HN Reader

There is an unwritten social contract here. Websites are willing to host and organise a vast number of content because that'll attract an audience for ads. If there are too may freeloaders resisting the ads then services won't host the content, and on the path to that the freeloaders are really just leeching off a system in an entitled way (unless their goal is to destroy the services they use in which case good on them for consistency and for picking a worthy target).

If people aren't going to be polite and accept that contract then fine, enforcement was always by an honour system. But strategically if a service's social contract doesn't work for someone then they shouldn't use that service - they'd just be feeding the beast. They should go make their own service work or investigate the long list of alternative platforms.

https://techcrunch.com/2022/12/22/fbi-ad-blocker/

https://web.archive.org/web/20230219020056/https://www.ic3.g...

And attention and privacy.

This notice is a great idea.

I might remove the "like" from the notice, since "uBlock Origin" is good, but some others are questionable or even outright malware.

BTW, note that the `ublockorigin.com` Web site that is linked to isn't by Raymond Hill, leader of uBlock Origin. It looks well-intended, and is nicely polished UX, but good practice would be to be careful (since it doesn't appear to be under Hill's control, and is an additional point of potential compromise in what would be very valuable malware). Hill seems to operate from <https://github.com/gorhill/uBlock>. One link that isn't too bad to view <https://github.com/gorhill/uBlock/blob/master/README.md>. Another that isn't great but OK is <https://github.com/gorhill/uBlock/wiki>.

Before that point I'd already spotted that limitation, but there might be an easy solution: get a domain added to a common block list used by DNS based blockers. If you get the right content from a resource on a host with that name (or the other test passes, so we test for both forms of blocker) show the message.

Of course there will be false positives if the page goes down or if they're is some other network issue, but no test like this will be perfect.

Anyone want to save me the research to find out the easiest way to get a domain on the lists? I have no objection to sacrificing a few £ per year on a name to use and I've got spare resource to serve the pile of tiny requests that'll go through because people aren't running a blocker.

EDIT: as a secondary note, I wouldn't just flip between “display:none” and “display:block” on one element upon detection result. That might cause visual disturbance in many page layouts as things load. I would leave a block of the same positioning and size properties in the flow in either case, either blank or with a message like “You'll be pleased to know that your ad blocker seems to be working.”, perhaps leaving the space blank (but still in the flow with the same dimensions) initially so an incorrect message isn't displayed if something (scripting being disabled client-side for instance) stops the tests running at all.

Perhaps only enables js when user clicks something.

We do not have to feel guilty to act against them.

Btw, yesterday Chromium told me Ublock Origin is no longer supported. Well, thank you, now I know why I wasn't using Chromium for anything other than MS365 stuff. It's working just fine on Firefox.

[1] https://home.cern/news/news/computing/computer-security-bloc...

This isn't really true. I ran an ad-supported site at one point with my content, just a small banner at the top of each page. The ads paid for a significant portion of my monthly rent. Getting a few dollars from the occasional viewer would not, since 99.99+% of people are not going to do that.

I don't like viewing ads, but let's not pretend like they don't make money for content creators. They absolutely do.

Of course highlighting this fact that the presence of an adblocker is detectable, unfortunately only results in escalating the cat-and-mouse game further.

I have also considered popularising a script that replaces the whole page's content with "JavaScript detected, please disable it to view this content and improve your security".

It's like a leech, and they want you to think it's a symbiotic relationship.

Bug report: There's a typo in the actual popup as shown to me, it says "extention". Consistently enough, the typo is present in the code snippet in the article:

    if (!document.cookie.includes("notice-shown")) {
        document.getElementById("ad-note-hidden").id = 'ad-note';
        document.getElementById("ad-note-content-wrapper").innerHTML = "No adblocker detected. " + 
        "Consider using an extention like <a href=https://ublockorigin.com/>uBlock Origin</a> to save time and bandwidth." +
        " <u onclick=hide()>Click here to close.</u>";
    }

Some services claim to turn "anonymous" visitors into actual email addresses (and some other basic info), likely via identity graphs (IP/device/hashed IDs).

I've heard of cases where people are getting outreached (via email) after just visiting a product website, even with an ad blocker on, using a private browser (Brave or similar).

Opensend is one example. They're pretty open about it in their FAQ [1].

[1] https://www.opensend.com/faq

That’s a pretty crazy statement. How often do you see loading a CSS stylesheet fail to load? Most sites are completely unusable without their stylesheets and I don’t recall the last time I saw a stylesheet fail to load.

It's also deeply paternalistic: Even if it is meant well - and I assume that's the case here - it implies the site operator knows better than the user what is good for them.

Finally, this will also lower the guards of less technical users for installing random plug-ins on website demand.

From a subjective gut feeling: Please do not do this. Let people decide what they need, and what they don't need.