Seems like it is an independent index now
1 Like
PriEco did use Google results in the past, as well as Bing, Brave search and Mojeek results. But I decided to move forward with an independent index
Itâs still relatively small, but I am committed to grow it as fast as possible
Would you find current PriEco results any helpful?
Just would like to tell you that we surpassed 200M results in our index
Index size is truly the highest priority for us and will keep on growing it as fast as possible to have all of the results you might need
4 Likes
PriEco never tracks or profiles you.
Promises are good but ultimately meaningless.
Itâs open source
Does not matter. Unless you intend people to self-host their own search engine
Minimal use of JavaScript
This is generally good. Reduces complexity and attack surface.
Hashed IP addresses are stored temporarily
Not collecting is important but preventing collection is the way.
I was happy to see youâre doing it the right way by having Tor Browser advertise the onion site:
http://priecovk7jsuh3tvkh62c6j4oep3l5bldigpzmay26rdpqz357t5dmad.onion/
Alas, itâs down.
Please fix it asap and strongly consider adding the link to the Onion Service to the clearnet websiteâs footer. Itâs good to let people know it exists as it is the only way to guarantee privacy on userâs end.
Hi thanks for the feedback!
Right, but what can be done about it? You canât verify what actually runs on a server from outside. I promised I donât do such things but donât know about any real way to prove it to you. There are independent audits. Not sure how much additional trust they provide + if there are some accessible for me but would be glad and up to.
Again true, it works for now mostly as an additional signal for trust. I do have plans for user self-hosting, but thatâs for much later.
JavaScript, here I will again disappoint you. That was in an old PriEco website. I promised here already to bring it back. So, will tell you when itâs back.
I mentioned it too, PriEco uses Cloudflare as of right now. I was very hesitant and didnât want to add it, but was forced to because of a bot traffic PriEco gets. Would really like to get rid of it and there are ideas in construction but nothing truly actionable for now.
That said Tor/Onion version doesnât use Cloudflare (read more)
I even forgot about the hashed IPs. PriEco no longer does that. Again trust me bro
Very sorry, thatâs entirely my fault/mistake. My guess is that it was down for max 18h. Itâs online again
The link is there
And when you open clearnet PriEco in tor browser it signals you â.onion availableâ
Very sorry for the inconvenience and I hope I can make your experience better
1 Like
Nothing, and we shouldnât have to! 
Privacy community usually cares about FOSS servers to
- To self-host when the threat model deems it necessary
- Have a peek at code smell and to spot issues and vulnerabilities
For third party services supporting Tor with preferably v3 Onions available is how user can enforce the server to respect user privacy. All the server needs to do is not obstruct that, and actually recommend the user takes security into their own hands. If this was mine, Iâd have say a link to a PeerTube/YouTube video that shows how to install Tor Browser and how to navigate to the onion site, and how to bookmark the link.
I promised here already to bring it back. So, will tell you when itâs back.
Itâs not the end of the world. Anonymity and lack of long term sessions keeps users as private as they can.
Very sorry, thatâs entirely my fault/mistake. My guess is that it was down for max 18h. Itâs online again
No need to apologize!
The link is there
This is my bad, I didnât realize I could scroll down. It apparently doesnât show even if I maximize the browser on 4K display. Would you consider e.g. a faint V shape arrow symbol to bottom of the visible page to signal thereâs more to be seen? It could be a button that scrolls the bottom visible. I get that you want to keep it as minimal as you can so Iâm trying to think of something that would work but wouldnât ruin it.
And when you open clearnet PriEco in tor browser it signals you â.onion availableâ
Yeah this was a positive surprise, I love it!
Youâre doing great work and I confirm the onion service works now. Thanks for speedy fix!
2 Likes
How does utilizing Google and Bing not compromise privacy?
Forgive me as I have limited knowledge in this area and mean no disrespect.
1 Like
This is pretty much impossible for a web search engine as you need to get the results from somewhere. You could create your own index, but this would require a lot of work and most likely it would be small. Maybe this is what some people are looking for and thatâs great
I do have PriEco website and index code online. I try to keep it readable, but I would understand if itâs harder to understand.
PriEco is on v3 onion domain. Thanks for telling me about your vision!
Yes, the tutorials would be helpful too.
You can even set PriEco and PriEco (Onion) as a default web search engine in Tor browser. Iâm not expert here, maybee there is a chance to fingerprint this by some websites. Would doubt it but I am not 100% sure.
True, I would think of as the easiest way to improve this would be by stopping maximizing the landing page to 100% browser window height and do like 90% or so, so that the footer peaks from the bottom. Thanks!
1 Like
Hi! PriEco acted as a proxy for Google and Bing results. PriEco passed to them your query and settings as preferred language and returned results to you without sharing your IP or other information.
PriEco no longer does it. Itâs independent in terms of web results.
Oh, to be 100% honest, PriEco does get result favicons from Google. The only information are individual URLs, no query, IP⌠This is actually easy to fix and make right to serve them from PriEco server instead
Havenât you considered sharing your index as well? This way users could replicate PriEco on their own infrastructure and get the same results without sacrificing privacy.
I have. For now, I just focus on improving PriEco itself.
As of now, I donât have the hardware and bandwidth to just let anyone download a full copy. There are public datasets you could use, although I would usually recommend to also process the data a bit.
PriEco index is very small (200M web pages) in comparison and designed to be very space efficient, but it wonât stay small forever.
I have this idea for a long time but yesterday was talking about it with someone
For more context:
but save them : I wanted to say that they would be creating a local copy
more active downloading: Is likely what you have in mind
Regardless, this idea is for much later. And about just being able to manually download a copy, donât have infrastructure for this. Maybe if you donated 
1 Like
@any1 and @maqp I got to it a bit later than I wanted to, but itâs finally here
PriEco can now function fully without JavaScript.
(Settings no longer auto-submit, you need to press that save button)
Benefits
PriEco auto-detects that your browser doesnât support JavaScript and sets itself up
(Optional) You can bookmark these URLs:
And use them as your first search query. They set up PriEco for you, in this case:
lang: en
loc: us
theme: dark (other option is light)
no-js: on (that js=1)
newtab: if you want to open clicked result links in a new tab
return to: tells PriEco where to redirect you after the preferences are set. In this case it makes privacyguides
Edit: You donât have to specify all parameters, only the ones you care about, if you donât specify the return_to the url redirects you to PriEco landing page
Negatives
PriEco is still behind Cloudflare, it is possible for it to inject some JS to the page, but if your browser doesnât support it, then it wouldnât be executed.
Same problem, if Cloudflare captchas you, you wouldnât be able to access PriEco as the captcha requires JS.
That said, it should be amazing to use it in a Tor browser with .onion domain which isnât behind Cloudflare
I hope I didnât miss anything, and it works well, if you experience any issues I would like to hear them
4 Likes
This does not currently work for me. I have to enable JS, set the NO JS option and then it works with JS disabled.
Edit: This works on Trivalent with JS disabled, but not on LibreWolf with JS disabled via uBlock Origin.
I tested it using Tor browser
Glad that PriEco works nonetheless
My rough guess is that disabled JS by Ublock prevents JS spoofing
I can give you what I do in PriEco
<noscript>
<meta http-equiv="refresh" content="0; url=/set?js=1&return_to={{{ current_path }}}">
<p>Configuring PriEco for No-JS mode...<a href="/set?js=1" class="link">Click here if you are not redirected automatically.</a></p>
</noscript>
This html should tell your browser if user doesnât have working JS, redirect them to an URL, which auto sets js cookie for you. Obviously I donât personally get any information about your JS usage. Itâs your browser who has to make the decision
Maybeee adding *##^noscript > meta[http-equiv="refresh"] to ublock My filters might fix it
I tried searching for a major retailer in my city - it gave me 20 results and none of them were the one I was looking for.
Iâm guessing the index needs to be built out further before it can be used on a daily basis, at least in my country.
2 Likes
PriEcoâs index is on the smaller side
I do try to grow it as fast as I can
It supports DuckDuckGo created concept: bangs. So, if PriEco doesnât find what youâre looking for, you can add ! query.
For example !ddg privacyquides
or privacy !brave guides
It can be anywhere in the query, just make sure you put there spaces
Iâm aware it isnât yet good enough to be used as the only web search engine for everyone. But I believe especially now itâs improving pretty fast.
I came back to this forum mainly to get feedback to get a better perspective of where PriEco is at.
I strongly appreciate even if you used PriEco time to time, suggested features or better ways to do things, or just that you tried it
2 Likes
Using it for a little while now, itâs totally a usable alternative when taking the bangs into account. Itâs missing a lot of sources, but itâs been twice now in only two days that I searched for something and got an unexpected ânicheâ result that was actually very interesting and helped me out, I guess there are advantages to the early stages of web indexing. I really like this project!!
Could it be that the !brave bang is not working currently? Iâm a DDG user primarily anyway, but I tested both and only one works.
3 Likes
Thank you for such nice feedback, you really made me happy!
There are not just DuckDuckGo and Brave search, there are more than 13000 available bangs
For me Brave search bang works, I canât make a guess of what could be the problem
Could you give me more information? Like query, browserâŚ
This could be the reason for finding niche results. Another reason might be that there arenât that many web search engines and especially western ones usually heavily optimize their results on Google.
Bing: Google: Bing Is Cheating, Copying Our Search Results
Brave search is much better, I donât have any prove they did but there have been some learning from Google back when it was called Tailscale under different management
PriEco doesnât do that yet, I did integrate for example PageRank algorithm and did learn about other search engines. But that was a knowledge I then applied directly to PriEco without data from other search engines.
That said there is still a huuge pile of knowledge covering how to make a better web search engine, and Iâm willing to do so much to improve PriEco. Obviously if any user generated data would be required, it would be opt-in with clear explanation.
I am not that keen on optimizing using Google as that would make PriEco still independent but too similar to Google (as I believe Bing and Brave search are). It would make it behave more like what user expect, but why be creating another Google clone? Iâd rather see what I can do without copying and see where it leads
