About search engines

Serax is a great advance since it allows to install and modify an efficient search engine and we can all install one or look for a trusted server, but it has a limiting problem, it is not p2p, it can not be federated, this prevents collaboration between different instances Way of competing with large corporations and their great resources)

Did you think about Yacy, the p2p search engine?

1 Like

And about;

Gigablast

Findx

They are decentraliced engines instead Yacy. Findx is based on Gigablast.

AFAIK findx is an actual search engine which indexes the web but Searx is a meta search engine. Searx just queries other engines and gives the results.

But findx has its own database which grows every second, it will be a huge waste of resources. By waste I mean, we can use it for other services. There are search engines out there which respect your privacy and Searx does too.

If you still want to use findx and find the results great then you can enable findx in your preferences! Disroot is already planning to deploy some services, there is a poll for some services. Vote over there by chosing from any of the four.

I think Searx will be good for now, just update your preferences as you wish!

I know

Where can i vote?

you can check the poll here:

https://poll.disroot.org/ff3Qbrs1Kh74pHtN

1 Like

But findx has its own database which grows every second, it will be a huge waste of resources

By resources, do you mean storage of the index? This is all just URLs and metadata, right? Just text. How much could storage space could it really use? If you mean bandwidth use, would this really be that much more than using Searx to query corporate search engines?

Besides, if Disroot, and RiseUp, and Framasoft and all the other freedom-respecting hosting collectives each hosted a node of a distributed search engine, you could each set a limit on how much storage and bandwidth your own node uses. But the combined effect would be a full search engine that doesn’t depend in any way on centralized, corporate-owned search engines. That would be the ultimate in de-gOgle-ifying the internet.

Idk, I haven’t worked with these stuff, but I guess even if the index has to store URL and metadata it will still be huge!

Still, Idk much. @muppeth How does this work and can disroot use findx instead of searx.

··· On May 9, 2018 12:58:20 PM UTC, strypey wrote: > > >> But findx has its own database which grows every second, it will be a >huge waste of resources > >By resources, do you mean storage of the index? This is all just URLs >and metadata, right? Just text. How much could storage space could it >really use? If you mean bandwidth use, would this really be that much >more than using Searx to query corporate search engines? > >Besides, if Disroot, and RiseUp, and Framasoft and all the other >freedom-respecting hosting collectives each hosted a node of a >[distributed search >engine](http://wiki.p2pfoundation.net/Distributed_Search_Engines), you >could each set a limit on how much storage and bandwidth your own node >uses. But the combined effect would be a full search engine that >doesn't depend in any way on centralized, corporate-owned search >engines. That would be the ultimate in [de-gOgle-ifying the >internet](https://degooglisons-internet.org/). > > > >-- >*Previous Replies* >you can check the poll here: > >https://poll.disroot.org/ff3Qbrs1Kh74pHtN > >Posted by muppeth on 03/26/2018 > >Where can i vote? > >Posted by elr on 03/26/2018 > >I know > >Posted by elr on 03/26/2018 > >AFAIK findx is an actual search engine which indexes the web but Searx >is a meta search engine. Searx just queries other engines and gives the >results. > >But findx has its own database which grows every second, it will be a >huge waste of resources. By waste I mean, we can use it for other >services. There are search engines out there which respect your privacy >and Searx does too. > >If you still want to use findx and find the results great then you can >enable findx in your preferences! Disroot is already planning to deploy >some services, there is a poll for some services. Vote over there by >chosing from any of the four. > >I think Searx will be good for now, just update your preferences as you >wish! > >Posted by Avizini on 03/24/2018 > >And about; > >Gigablast >- http://gigablast.com/ >- https://github.com/gigablast/open-source-search-engine > >Findx >- https://www.findx.com/ >- https://github.com/privacore/open-source-search-engine > >They are decentraliced engines instead Yacy. Findx is based on >Gigablast. > >Posted by elr on 03/23/2018 > > > >--- >[Visit Topic](http://forum.disroot.org/t/about-search-engines/1894/7) >or reply to this email to respond. > >To unsubscribe from these emails, [click >here](http://forum.disroot.org/email/unsubscribe/bedd80991c64c82bb9177ab2a16a19c2aeec8b4c4c983c740bb87a5d76cfaa79).

Avizini

I guess even if the index has to store URL and metadata it will still be huge!

As I said, if it is a node in a distributed search engine like YaCy, the people operating the node can choose how much of the index to store locally. I believe it may be possible to add a YaCy node as one of the search engines meta-searched by Searx. It may also be possible to feed the search results obtained from the other engines back to the YaCy node, helping to grow the overall index available to the YaCy network.

How does this work and can disroot use findx instead of searx.

To be clear, I was responding in support of @juansantiago’s suggestion of YaCy. I’m not sure what @elr means when they describe Gigablast and Findx as “decentralized”. Both are free code (“open source”), but I can’t see anything on their homepages or in their repo, that suggests if you ran two or more instances of them, they could connect to each other the way YaCy peers can.

Hosting YaCy is definatelyl something we want to do as disroot. When making decission on which search engine to host, we knew going for yaCy at the early stage of the porject would not be a good idea. We went for searx because it does amazing job as a metasearch engine.

We will in the near future revisit the idea we’ve had in the beginning, which is hosting an populating the search on YaCy. For this of course you need dedicated people that want to take on this challenge and maintain the health of the engine. We would also like to involve other services (like foramasoft you mentioned) specially with the recent growing interest in fediverse in general (mastodon, diaspora, hubzilla etc). I think we are gettin the critical mas that is needed to make YaCy a relevant trully decentralized search engine. Plus we can add it as one of the searx search engines too.

I guess (specially if there is enough people to volounteer) in near future we will work on proper camaign.

Awesome! I think approaching it a shared project with other user-respecting hosting groups is the way to go. Not only because this will result in a more powerful and useful network of YaCy peers, but because your tech teams can support each other in learning how to make the best use of YaCy, limit the resources it uses on each team’s servers etc. Also, working as a collecting of hosting groups means you’re more likely to do thorough documentation as you go, making it easier for other folks running servers to follow your lead and set up their own YaCy peer.

My sysadmin skills are incredibly basic, but I’m keen to upskill, and to help with some of the communication between projects, and documentation.

sorry I meant centraliced, this is why I said “instead of YaCy”.

Anyways, findX is dead and devs arount it did not recommend using their code or Gigablast code because, in their experience, it’s an amount of shit

Any progress on this? I know you’ve been fighting technical fires on other fronts, so no pressure. If there’s anything I can do to help with the human<>human communication with other hosting groups (about collaboration on YaCy or federated search in general), I’d be happy to help. If there’s sufficient interest, a federated search board could be created on talk.feneas.org (the forum of the Federated Networks Association).

Atm we do not have resources (human resources) to work on this. Our roadmap for upcoming months is full and we are running behind the plan already. Maybe as a side project at some point we could play around with it but to be honest there is already a long list of side projects lining up.