Due to the nature of fediverse, you can’t just do a “site:lemmy” search as you’d do with reddit (site:reddit.com). I was searching for ways to do it on my firefox browsers and I came to a solution that searches 21 different instances simultaneously, which I consider to be a big part of lemmy (I also include kbin instances). I am using firefox with DDGO (which has a 500-character limit per search query).
There’s gonna be a text you will have to copy, I’ll just write it once here hidden inside the spoiler:
"URL string"
https://duckduckgo.com/%s (site:lemmy.world |site:kbin.social |site:lemm.ee |site:programming.dev |site:lemmy.ml |site:sh.itjust.works |site:feddit.de |site:lemmy.ca |site:beehaw.org |site:lemmy.dbzer0.com |site:lemmy.blahaj.zone |site:discuss.tchncs.de |site:sopuli.xyz |site:lemmy.one |site:feddit.uk |site:midwest.social |site:infosec.pub |site:slrpnk.net |site:mander.xyz |site:ttrpg.network |site:lemdro.id)
Replace it when I ask you to write the “URL string”.
There are different processes, depending on the type of device/OS (computer, mobile/android etc.). I will list some of them and you can pick the one you want:
-
Firefox Android: Go to search settings->manage alternative search engines->add search engine. Pick a name you want (eg.
Lemmy
) and on the search string URL paste the “URL string” I’ve written above. Now next time you want to search Lemmy, just select “Lemmy” as your search engine. -
Firefox Computer V1: Create a bookmark with a name you want (eg. Lemmy search). In the URL field paste the “URL string” I’ve written above. In the keyword field add a keyword to use as a prefix to your searches (eg.
@l
). Now to search Lemmy, you can type@l "random search"
in the address bar. -
Firefox Computer V2: Open a new tab and type
about:config
in the address bar. In the search box type:browser.urlbar.update2.engineAliasRefresh
Click on the little+
symbol on the right. Go to firefox Settings->Search or enter this in the address bar:about:preferences#search
In the “Search Shortcuts” section you should notice a new “Add” button. Just like on android, press “Add”, insert a name (eg.Lemmy
), add the “URL string” from above and once you’re done you can add a keyword (eg.@lemmy
). If you want, you can now go back and disablebrowser.urlbar.update2.engineAliasRefresh
(press the trash can icon) (I don’t know if this will crash anything).
I really wanted to search lemmy, but not being able to do a simple site:lemmy detered me from doing so.
So there you have it! The added text is 412 characters which leaves 88 characters for your searches. You can change the url to add/remove instances you want. If anyone can improve it please (like shrinking the url or if you manage to search all instances) it will be appreciated.
Credits:
-
The idea came from this site (which I found on lemmy): https://fedi-search.com/ I asked it to search lemmy on ddgo and gave me a similar string to the one I have above, only limited to fewer instances.
-
I found the biggest instances I wanted to include in the search using the csv from this github page: https://github.com/maltfield/awesome-lemmy-instances/blob/main/awesome-lemmy-instances.csv
-
Lastly, I found the computer process from 2 answers on this stackoverflow question: https://superuser.com/questions/7327/how-to-add-a-custom-search-engine-to-firefox
There’s https://fedi-search.com/ and https://www.search-lemmy.com/ as well. I’m not sure if that latter one is still active though, it certainly appears to be down at the moment.
Search Lemmy is on hold, they had performance issues with accessing lemmy and are waiting for a fix
Yep I checked them both (in the credits). What I did is kinda like the fedi-search, but with more/different instances.
I’d like to check search-lemmy, but it’s down for me as well…
fedi-search has an error message for me.
Instance has been ratelimited
I had the same error when using Whoogle. I didn’t get an error with other search engines.
I’ve been having issues with search-lemmy for a while now, it doesn’t really show anything recent
I have to think, on some BBS forum somewhere, before search engines took off, someone made a script out of
curl
,grep
, and a list of popular domains to search this new thing called “the Internet”.This giant aggregation string you’ve created isn’t something that should need to be done manually, this is something search engines should just do for us. Instead they’re focused on ads and listicles…
Maybe duckduckgo will eventually add a search for lemmy like they did for reddit… Shrug
I don’t think it should be necessary to tell a search engine where the most relevant results are located. That’s literally the only point of a search engine. I want to enter a search term, and if the best results are on reddit, I want to see those, or if it’s on Lemmy I want to see that, or if the best result is somewhere else I don’t know about, THAT’S what I want to see. The fact that we have to manually tell search engines where to search is completely backwards.
Not ranting at you, you’ve done good work. Just disappointed at the current state of “search” engines.
Ahh now I get what you mean, ok ok :)
Edit: My 100th comment on the platform!🎉
Awesome. I submitted this as a Bang on DuckDuckGo. I don’t know if they accept self-links, but it’s worth a try.
If you use Kagi, you can add this “lens” to limit search to these 10 largest Lemmy instances: https://kagi.com/lenses/qB39IMQCFQRpBL8fPKGmQ1wMw9nWGHNc
Instances: lemmy.world, lemmy.ml, kbin.social, lemm.ee, programming.dev, sh.itjust.works, lemmy.ca, beehaw.org, lemmy.dbzer0.com, lemmy.blahaj.zone . I’d add more, but the limit is 10. You can edit the instances list when you save the lens.
Oooooh this is neat. Thanks!
How do you find Kagi? I’ve seen it popping up more and more recently
I suggest going straight to https://kagi.com, or you can also use your current search engine to find it using query “kagi”.
But to be serious, it’s great. The search results themselves feel like how google used to feel years ago before the enshittification, and there are many useful features, like changing priority of different websites. I for example have removed all pinterest sites, reduced instagram and increased wikipedia weighting. They have a free 200 search trial, give it a go!
I’m a fan so far myself. It’s not always perfect but you can take your search to any of the other popular search engines right from Kagi if you need to.
It’s got a quick answer button which is using one of them new fangled AIs to pull an answer for your search out of the results, and then lists the sources for each bit it’s telling you. You can also “summarize a page” for any of the results.
You have these lenses that people were mentioning and one of the default ones that I use all the time is “forums”. So if my results are a bunch of shitty articles I click that and it shows me only sites like reddit instead.
I’m actually still evaluating it. It can returns results normally buried on google, and you can block or prioritize sites permanently on your search settings. So far it’s pretty good, but it’s still too early for me to decide if it’s actually worth the money, though I have signed up for unlimited plan so I can review it further. Google and Bing have this perverse incentive where making their search a more ineffective would result in large increase in revenue because you’ll search more and thus see more ads. The premise of paid search engine that don’t have any incentive to make you do more searches seems appealing to me. In fact, Kagi has huge incentive to make their users search as little as possible (and thus must deliver good search result) in order to be profitable because they said it them cost 1.5 cent per search.
There is a trial account with no credit card required that can be used for 200 searches.
Thanks, I was wondering whether this could be achieved using lenses
This seems like a good candidate for a bookmarklet that would append the (site:…) parts to an existing DuckDuckGo search result URL. Then you could just do a normal search followed by clicking the bookmarklet.
Why not just use your instance’s search function?
I dont know if can do a “fuzzy” search for me. Plus it may be showing my searches to the instance owner so ugh, dunno.
Because lemmy search sucks. Its very specific, and usually the most relavant stuff is buried by tangetially related things.
I think it is still miles better than Reddit… But I can be wrong.
If they are anything like me, they are used to reddit search being hot garbage, so they prefer to use a search engine. I have no clue if lemmy search is any good
One reason might be: Each instance only has a partial knowledge of content in Lemmy. It can be unaware of certain communities on other instances, if your instance has not discovered them yet. Hence the need for all these meta-search tools.
fedi-search method is good too. Its for google tho.
‘minecraft (intext:“modlog” & “instances” & “docs” & “code” & “join lemmy” | intext:“powered by kbin”)’
They seem to be able to search through all lemmy/kbin instances.
Ohhhh so you force it to search words that almost only and every lemmy or kbin page contain. Thats smart, might try it as well.
This might work for now, but some people will abuse this at some time to draw traffic to their cluckbait sites. Like in the old days with “index of” “parent dir” to find open dirs.
Yikes…
._.
For myself, I just created a
!lemmy
custom bang in Kagi with this URL:https://lemmy.ml/search?q=%s
(URL encoding on.)
You can replace the domain name with your instance’s and it should work the same.
Kagi is great.
This would only search one instance, is that right?
Correct. However, the instance will contain any federated content it knows of aswell. As long as there is at least one subscriber to the community from your instance, you should be able to find the community’s posts.
Nice, thank you. This was the piece I was curious about!
Yes
That’s actually pretty handy. Cheers!
I was literally wondering about this exact thing today, thanks for this!
Hopefully, one day we will have it SEO
Removed by mod
It works, tried now. Thanks.
Haha, happy it worked for you. I’m just a bit worried that it may somehow get confused with all the parameters and limit the quantinty of the results.
Thanks! This is very useful. Saved.
Really useful!! Thanks for sharing