How can we Google for Lemmy content instead of Reddit?
I am sorry the question is confusing.
But some Google searches give much better results if you add "reddit" to the end of your query. This ends up generating a lot of traffic for Reddit.
Anyone found a way to search something but hint Google to look at Lemmy?
Try kagi search! It actually has a “lense” or search option that lets you directly search federated services like lemmy: https://help.kagi.com/kagi/features/lenses.html
This is awesome! Bookmarked
You can add "site:[whatever.tld]" to a Google search to restrict results to the specified domain. But as mentioned elsewhere many instances block indexing.
Conversely, you can add -site:Reddit.com and exclude Reddit results altogether.
That way you can search one instance but not the fediverse
Maybe not helpful, but Kagi search includes an option to search forums, which includes Lemmy. They have or had a dedicated Lemmy search, but I don't see it on my end right now.
This is definitely helpful. I hadn't noticed it. Kagi keeps on giving. Thank you!
It's still there for me, called Fediverse Forums. It's a lens you can disable or reorder so maybe you did and forgot about it? It's also different to Forums lens which doesn't seem to include Fediverse for some reason.
Why are yall paying for search
Better than the search paying to use you.
Add this at the end of your search query to omit results from Reddit:
-site:https://www.reddit.com/
To specifically include results from Lemmy, you can just add a plus sign and replace the site name:
+site:https://lemmy.world/
Believe it or not Lemmy.world is not the only instance.
Yeah I'm new to Lemmy and still getting used to the whole instance thing. Apparently you can use "related:" to include results that are similar. In the below example, I did a Google search for "ukraine related:lemmy -site:reddit.com".
I'm guessing that search would turn up results for any instance LW is federated with? If not that would be disappointing!
Just as a FYI:
You don’t need the + to include things on Google. Anything you type that isn’t prefaced with a minus symbol already means “include this”. Also, using quotes means you want an exact match.
Additionally, you could shorten to this:
Search keywords here -site:Reddit.com site:lemmy.world “exact match keywords, if you want”
And just for funsies—you can use related: as an operator to find sites that are similar to something.
You don’t need the + to include things on Google.
Yeah I don't know why I added that. Kinda dumb on my part.
And just for funsies—you can use related: as an operator to find sites that are similar to something.
I think this is what OP was really looking for. I did a Google search for "ukraine related:lemmy -site:reddit.com" and got this, which seems to be the kind of results OP is looking for:
At least it's not pulling up mostly Motorhead-related content like it used to...
Reddit has a decade long corpus of valuable knowledge from millions of individuals... Lemmy just doesn't have that scale or earned trust yet.
Yeah I don't browse Reddit anymore, but I still Google it for BIFL products and such. Lemmy just doesn't have the content for those use cases.
You should definitely post in a relevant community to help build the proverbial field of dreams!
SearXNG also shows lemmy results in the social tab.
Why not just open your Lemmy app of choice and click search?
I currently don't do that (can not replace the treasure trove in reddit yet) but you can edit default search engine url and add something like
site:lemmy.world/ site:lemmy.ml/ site:lemm.ee/ site:sh.itjust.works/ site:lemmy.dbzer0.com/ site:lemmy.ca/ site:programming.dev/ site:lemmy.blahaj.zone/ site:discuss.tchncs.de/ site:spouli.xyz/
and make a seperate search engine shortcut for this. But as others have said, google is not great for this.
If you use searxng, then you can also include instance searches in default results . There is a seperate social media page(just like you have image or video tab), enable lemmy and mastodon stuff and use that.
Don't use Google, use Startpage.com instead, and add "lemmy" to your search terms.
There was a Lemmy search engine but I don't think it is up any more.
Stract has a Fediverse "optic" but it doesn't appear to be working.
Probably if you tell Google to search site:lemmy.world that is likely to be good enough.
That makes me wonder if the fediverse is disadvantaged by SEO. Each instance is likely "just" a domain to search engines.
Just change "reddit" for the Lemmy instance of your choice and it is functionally the same.
-site:reddit.com should exclude results from that site.
Then you will find only the shitty articles
I want to whitelist Lemmy, not blacklist Reddit
I don't know if the OR operator + grouping still works, but you could try something like (site:fedia.org OR site:feddit.uk) or whatever. I know at some point some operators and grouping symbol usage got dropped by google. I'm not sure about other search engines.
a lot of lemmy instances block google indexing purposefully. you would need to search using the instance search capabilities.
Has had it for a while now, yeah.
One day I will be able to afford a subscription based search engine..
Has had it for a while and can easily be toggled between normal search and lemmy search, forums, academic, pdfs, etc. It's glorious.
Hopefully I'm not opening a can of worms by asking this, but, why?
To be more precise, why not let any search engine index? It seems like it'll grow Lemmy if people can use its data to search through.
This comment is licensed under CC BY-NC-SA 4.0
Because growth through opening up for capital interests is going to end badly. See Internet since 20 years
That seems rather counterproductive imo. The only way to get useful results on any search engine is to input "reddit" at the end of your search ime. So it seems like it's limiting the discovery and knowledge base of things here by not allowing for the same thing for Lemmy. Idk how to word that correctly.
More so the only way to get results that aren't ad-filled webpages that are paid to be top Google results is to search for reddit posts. Unfortunately a lot of the newer reddit stuff seems to be regurgitated bot accounts reposting stuff, so most useful stuff on Reddit is from before the enshittification.
Seems like the correct way for them to index the site would be to create their own instance that federates with the rest and then index the content on the server. There is no need to scan each individual instance when there is a publishing protocol that handles the content stream.