.
I don’t think it’s searchable yet. But there’s solutions like Lemmy Search which might make it a reality.
This is amazing, thanks!
You would have gotten your answer by asking google that exact same question - yes.
But the way federation works is terrible for search engine optimization as it fragments the userbase to a million smaller sites - instead of whatever hundreds of thousands of overall lemmy users there is, Google will only see the few top ones as being anything relevant. For the kind of search where you add "Lemmy" at the end like you can with reddit, it is (currently) probably better to use "site:lemmy.world" as it is by far the largest one.