"Test your adblocker" websites can harm users and the adblocker ecosystem | Hacker News Reader

"Test your adblocker" websites can harm users and the adblocker ecosystem

brave.com

71 pointsReadCarlBarks25 comments

I'm a bit baffled that automated ad blocker tests exist in the first place. Historically the main difficulty in writing an ad blocker are the rules. How are you going to accurately test these rules by ... testing if they match your imagined rules.

I guess there is always an opportunity to peddle a useless metric. If you actually wanted to do this right the obvious way would be manual review: pick X websites, visit them with each ad blocker, deduct points for visible ads and lost functionality. Repeat for each ad blocker, then rank them by score. That would also give you a nice score whether ad blockers or the advertising industry is currently "winning" the ad war

Ad blockers are just the embodiment of the internet’s social convention: send whatever bits you want, and I’ll render them however I want. If somebody wants to sent me ads, I prefer not to render them. But they don’t burn my eyeballs out.

They are more like a picket fence, than a fortification with barbed wire and all that.

If a site is really persistent about circumventing my adblocker, they are basically hostile to me and I should just leave.

> If a site is really persistent about circumventing my adblocker, they are basically hostile to me and I should just leave.

Since you're not generating any revenue and are costing them bandwidth, the site likely prefers it that way...

Yes, I agreed. To be clear I’m not doing the silly thing of threatening to leave a site I’m not going to pay for. I’m describing the result that makes everybody happy.

This reminds me of how it's said that threatening to hire a lawyer is meaningless. Actually hiring a lawyer is serious.

Jordan Peterson elegantly demonsrated this recently

>They are more like a picket fence, than a fortification with barbed wire and all that.

I disagree. I consider an adblocker (Ublock specifically plus Cookie AutoDelete) like a condom for my web browser. The name adblocker is really a misnomer, since it blocks much more nefarious things like tracking and fingerprinting. The name comes from a time when the internet was much less hostile and adversarial towards its users. Additionally, browsing the internet in the EU is a miserable experience because of unrelenting and useless cookie popups, made worse when one manages cookies with CAD.

If all I had to deal with was just ads, at a reasonable volume, I wouldn't care so much.

I run ublock origin and noscript. So, my way of looking at it might be very influenced by the software I use.

But I think of these as pretty different use-cases. It is better not to run JavaScript at all from malicious sites, I think I put most of your “condom” use-case in there. Ublock is just there to make the site pretty.

> send whatever bits you want, and I’ll render them however I want

> If a site is really persistent about circumventing my adblocker, they are basically hostile to me and I should just leave

You need to pick one of these.

You either are ok when them sending you any bits they want and you render whatever you want, or not. These bits might influence your adblocker - you're free to overcome that and render it the way you want. Or not, if you're not willing to/ not capable to do that.

I don’t think that’s true. The first describes the social convention. The second describes a reasonable response to sites that don’t follow it.

I don’t think there’s anything immoral about engaging in the back and forth. I just think it isn’t worthwhile to go very far with it.

There's a few sites I visit that show first party ads. They are all related to the normal audience of the site itself, rather than based on my history. I see this as a fair exchange.

I think the article reaches the right conclusion in this matter: This kind of thing is a test of an interchangeable list of rulesets instead of a test of the blocker using them.

Isn't a high-quality ad blocker pretty self evident without a dedicated test page? Seems like these sites are directly pushing Goodhart’s Law.

adblocker ecosystem everything is a market

welll, i just use all of them then. to me, those are more real world scenario than the one they link.

Ok, where’s your adblocker test then?

Right, either release a better quality Adblock tester or stop whining? People need to test somehow.

What do you expect from an advertising blog post? Of course they're going to shit on everyone else while claiming theirs is best in class. Otherwise, what would be the point. Does anyone other than bots/scrapers read these things? This is a classic marketing ploy that's just a twist of "repeat the lie often enough it becomes truth".

They offer some testing sites at the bottom of the article: "Brave will continue to work with legitimate testing sites like https://privacytests.org and https://coveryourtracks.eff.org"

Neither of those solve the problem of “is my adblocker on”. Maybe the Brave people should have done some research on why people use these sites rather than making a long blog post explaining why they are bad, actually, while not providing any alternatives for the things people want.

"Will continue to work with"

That's like saying bulletproof glass makers working with gun makers. How can you trust either when both are "working together?"

This comes across as some sites make Brave look bad, and they are calling them out for whatever reason.

Also of note - De Beers owns the diamond certification company, which is convinient because why not.

Are you suggesting that the EFF shouldn't be trusted because Brave works with them to suggest new tests and point out broken ones?

"broken" you say? I'm curious what Brave would consider to be broken. "Gee, we found one of your tests is _incorrect_, and we think this is the right way to do it"

The article attacks the notion of acceptance testing on websites due to a nebulous fear that good != perfect and testing against the moving target of advertisements is too time consuming and potentially misleading, and instead suggests to test+rate adblockers based on a set of hand picked criteria that suits the product they're advertising. Unsurprisingly, half of the reasons are due to the advantages of adblock being first class native in their product vs a second class extension.

It sure isn't hard to be #1 in some category if you get to define it, but the comparison is kind of muddled for me when you consider the significant amount of time and effort towards maintaining and monetizing an entire platform of security vs maintaining an extension on an already popular platform. Harm reduction isn't a zero sum game.

If you really wanted to know, most of the tests for Brave's adblock engine are at https://github.com/brave/adblock-rust/tree/master/tests

Those are automated unit and integration tests with controlled filter data as inputs. That's the only practically useful kind of test for an adblocker.

Reply on news.ycombinator.com