Manifest V3, webRequest, and ad blockers
-
Vivaldi comes with its own built-in Ad and Tracker Blocker, tons of customization options, a built-in Mail and Calendar, and more. Will it follow Google’s strict reliance on Manifest V3? Vivaldi Developer, Julien Picalausa takes a deep dive into this matter.
Click here to see the full blog post
-
@julien_picalausa About “Could we keep using our ad-blocking extensions in Vivaldi?”: Even if Vivaldi somehow managed to keep extensions relying on manifest v2 functional, these extensions will disappear from the webstore when the switch to v3 happens. I feel it would make more sense to try and bring advanced functionality to the inbuilt blocker, so people don’t feel the need to use an extension in the first place.
-
So, basically this means that we all lose the powerful features that Ublock Origin has at this moment. And that using Vivaldi's own blocker is going to be similar (though I expect less features) as using the possible upcoming "Ublock Origin Lite" that throws away pretty much every powerful Ublock Origin feature.
I had hoped that Vivaldi would have done more. I had hopes that Vivaldi fighting against this V3 bs would mean that using the original Ublock Origin would be still possible.Google. Don't be evil, Eh? That company has way too much power.
Seems that I'll be forcing myself to use internet less and less from now on, I can't go back to ads ADS ads ADS popping up everywhere. Ads are the cancer of internet (and all kinds of JavaScript heavy flashy, unusable web3.0, or whatever the trendy word today is, design mess).
-
It's good to hear that Vivaldi's internal ad blocking system will keep working, and extensions might work for a little while longer yet through the enterprise code being available. That's all very good to hear.
But what bothers me is it means that the granularity offered by extensions will be lost eventually. For example, Vivaldi can only toggle between (Off, Tracker Blocker, and Ad Blocker). But in uBlock Origin and similar "switchboard" style extensions you get individual control of domains that load other resources (not to mention the visual filtering capabilities, webfont blocking, easy one-click disabling javascript).
This is what I really think is being lost with the change to MV3, and its removal of programmatic blocking web requests. At the end of the day, if the ad blocking ability was removed altogether users could always use a special host file to block it at an OS or router level. But having user control of specific 3rd party resources on each site is a special feature that can only really be offered by the web browser, or an extension.
-
@Preorian I’m using the internal blocker as we speak. It can do much more than ublock lite, because there aren’t any limitations on the number of filter entries. You just have to learn how to set it up properly (adding additional lists, creating your own custom filters, etc).
-
@luetage: Yes, that is an issue. The only real fix there would be for us to have our own extension store, which would be quite a bit of extra work. Side-loading will always be an option of course. I suppose it's going to be a bit of a question of whether extension makers want to talk to us about this as well.
Software developer at Vivaldi since January 2017.
-
@preorian: No, this means that the way things look at the moment, we expect that the Vivaldi inbuilt adblocker will keep all the features it currently has and potentially more, as we have plans to keep expanding on it. We are not subject to the same limitations as extensions in general.
-
@julien_picalausa said in Manifest V3, webRequest, and ad blockers:
@preorian: No, this means that the way things look at the moment, we expect that the Vivaldi inbuilt adblocker will keep all the features it currently has and potentially more, as we have plans to keep expanding on it. We are not subject to the same limitations as extensions in general.
Ok. Still I'm skeptical that current uBlock Origin users who are using the more advanced features will have to settle for a lot less with Vivaldi. Sorry for saying, but I can't see you topping what gorhill has done with uBlock Origin, honing everything year after year working hard. Maybe you could contact him and recruit him? (though he seems to be a kind of ahole person, his responses to many people are really harsh (to put it mildly)).
-
unfortunately the Mv2 version of the API is already severely limited, while it makes sense to use it for the time being you probably should be working on a native solution instead
or adopting and improving existing one, Brave has a quite capable solution, even if the configurability isn't there yet the internals seem to do the right job, should be enough to replicate the FF uBO -
@preorian: Chromium doesn't offer full uBO capabilities anyway, Mv3 is a chance to get browser vendors to implement them natively
-
@zakius said in Manifest V3, webRequest, and ad blockers:
Mv3 is a chance
Let's be clear here, Mv3 is a detriment and a way for google (*phtui*) to protect their ad-scheme/revenue, not a "chance" to make something better.
-
@Preorian it's bad enough it may push browser vendors to offer more than terrible Mv2 extensions can do now, so in that way it is a chance
Chromium variant of uBO is severely crippled due to API limitations and bugs but is considered good enough by many, when even that becomes unavailable it may motivate teams to start working on native, fully fledged content blockers -
@zakius said in Manifest V3, webRequest, and ad blockers:
to offer more than terrible Mv2 extensions can do now
Oh please, this is pure nonsense.. "terrible" Mv2 extensions? Really? Can't take you seriously after that. Like i wrote earlier, I very much doubt that Vivaldi would put so much effort and manpower that they would reach even close to what gorhill have over the years with uBO. And also, there's a trend with how Vivaldi develops new features with just basic functions, they implement them and then might fix them a bit (if at all) but that's it, many of the new features are just implemented and then abandoned with basic functions, the focus seems to be to implement as many new features as possible, which unfortunately has left a mark on the quality and usability of the new features. Also I have seen lately that the quality overall has degraded, there are more and more bugs. I never thought I would say this after switching to Vivaldi (Firefox user after good-old-opera was sold who waited for Vivaldi to became good enough as a daily driver), but Firefox (with all it's faults and annoyances) starts to seem as an viable option. Vivaldi should focus more on stability at the moment.
-
Mv2 are bad and you can't deny that, they can't do basic things
even WE are less crippled, though still a bad joke when compared to Firefox extensions and everyone knows it
the only browser that has even more limited extensions capabilities than Mv2 is Safari and that shouldn't surprise anyone eitherand sure, I know V is mostly focusing on "can we slash this on our checklist already?" and that's frustrating, but what else we have in the store? it's either Mozilla getting proper management or Vivaldi polishing things that desperately need attention for years, browsers are way too complex and these two are our only probable options, even if unlikely
and when it comes to content blocking in Chromium, the under the hood work is already done and open source, it just needs more configurability (which obviously can be a huge pain and resources drain and I am well aware of that) but at least the filtering engine is there and has all the required capabilities so there's much less work to do if someone decides to use it
so we have to show we expect configurability from... browser that promises it, it shouldn't be too much to ask when that's the whole point of V's existence, right?
I'm not holding my breath, I'm not keeping my hopes high, but if there's a chance let's show we need this
and Quantum is the only viable option for now, thanks to the hackability, but at some point it may lose that
-
In Vivaldi, I trust!
Thank you!
-
@preorian: The code of ublock is open source. It's entirely possible to take inspiration from it for our own implementation. It isn't a fast process, but it is much faster than reinventing everything ourselves. The only reason development on our adblocker have stalled is because I have been a bit too busy with sync as of late.
Software developer at Vivaldi since January 2017.
-
@zakius: Our solution is native. Fully implemented in C++.
Software developer at Vivaldi since January 2017.
-
@julien_picalausa Any chance we some day can have a logger like uBO to show what requests are being blocked and why? It would really help troubleshoot blocker issues
Or maybe is there a way to see some log at the moment? Maybe something in the console?
🎻Volunteer helper · Forum moderator · Sopranos tester 🛠️Troubleshooting 🐛Report a bug 📜Markdown help
🦆"With a rubber duck, one's never alone" -Douglas Adams🦆 -
@julien_picalausa if it is native it shouldn't rely on the extensions API at all, though may use some of the internal parts used to implement it, possibly the wording of the article was misleading
-
@zakius: It's not relying on the extension API. It's relying on an underlying, internal API, which is also used by the extension API.
