Page 1 of 1

Opinions for Searx?

Posted: 2017-09-06 22:35
by Mad_7
Hi! :D

I'm sure that everyone knows of DuckDuckGo and StartPage.
I suppose it's a matter of personal taste, but seems to me that StartPage is the better option (since it's based outside US).

Does anyone use SearX? (https://asciimoo.github.io/searx/).
Seems very promising to me, but has anyone experience with it?
How it compares, with the above mentioned?

An interesting fact, is that it has several "instances" you can choose from, or even Searxes (for picking a random one).
(https://github.com/asciimoo/searx/wiki/Searx-instances).

What do you think about selecting an instance?
Just with server's geographic location?
Do you trust the person (or company) who owns the server?
Or which one will you trust and why?

I suppose, compiling and running it on your own machine is the better option, but isn't for the average Joe (so you can't suggest it, to your friends).
Well...what's your opinion? :D

Re: Opinions for Searx?

Posted: 2017-09-08 18:15
by kedaha
Thanks Mad_7 for posting about Searx; I'd never heard of it but it's interesting like the other search engine talked about, like Deep Search, in viewtopic.php?f=3&t=134539
I'll certainly try it. I think a different internet is possible if one uses alternative search engines to the dominant ones which have become like virtual shopping malls, always trying to sell you something.

Re: Opinions for Searx?

Posted: 2017-09-08 23:20
by Bulkley
Searx looks interesting. It somewhat reminds me of the old net crawlers from a long time ago. Anyway I thought I'd give it a try since Searx is in Testing. Unfortunately it wants to bring down a long string of dependencies. As I am using Stable it is a risk I'd rather avoid. If/when it comes into Stretch backport I'll try it.

Thanks for posting this.

Re: Opinions for Searx?

Posted: 2017-09-09 06:49
by debiman
^ maybe i'm misunderstanding something here.
why do you need to install this?
i added a search to firefox, that currently takes me to https://searx.abenthung.it/
i am not quite sure how all this works, but so far it works, i get good results.
and it would be really nice to find full alternatives to big G, bing, yahoo...

Re: Opinions for Searx?

Posted: 2017-09-09 12:27
by kedaha
Bulkley wrote:Searx looks interesting. It somewhat reminds me of the old net crawlers from a long time ago. Anyway I thought I'd give it a try since Searx is in Testing. Unfortunately it wants to bring down a long string of dependencies. As I am using Stable it is a risk I'd rather avoid. If/when it comes into Stretch backport I'll try it.

Code: Select all

$ rmadison searx
searx      | 0.12.0+dfsg1-1 | testing    | source, all
searx      | 0.12.0+dfsg1-1 | unstable   | source, all
It's true that it brings in a long string of dependencies but once you've backported it, remove searx-build-deps and run apt-get autoremove as detailed in SimpleBackportCreation.
If you wish to backport it yourself, just kick off with:

Code: Select all

$ mdir searx && cd searx
user@debian:~/searx$ dget -x http://http.debian.net/debian/pool/main/s/searx/searx_0.12.0+dfsg1-1.dsc
and

Code: Select all

root@debian:/home/user/searx/searx-0.12.0+dfsg1# mk-build-deps --install --remove
and follow the procedure detailed in the link.
The backport is easily created but I haven't figured out exactly how it runs yet.

Re: Opinions for Searx?

Posted: 2017-09-09 17:12
by Bulkley
kedaha, Thanks for the instructions. When I get time I'll dig in.

Re: Opinions for Searx?

Posted: 2017-09-09 20:02
by kedaha
Bulkley wrote:kedaha, Thanks for the instructions. When I get time I'll dig in.
It only takes a few minutes.
I've found more instructions at searx/dev/install/installation.html. I'm thinking about running it from my server on apache2 rather than on my desktop but I'm still uncertain how to get the installed debian packages (python3-searx and searx) working.
Meanwhile I've been trying it and like it a lot.
debiman wrote: maybe i'm misunderstanding something here.
why do you need to install this?
I've been reading up about it. The whole idea is not to use instances of it hosted by someone else but to run it oneself, as ones own search engine.