This is one of the better coverages of Manifest v3. All the articles that have been upvoted on various link sharing sites make the claim, or seriously imply that you won't be able to block ads any...
This is one of the better coverages of Manifest v3. All the articles that have been upvoted on various link sharing sites make the claim, or seriously imply that you won't be able to block ads any longer, which is of course not true.
This article does a good job of explaining the actual API changes and tries to guess at the reasoning behind them. I also like that he explored alternative approaches that Google could have taken instead.
To comment on some points:
Let me skip the performance topic here for reasons of brevity, with the addendum that uBlock Origin adds ~14 microseconds on average to each request it inspects - not exactly a burning performance problem
So, I'd rather he had actually explored it. uBlock Origin is probably the most efficient ad blocker, but it's not the most popular. I'd be curious to know what performance impact AdBlockPlus and others have. It's worth remembering that these blockers will also be forced into the new system, so it's a bit premature to call the performance improvement invalid.
“Apple is doing the same”
I made this argument previously, and this section makes a good case for why Apple may have done so and why it made more sense for them than for Google.
“Google ads can be trivially blocked even after this change”
I made this argument as well. This section though seems a bit... speculative. If Google Ads really move to a system of ever-moving targets to evade what's possible in DNT then I'll gladly eat my words. I think it's a pretty unlikely outcome though, and I'm not very convinced by the argument.
I don't fully land with the author on the conclusion, but I respect their methodology. This was a good article.
This one was done in February, some time after the Manifest V3 controversy started: https://whotracks.me/blog/adblockers_performance_study.html But if we are going to be realistic, adding 14...
I'd be curious to know what performance impact AdBlockPlus and others have.
But if we are going to be realistic, adding 14 microseconds to block an ad that would take from several seconds to minutes to load and drain battery and data is a really good trade-off. That's where the argument against adblockers' performance falls apart. I install uBlock Origin in all my low end devices because it makes a huge difference in performance when browsing the web.
Edit: PS: there was a little issue with the way they measured uBlock Origin's performance in that article and it was discussed in uBlock's issue tracker but I can't find the discussion right now.
Most of the conversation about the changes to the Web Request API is little better than conspiracy thinking. Essentially, Google sells ads, therefore they hate ad blockers, therefore any change...
This section though seems a bit... speculative
Most of the conversation about the changes to the Web Request API is little better than conspiracy thinking.
Essentially, Google sells ads, therefore they hate ad blockers, therefore any change that makes ad blocking more difficult must be a nefarious scheme that harms users.
The fact is that the current Web Request API gives the extension the ability to read and change everything the user does on the web, on every site. That means credentials, social media, banking information, emails, everything. It's incredibly dangerous, and extensions abuse these permissions.
I have uMatrix and uBlock Origin installed right now. I trust them, but they shouldn't need access to everything just to apply a regex to the resources the page requests.
The new Declarative Net Request API does just that, and with a lot less CPU and RAM than the current method.
If the new API makes ad blocking possible with greater security, greater speed, and lower resource utilization, then that's a win for users and a win for content blocking, even if current blockers need to be reworked to achieve that.
This is one of the better coverages of Manifest v3. All the articles that have been upvoted on various link sharing sites make the claim, or seriously imply that you won't be able to block ads any longer, which is of course not true.
This article does a good job of explaining the actual API changes and tries to guess at the reasoning behind them. I also like that he explored alternative approaches that Google could have taken instead.
To comment on some points:
So, I'd rather he had actually explored it. uBlock Origin is probably the most efficient ad blocker, but it's not the most popular. I'd be curious to know what performance impact AdBlockPlus and others have. It's worth remembering that these blockers will also be forced into the new system, so it's a bit premature to call the performance improvement invalid.
I made this argument previously, and this section makes a good case for why Apple may have done so and why it made more sense for them than for Google.
I made this argument as well. This section though seems a bit... speculative. If Google Ads really move to a system of ever-moving targets to evade what's possible in DNT then I'll gladly eat my words. I think it's a pretty unlikely outcome though, and I'm not very convinced by the argument.
I don't fully land with the author on the conclusion, but I respect their methodology. This was a good article.
This one was done in February, some time after the Manifest V3 controversy started: https://whotracks.me/blog/adblockers_performance_study.html
But if we are going to be realistic, adding 14 microseconds to block an ad that would take from several seconds to minutes to load and drain battery and data is a really good trade-off. That's where the argument against adblockers' performance falls apart. I install uBlock Origin in all my low end devices because it makes a huge difference in performance when browsing the web.
Edit: PS: there was a little issue with the way they measured uBlock Origin's performance in that article and it was discussed in uBlock's issue tracker but I can't find the discussion right now.
Thanks for the link! I'll give it a read.
Most of the conversation about the changes to the Web Request API is little better than conspiracy thinking.
Essentially, Google sells ads, therefore they hate ad blockers, therefore any change that makes ad blocking more difficult must be a nefarious scheme that harms users.
The fact is that the current Web Request API gives the extension the ability to read and change everything the user does on the web, on every site. That means credentials, social media, banking information, emails, everything. It's incredibly dangerous, and extensions abuse these permissions.
I have uMatrix and uBlock Origin installed right now. I trust them, but they shouldn't need access to everything just to apply a regex to the resources the page requests.
The new Declarative Net Request API does just that, and with a lot less CPU and RAM than the current method.
If the new API makes ad blocking possible with greater security, greater speed, and lower resource utilization, then that's a win for users and a win for content blocking, even if current blockers need to be reworked to achieve that.
Enough with the conspiracy thinking.