Change approach : make uBlock an application instead of a browser plugin ( a.k.a. : save uBlock from Manfest V3)

[thealio]

Hi
I wanna be pretty short and go directly to the point of this discussion.

Referring to #2261

Manifest V3 will limit the power and the functions of browsers extensions , especially adblockers.

The new uBlock Origin Manifest V3 will be actually less powerful and less usable .

Also , the browser-puglin nature of the project limits this to be used only for chromium and firefox based browsers, kicking out the rest of the other browsers.

I' like to suggest another development approach: made Ublock a standalone application , that works in background, becoming browsers- independent.

This is the only solution to the problems I mentioned above in my opinion.

[MasterKia]

Are you willing to put the effort?

[thealio]

If you're talking about donations , of course.
Otherwise, I don't understand your point, because to develope a less featured extension will mean lots of efforts, because you 'll have to arrange it to manifest v3 limitations, and answer to a lot of users that will not understand why uBlock don't block ads like before, and all the related issues.

[gorhill]

I' like to suggest another development approach: made Ublock a standalone application

@thealio Why ask someone else to implement your idea? If you strongly believe your idea is the way to go, why not just work on it?

[thealio]

Because you're the main Dev, and this is a discussion tread to share ideas.
ublock will be replaced by uBlock lite, that will not be as powerful as the manifest v2 version, and I think developing a bad copy of the manifest v2 version is not a solution.
Your answer is silly, the discussion aims to suggest and share idea, your "do it yourself" answer has no sense in my opinion

[gorhill]

your "do it yourself" answer has no sense in my opinion

You idea is trivial, it has been talked over and over over the years.

If I am not doing this now, it's not because I could not have come up with such a trivial idea. It's because I do not have to the time and motivation to undertake such involved and difficult project.

I know how content blockers work, and I know the huge amount of work this would require, and the fact that this would require to deal with security-sensitive issues. I have no interest in getting involved in all this, I have a real life outside GitHub. AdGuard offers such product, you can already get this from them.

Whoever thinks it's a good idea, please undertake it yourself, good open source projects are not born out of a random person on the internet suggesting another person to implement their idea, it's people scratching their own itch in the first place.

People asking that I work on their idea has always been a recurring theme, hence why I use to put a warning against this in the project's contributing file:

Do NOT
...

• Submit "revolutionary ideas".
  ...

TLDR: Please scratch your own itch, don't ask others to do it for you. uBO is GPLv3, whatever is found to be useful in it can be reused in another GPLv3 project.

[krystian3w]

I might only suggest that someone write a user manual for the uBo in the "Node" version:

• how to update the base filters in it
• how to change your subscription
• how to add a completely customised list like letsblock.it
• how to save something to the My Filters panel as it exists there
  • whether firewall mode (medium/hard) can be used in it

Maybe then you* (* - 2-nd person plural) won't need to write a uBlock Origin Browser from scratch based on Electron.js or NW.js.

[thealio]

@gorhill
I just suggested something that seemed to me a solution for a problem.
This is a discission thread where people can gently ask , to receive gently answers.
Nobody should blame you for agree or disagree to users ideas to improve the project.
I'm not asking you to do something I should do myself, this is intended only a brainstorming thread.
BTW , YOU are the main developer , so YOU have to "scratch your own itch" on this project, not me, of course your can agree or disagree with users' ideas and reject them , no problem.
The problem sussist if you answer with such things like "do it by yourself" or "scratch your own itch", because I'm not a developer, you are, and this kind of answers sound like very silly and not helpful at all.

[Yuki2718]

If you're talking about donations , of course.

Donation doesn't help anything. All of us are volunteers spending spare time for this project and are not working for $$$.

[stephenhawk8054]

YOU have to "scratch your own itch"

The idea is suggested by you. This is not his own itch, and he doesn't have plans for it. The inside codes and security models between an extension and a standalone program are 2 different worlds.

This is a voluntary project. Everyone involved here are volunteers, contributions are done by each one's work including detailed reports and resolutions for the issues. This is not donation model where anyone can throw the money and ask others to work on some random ideas.

[thealio]

I'm not talking about a plugin like flash.
I'm talking about a standalone applications

[Yuki2718]

Do you even understand what you are suggesting? If you think converting browser extension (based on javascript) to native app is so trivial, try to do by yourself first. Only then you will understand what you're suggesting.

[gwarser]

MV3 api is limited, but outside the browser there is no api at all.

[krystian3w]

I doubt that turning uBo into a native application like AdGuard will be easy.
I tend to rule out a fork based on Electron.js or NW.js.

[Sp0kzz]

brave shields is built-in and uses uBo filters

[krystian3w]

But no support :nth-child / :nth-of-type…, so 30+ filters can not by used without rewrite to something else e.g. with :has-text (by human, not simple script in Rust).