11 ms·

> Their adblocker is just a fork of uBlock Origin, This does not appear to be true. Here is the github repo for their open source adblock engine written in ru

by wmitty 5y ago

> Their adblocker is just a fork of uBlock Origin, This does not appear to be true. Here is the github repo for their open source adblock engine written in rust: https://github.com/brave/adblock-rust https://github.com/brave/adblock-rust Here is a (somewhat dated) article describing it by the authors: https://brave.com/improved-ad-blocker-performance/ https://brave.com/improved-ad-blocker-performance/ > Google will take decisions that benefit their advertisement business, like making impossible to use adblockers on any Chromium based browser. Because the brave adblocker is integrated directly into the browser (ie. not an extension) the Manifest V3 limitations don't apply.

fallat 5y ago
Why not just use Ungoogled-Chromium?
- paco3346 5y ago
  I switched because Google removed the ability to log in and sync settings, history, password, etc. (I realize that in this case I'm directly giving Google my data) but it was a super nice feature. Brave's Sync v2 works decently well.
- lorlou 5y ago
  It doesn't seem to include an automatic updater.
  - k4rli 5y ago
    Simply use a package manager.
    - fastball 5y ago
      [insert link to infamous HN Dropbox comment here]
      - nyberg 5y ago
        Use GNU GUIX to manage it. It's been packaged for quite a while now
      - weird-eye-issue 5y ago
        This is the second reference to that in this thread. It's getting pretty old and I don't even think it's relevant
        andai 5y ago
        Yeah just download it over FTP bro!
- jsf01 5y ago
  Third party untrusted binaries last I checked
  - gruez 5y ago
    You can build it yourself, but even with a midrange desktop it'll take you at least an hour to build. A laptop would probably take 2-3 at least.
    - ben940830298432 5y ago
      Are you going to read the source to confirm nothing malicious was added?
      - gruez 5y ago
        There's around 4.9k lines of python code and 15.9k lines of patches. That doesn't seem that hard to scrutinize. From a threat model point of view you should be more worried about supply chain attacks from all the third party programs/libraries you have installed on your computer.
  - j-james 5y ago
    You can pull trusted binaries from OpenBuildService now.
- andai 5y ago
  I tried this a year ago. Had some trouble first downloading this (afaik the project only provided sources, not binaries, so you had to trust some random guy's website to download the .exe), then it randomly crashed within 5 minutes every time I launched it, then I deleted it.
kkoncevicius 5y ago
There are more lies in that article. This one for example is so often repeated but untrue: > Rewards is their shitty program that will replace ads displayed on websites with their own. Brave doesn't replace ads with their own. Brave ads are displayed as desktop pop-ups. They can also be easily disabled (which, surprise, the author doesn't mention because of his bias). And the idea behind Brave ads is to give you tokens which are then distributed to the content creators you engaged with. This is the default setting. Their idea is not to shovel you with ads or offer you "get rich with crypto" schemes. Idea is to block ads but still provide revenue to the content, based on how many users engage with that content. When I see people saying "Brave replaces ads with their own" I have to wonder if they have tried using Brave themselves before writing these critique articles.
- Belphemur 5y ago
  To play devil advocate. On one side, Brave come with an adblocker that will remove any ads from the website you're visiting. On the other, they provide their own ads through the reward program. So it can be seen as "replacing website ads by its own". I approve that line of reasoning, but I think that what the author meant.
  - Belphemur 5y ago
    Edit: I don't approve that line of reasoning, but I think that what the author meant.
  - kkoncevicius 5y ago
    To play the devil's devil's advocate :) Brave allows you to do whatever you want. You can see publisher ads without Brave ads. You can see Brave ads without publishers ads. You can see both. Or you can disable both. Since individual users can achieve any configuration of ads they like, to me it seems that some people are only unhappy with this because they want to push their moral stances on everyone else. Like, for example, stating that the ability to block publisher ads while enabling Brave ads is immoral and shouldn't be allowed.
  - ABCLAW 5y ago
pmurt7 5y ago
> If earning half a penny in a month is okay for you, in exchange of your privacy, because of course, they’re tracking you with Rewards, then enjoy your money. Lie. Brave doesn't track you. Your ad data never leave your machine (a bit like your bookmarks). The ad engine works privately on your computer and not on Brave server.
- ehutch79 5y ago
  If it's fetching ads, it has to 100% be sending some data to someone, who is likely able to correlate it and track you. It doesn't take much.
  - pmurt7 5y ago
    The entire ad catalog is sent on your machine and some ad engine running inside the browser decides which ads to show you. It's funny seeing all these folks nitpicking at Brave but who are fine using Google or Microsoft every day
    - ehutch79 5y ago
      I don't really care about brave either way, it's just dubious that the ads are somehow untrackable when you apparently get credit for seeing them some how?
      - jonathansampson 5y ago
        We use zero-knowledge proofs and blinded tokens to track when an ad has been viewed by a user. But there is no user data involved here. The magic of cryptography is that you can prove you viewed the ad without telling us anything about you
        x4e 5y ago
        Do you have any reading material about how you achieved this? I can't really see how zero-knowledge proofs could solve this. There is no cryptographic way to prove that software executing on a clients machine triggered a notification. Especially on Linux where an open source notification manager could be modified to reject it.
causality0 5y ago
Does integrating it into the browser have any performance benefits over using an extension?
- pmurt7 5y ago
  Brave ad blocker is written in Rust and browser extensions in JavaScript, so it should be faster
  - jonathansampson 5y ago
    Not only faster, but we aren't beholden to the APIs offered by Google and others. Manifest v3 threatened the existence of popular content-blockers like uBlock Origin. Since we are the browser, we aren't so limited. A recent example of how we are able to do more was with the introduction of CNAME blocking, which allowed us to identify when a third-party tracker had managed to be requested from a first-party URL: https://brave.com/privacy-updates-6/ https://brave.com/privacy-updates-6/.
    - causality0 5y ago
      Hey, thanks. One of my favorite computers is a Surface 3 running a Cherry Trail CPU. I tried Brave out and it's noticeably snappier on the old hardware than Firefox or Chrome.
  - x4e 5y ago
    uBlock Origin is using Web Assembly now for certain critical code. Anyway performance differences between the two have always been negligible and often fluctuating.
shilad 5y ago
The Epic Privacy Browser Team is integrating uBlock into Epic in their next update and didn't find a significant degradation in performance from any Chrome limitations, nor a significant performance improvement in Brave's implementation. Epic's mobile browsers were built on Brave/Chromium, but now that Brave has endpoint and other dependencies as mentioned it doesn't explain, it isn't possible to continue to build on them or even test them since Brave features don't work in outsider builds.
- jonathansampson 5y ago
  "…now that Brave has endpoint and other dependencies as mentioned it doesn't explain…" What unexplained endpoints/dependencies does Brave have? I believe I demonstrated otherwise (with links to external resources) here: https://news.ycombinator.com/item?id=27552530 https://news.ycombinator.com/item?id=27552530.
- nieve 5y ago
  > HTML filtering is the ability to filter the response body of HTML documents before it is parsed by the browser. > For example, this allows the removal of specific tags in HTML documents before they are parsed and executed by the browser, something not possible in a reliable manner in other browsers. This feature requires the webRequest.filterResponseData() API, currently only available in Firefox. https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-best-on-Firefox https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-b... I'm going to trust Gorhill on this one. If a significant feature _does not exist_ in Epic then the only way that it couldn't hurt performance is if it was somehow useless. I suspect the Epic people (accidentally?) didn't measure that aspect.
freediver 5y ago
I am curious why doesn’t Brave block Google ads on its standard (default) ad-blocker settings?

by wmitty 5y ago

fallat 5y ago

Why not just use Ungoogled-Chromium?

paco3346 5y ago
I switched because Google removed the ability to log in and sync settings, history, password, etc. (I realize that in this case I'm directly giving Google my data) but it was a super nice feature. Brave's Sync v2 works decently well.
lorlou 5y ago
It doesn't seem to include an automatic updater.
- k4rli 5y ago
  Simply use a package manager.
  - fastball 5y ago
    [insert link to infamous HN Dropbox comment here]
    - nyberg 5y ago
      Use GNU GUIX to manage it. It's been packaged for quite a while now
    - weird-eye-issue 5y ago
      This is the second reference to that in this thread. It's getting pretty old and I don't even think it's relevant
      - andai 5y ago
        Yeah just download it over FTP bro!
jsf01 5y ago
Third party untrusted binaries last I checked
- gruez 5y ago
  You can build it yourself, but even with a midrange desktop it'll take you at least an hour to build. A laptop would probably take 2-3 at least.
  - ben940830298432 5y ago
    Are you going to read the source to confirm nothing malicious was added?
    - gruez 5y ago
      There's around 4.9k lines of python code and 15.9k lines of patches. That doesn't seem that hard to scrutinize. From a threat model point of view you should be more worried about supply chain attacks from all the third party programs/libraries you have installed on your computer.
- j-james 5y ago
  You can pull trusted binaries from OpenBuildService now.
andai 5y ago
I tried this a year ago. Had some trouble first downloading this (afaik the project only provided sources, not binaries, so you had to trust some random guy's website to download the .exe), then it randomly crashed within 5 minutes every time I launched it, then I deleted it.

kkoncevicius 5y ago

There are more lies in that article. This one for example is so often repeated but untrue: > Rewards is their shitty program that will replace ads displayed on websites with their own. Brave doesn't replace ads with their own. Brave ads are displayed as desktop pop-ups. They can also be easily disabled (which, surprise, the author doesn't mention because of his bias). And the idea behind Brave ads is to give you tokens which are then distributed to the content creators you engaged with. This is the default setting. Their idea is not to shovel you with ads or offer you "get rich with crypto" schemes. Idea is to block ads but still provide revenue to the content, based on how many users engage with that content. When I see people saying "Brave replaces ads with their own" I have to wonder if they have tried using Brave themselves before writing these critique articles.

Belphemur 5y ago
To play devil advocate. On one side, Brave come with an adblocker that will remove any ads from the website you're visiting. On the other, they provide their own ads through the reward program. So it can be seen as "replacing website ads by its own". I approve that line of reasoning, but I think that what the author meant.
- Belphemur 5y ago
  Edit: I don't approve that line of reasoning, but I think that what the author meant.
- kkoncevicius 5y ago
  To play the devil's devil's advocate :) Brave allows you to do whatever you want. You can see publisher ads without Brave ads. You can see Brave ads without publishers ads. You can see both. Or you can disable both. Since individual users can achieve any configuration of ads they like, to me it seems that some people are only unhappy with this because they want to push their moral stances on everyone else. Like, for example, stating that the ability to block publisher ads while enabling Brave ads is immoral and shouldn't be allowed.
- ABCLAW 5y ago

pmurt7 5y ago

> If earning half a penny in a month is okay for you, in exchange of your privacy, because of course, they’re tracking you with Rewards, then enjoy your money. Lie. Brave doesn't track you. Your ad data never leave your machine (a bit like your bookmarks). The ad engine works privately on your computer and not on Brave server.

ehutch79 5y ago
If it's fetching ads, it has to 100% be sending some data to someone, who is likely able to correlate it and track you. It doesn't take much.
- pmurt7 5y ago
  The entire ad catalog is sent on your machine and some ad engine running inside the browser decides which ads to show you. It's funny seeing all these folks nitpicking at Brave but who are fine using Google or Microsoft every day
  - ehutch79 5y ago
    I don't really care about brave either way, it's just dubious that the ads are somehow untrackable when you apparently get credit for seeing them some how?
    - jonathansampson 5y ago
      We use zero-knowledge proofs and blinded tokens to track when an ad has been viewed by a user. But there is no user data involved here. The magic of cryptography is that you can prove you viewed the ad without telling us anything about you
      - x4e 5y ago
        Do you have any reading material about how you achieved this? I can't really see how zero-knowledge proofs could solve this. There is no cryptographic way to prove that software executing on a clients machine triggered a notification. Especially on Linux where an open source notification manager could be modified to reject it.

causality0 5y ago

Does integrating it into the browser have any performance benefits over using an extension?

pmurt7 5y ago
Brave ad blocker is written in Rust and browser extensions in JavaScript, so it should be faster
- jonathansampson 5y ago
  Not only faster, but we aren't beholden to the APIs offered by Google and others. Manifest v3 threatened the existence of popular content-blockers like uBlock Origin. Since we are the browser, we aren't so limited. A recent example of how we are able to do more was with the introduction of CNAME blocking, which allowed us to identify when a third-party tracker had managed to be requested from a first-party URL: https://brave.com/privacy-updates-6/ https://brave.com/privacy-updates-6/.
  - causality0 5y ago
    Hey, thanks. One of my favorite computers is a Surface 3 running a Cherry Trail CPU. I tried Brave out and it's noticeably snappier on the old hardware than Firefox or Chrome.
- x4e 5y ago
  uBlock Origin is using Web Assembly now for certain critical code. Anyway performance differences between the two have always been negligible and often fluctuating.

shilad 5y ago

The Epic Privacy Browser Team is integrating uBlock into Epic in their next update and didn't find a significant degradation in performance from any Chrome limitations, nor a significant performance improvement in Brave's implementation. Epic's mobile browsers were built on Brave/Chromium, but now that Brave has endpoint and other dependencies as mentioned it doesn't explain, it isn't possible to continue to build on them or even test them since Brave features don't work in outsider builds.

jonathansampson 5y ago
"…now that Brave has endpoint and other dependencies as mentioned it doesn't explain…" What unexplained endpoints/dependencies does Brave have? I believe I demonstrated otherwise (with links to external resources) here: https://news.ycombinator.com/item?id=27552530 https://news.ycombinator.com/item?id=27552530.
nieve 5y ago
> HTML filtering is the ability to filter the response body of HTML documents before it is parsed by the browser. > For example, this allows the removal of specific tags in HTML documents before they are parsed and executed by the browser, something not possible in a reliable manner in other browsers. This feature requires the webRequest.filterResponseData() API, currently only available in Firefox. https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-best-on-Firefox https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-b... I'm going to trust Gorhill on this one. If a significant feature _does not exist_ in Epic then the only way that it couldn't hurt performance is if it was somehow useless. I suspect the Epic people (accidentally?) didn't measure that aspect.

freediver 5y ago

I am curious why doesn’t Brave block Google ads on its standard (default) ad-blocker settings?