# My recent changes towards more privacy friendlier life **URL:** https://discuss.privacyguides.net/t/my-recent-changes-towards-more-privacy-friendlier-life/38166 **Category:** General **Created:** 2026-05-27T19:14:38Z **Posts:** 43 ## Post 1 by @robots.txt — 2026-05-27T19:14:38Z ![image](https://forum-uploads.privacyguidesusercontent.com/original/3X/9/5/951669117f1671e100c299e28294fea6eba3b97c.png) The most recent and I think hardest changes have been transitioning away from the google services because they are really good at what they do but the downside are the privacy concerns so in the end I don’t feel so bad about it. I just have to get adjusted to these new ones and I think it will be okay. It took while to move out from the bitwarden and I’m glad it is finally done. My current strategy is to keep the password file all the time encrypted and spread it to several devices. The most recent backup goes to cloud and from where I update it to the other machines and other backup methods. I don’t find it burden but actually fun because I know I have full control over them. Also, I wanted to move away from the browser extensions to be less fingerprintable and just use fewer extensions overall. My long term goal which I will probably add here is that I focus more on the self hosting and services that are actually safe from leaks. My another goal is to build linux phone from scratch and I have been researching this topic for while now. This is something I would like to build by myself. I would like to include physical killswitches too. If you have any feed back or questions let me know --- ## Post 2 by @FranklyFlawless — 2026-05-27T21:55:29Z What are you replacing DuckDuckGo with? --- ## Post 3 by @Expert4870 — 2026-05-27T22:29:49Z I recommend LibRedirect to make it easier for you to use the frontends. I think it blocks any network requests from the target site which is pretty cool. --- ## Post 4 by @FranklyFlawless — 2026-05-28T01:13:58Z It does not by default, and embedded injections/interceptions are only supported on specific front-ends. --- ## Post 5 by @Expert4870 — 2026-05-28T01:22:50Z This doesn’t mean that it does? > webRequest, webRequestBlocking: To block http requests or redirect them (e.g. to redirect to Invidious before it can reach YouTube’s servers). > **[LibRedirect - Privacy-friendly Redirector](https://libredirect.github.io/faq.html#why_all_those_permissions)** > > A web extension that redirects YouTube, Twitter, Instagram, etc. requests to alternative privacy-friendly frontends --- ## Post 6 by @Bumbashirovich — 2026-05-28T01:30:50Z Does anyone store the KeePass password database on Ente Locker? --- ## Post 7 by @FranklyFlawless — 2026-05-28T01:31:03Z There are two mechanisms at play here: 1. URL redirection in your address bar, from [youtube.com](http://youtube.com) to [invidious-instance.org](http://invidious-instance.org) 2. Embedded redirection, when navigating within third-party websites The former is what is the primary use case for LibRedirect, but embeds are a separate case. The term “target site” is ambiguous and could mean two options: 1. The main site, AKA, YouTube 2. Some random third-party site that uses YouTube embedded videos (or other embeds supported by LibRedirect) The latter is not covered by default, it is opt-in, so for example, on Privacy Guides, if someone embeds a YouTube video in a topic, it will not redirect to an Invidious instance unless you have configured it to also redirect embeds as well. --- ## Post 8 by @SwampTrainer — 2026-05-28T05:19:55Z Just wanted to say keep up the good work! --- ## Post 9 by @WhinyHamletPayer — 2026-05-28T14:08:33Z Avoid Linux phones > **[Linux Phones | Madaidan's Insecurities](https://madaidans-insecurities.github.io/linux-phones.html)** --- ## Post 10 by @robots.txt — 2026-05-28T14:12:16Z I have not decided yet. I have been researching searxng which I would love to host locally without opening ports to the internet but I don’t know if it’s possible But I’m also using the startpage sometimes. What I really dislike is that where the search engines are heading because it is almost impossible to find mini blogs who have researched / shared their opinion on matter x or any other useful websites. Either the search results are generated with AI or they offer “centralized” websites (for example geeksforgeeks with programming related questions) --- ## Post 11 by @robots.txt — 2026-05-28T14:13:04Z How does the network request blocking work with the LibRedirect? --- ## Post 12 by @robots.txt — 2026-05-28T14:13:46Z is Ente Locker like cloud hosting service? --- ## Post 13 by @robots.txt — 2026-05-28T14:14:59Z Thank you! I will keep updating the chart in the future :slight_smile: --- ## Post 14 by @dada_goose — 2026-05-28T14:29:06Z You may have already seen it but have you had a look at Kagi? It has a “small web” lens which specifically searches for things like personal blogs or small independent sites. It’s paid of course but I have to say, as a doubter for a long time and only a recent convert, I absolutely love it. The results are great, it’s privacy focussed and it’s incredibly customizable. If you want to you can even write custom CSS to make it look exactly how you want, I’ve had a lot of fun with that feature recently. --- ## Post 15 by @seize — 2026-05-28T14:35:24Z Sort of… From what I have gathered (I don’t use Entered Locker) it is a mix between “traditional cloud storage” like google drive, but with E2EE[[1]](#footnote-155691-1) and something more like bitwarden or proton pass and thier “secure vaults”. References: - [https://alternativeto.net/news/2026/2/ente-launches-locker-a-new-open-source-secure-vault-with-end-to-end-encryption/](https://alternativeto.net/news/2026/2/ente-launches-locker-a-new-open-source-secure-vault-with-end-to-end-encryption/) - [Ente Locker - Safe space for your most important documents](https://ente.com/locker) * * * 1. End to end encryption [↩︎](#footnote-ref-155691-1) --- ## Post 16 by @FranklyFlawless — 2026-05-29T03:15:14Z > [@robots.txt](#): > > I have been researching searxng which I would love to host locally without opening ports to the internet but I don’t know if it’s possible Yes, you need to set the SearXNG’s IP address to `127.0.0.1`, which than resolves to localhost. After deployment, you can navigate to `http://localhost:port`, where `port` is the container port number you assigned SearXNG with. > [@robots.txt](#): > > What I really dislike is that where the search engines are heading because it is almost impossible to find mini blogs who have researched / shared their opinion on matter x or any other useful websites. Right, if you want to take matters into your own hands, you may want to use your own crawler and indexer, although you may discover that truly independent voices are few and far between. --- ## Post 17 by @sahqw — 2026-05-29T14:22:09Z I use DuckDuckGo a lot, and I’ve started to report AI generated websites. They have that in their feedback menu. However, this does not seem to do anything. They’re still showing up in results and I’m probably one of a few people reporting them. --- ## Post 18 by @FranklyFlawless — 2026-05-29T14:45:43Z I understand, you may be interested in this topic: > [@uBlock Origin filter list to block AI generated content and websites](https://discuss.privacyguides.net/t/ublock-origin-filter-list-to-block-ai-generated-content-and-websites/35670): > > Here’s a new filter list I found that one can use alongside other filter lists within uBO. Let’s keep the web human! --- ## Post 19 by @sahqw — 2026-05-29T18:36:47Z Very nice, thank you for sharing. --- ## Post 20 by @ikelatomig — 2026-06-01T02:27:58Z For the record. Embeds are not redirected because it would be a process of injecting scripts to replace embeds. For normal redirections, it only changes the URL. As a matter of trust, we disabled embed redirections by default. Since we haven’t started to port to Mv3. --- ## Post 21 by @ikelatomig — 2026-06-01T02:29:44Z It watches the network of browser, if [youtube.com](http://youtube.com) is requested like a security guard. Blocks it before the request gets out of the browser to YouTube server and redirects to Invidious. --- ## Post 22 by @seize — 2026-06-01T02:39:31Z > [@ikelatomig](#): > > As a matter of trust, we disabled embed redirections by default. Since we haven’t started to port to Mv3. @ikelatomig Are you a developer for libredirect? If so it would be advisable for you to reach out to the team here so that your profile can properly reflect this. --- ## Post 23 by @ikelatomig — 2026-06-01T02:46:10Z I am one of the maintainers you can see it in Github repo. Not the lead developer. So, I didn’t request for it explicitly. --- ## Post 24 by @seize — 2026-06-01T02:51:27Z Fair, I was just a little confused :sweat_smile:. Thanks for clarifying. --- ## Post 25 by @FranklyFlawless — 2026-06-01T03:15:29Z Understood, does Manifest V3 change the landscape for embeds or URL redirection? --- ## Post 26 by @ikelatomig — 2026-06-01T03:53:54Z Well, it does by a large degree. With Mv3, Libredirect would not see any URL at all. Libredirect provides a set of rules to the browser. The browser handles it. Similar to how Predirect works which is built on Mv3. Currently, it monitors requests and manually does everything as I said earlier. The extensive permissions is mostly for **convenient** features. If you are user who values strict security and has trust on permissionless model. Go for predirect. Libredirect is meant for convenience of the users. So, it will need some permissions that allows delivery of those features. Essentially it is based on trust placed by the users on the developer despite being open-source. When moving to Mv3, for embeds, a tooltip will be placed for explanation and additional permission granted by the user, (i.e. Optional) and not declared as required permission on install itself. It will still require “Access your data for all websites” because the extension will need to inject a script that replace the embed URL. Nothing much. So, when user loads a website embeds are redirected automatically. If the user don’t like that permission. You can invoke a shortcut key that initiates the extension to replace embeds on a per-site + per-input basis without permission at all via permissionless model. * * * Same goes for Copy Raw which uses `all_urls` permission, it could be disabled if permission is declared optional and enabled by the user itself * * * What could be done to improve is having a better on-boarding experience that explains what is what and make the user configure as per their needs along with a default option where it is configured optimally for the majority of the users. After porting, I or other plan to extensively document permission system and use cases. So the user understands. But it will take a long time, some years for sure. Both Mv2 and Mv3 will co-exist like uBO and uBO lite. That is all. If Mv2 and Mv3 gets feature parity. Maybe then Mv2 will be sunsetted. So basically, it will be trustless/permissionless system and permissions are requested via controls that user understands and allows as needed. * * * From a user standpoint of view, Mv3 is good as it creates a balance of permissions and revokes access an extension could have and gives user privacy & security. But from a developer point of view, it would be either nightmare or difficult. But not impossible. Mv2 was like painting in the flat ground with no boundaries. Mv3 is like painting a sheet of paper you are given. The FAQ about Mv3 was written before Mv3 was fully adopted. So it could be errors or true based on how the Mv3 port of the extension turns out to be. * * * Forget to say this, the reason embeds are disabled by default is that. Certain sites have strict CORS which block the replaced embeds from loading at all. That is also one of the main reasons that embeds are disabled. We don’t want users to have a negative experience while browsing. Only YouTube embeds works fine that too if the site allows it. If it blocks other domains via CORS. That’s it. Nothing else works. Nitter can be replaced but the instance disabled it or the project didn’t fix it after X rebrand. There aren’t much embeds at all apart from these. Imgur, I don’t even know. Need to test. * * * Sorry for the long message. --- ## Post 27 by @FranklyFlawless — 2026-06-01T03:59:17Z Thank you for the information, understanding the web extension landscape is useful, because I am pretty interested in formulating my own technical solution towards my situation. In particular, I understand that privacy front-ends and static filter rules are insufficient because upstream changes the rules and the development work required afterwards is unscalable against them. --- ## Post 28 by @ikelatomig — 2026-06-01T16:08:32Z > [@FranklyFlawless](#): > > because upstream changes the rules and the development work required afterwards is unscalable against them. It is a cat & mouse game. This is the pain that frontend developers face mainly. If it is blocked by original service. Many instances will shut down. It will trigger a chain of events and discourage public hosting. Say Bibliogram one of the well know frontends for Instagram was blocked by meta from scraping since they can’t profile users and at the same they sent a cease and desist letter to the project too. Woth court battles. IIRC Then they shutdown. Now there is one in Libredirect. Working as of now. It is also matter of time. For Libredirect it is just a matter of domain change. But front-end devs and instance maintainers are the real saviors of privacy in digital space. Donate to them, if you can, than us. Many instances shut down due to increasing costs in VPS. --- ## Post 29 by @conn7867 — 2026-06-01T21:53:58Z You know what they say, cash is King. --- ## Post 30 by @FranklyFlawless — 2026-06-02T04:06:51Z I stopped being a privacy front-end operator because I became tired of upstream breakage. I would rather achieve a more persistent solution that does not involve fragility. In addition, my pull requests on Codeberg towards privacy front-ends and web extensions were being ignored over months, so I have stopped contributing to the commons entirely. I get more accomplished working alone and refusing external contributions instead. --- ## Post 31 by @mygdoors — 2026-06-02T14:49:36Z Try Kagi and their “small web” initiative. --- ## Post 32 by @FranklyFlawless — 2026-06-02T16:28:07Z Not interested, they are far from a good fit. --- ## Post 33 by @anonymous555 — 2026-06-02T20:17:51Z Hello, Why did you hearten / “like” that reply though :slightly_smiling_face: ? --- ## Post 34 by @Expert4870 — 2026-06-02T21:39:05Z Farming points --- ## Post 35 by @FranklyFlawless — 2026-06-03T09:49:39Z It means I made the time and effort to read your post. --- ## Post 36 by @vassilis — 2026-06-03T15:32:54Z I’ve been thinking about switching to a linux phone for a while now, but ultimately I tend to gravitate towards a feature phone, the ones from the ‘old times’. No sensors, no cameras, nothing. Just the keys to make a call and send an SMS. What I’d love to replace my smartphone with is a linux tablet. It has to be small enough to carry around with me at all times and preferably operate on Debian. However I haven’t been able to find such a device (yet). --- ## Post 37 by @FranklyFlawless — 2026-06-04T04:01:16Z Linux tablets are easy to find, but the reality of using them as a daily driver may frustrate you if your background is coming from Big Tech devices. --- ## Post 38 by @vassilis — 2026-06-04T19:46:56Z I’d love some recommendations, if you have any! It’s been a while since I divorced big tech. I won’t miss anything important by replacing my phone with a Linux tablet :wink: Thanks in advance! --- ## Post 39 by @FranklyFlawless — 2026-06-05T22:27:46Z Okay, I have one recommendation: > **[StarLite 12.5-inch Linux 2-in-1](https://us.starlabs.systems/pages/starlite?shpxid=27dbe7a3-a330-4fe6-822b-a4ab4bae6bc0)** > > StarLite is a 12.5-inch Linux tablet and laptop with a 3:2 touch display, Intel Processor N350, coreboot firmware, pen support and a removable keyboard. --- ## Post 40 by @anonymous555 — 2026-06-06T06:15:54Z This is not how likes are meant to use… --- ## Post 41 by @FranklyFlawless — 2026-06-06T06:41:22Z The only thing that matters to me is that I use them at all to begin with. --- ## Post 42 by @krystalline — 2026-06-06T13:21:14Z In general a report like this will not take immediate effect, because of the abuse potential. Either that puts it in a manual queue which I doubt, or it depends on many people reporting the same. I expect the feature is quite obscure and that getting an ai site filtered would take a while. An addon with a quicker flow of review or custom additions would be my choice to get quicker results. And yet more and more of those sites show up… --- ## Post 43 by @privacylover — 2026-06-06T14:55:01Z > [@vassilis](#): > > What I’d love to replace my smartphone with is a linux tablet. It has to be small enough to carry around with me at all times and preferably operate on Debian. It runs Ubuntu Touch: > **[Volla Tablet | Volla Devices](https://volla.online/en/volla-tablet/)**