5 ms·

I wonder why content website don't use Canvas rendering in order to block all ad-blockers. You either see the canvas, with ads in it or don't see anything at al

by msoad 11y ago

I wonder why content website don't use Canvas rendering in order to block all ad-blockers. You either see the canvas, with ads in it or don't see anything at all. There need to be a compatible ad-network as well. Accessibility is going to be an issue but if you can solve all of that there is a startup idea right there!

SquareWheel 11y ago
As a user, that would certainly turn me off. The entire page being in canvas fundamentally breaks native functionality. Text selection, copy/paste, navigation keys. Most of these could be simulated but it would never feel truly native. Heck, I get annoyed when on Snopes simply because I can't select any text (which I use to keep my place in the article).
jacquesm 11y ago
It's interesting that you hit on this solution which most likely has been debated in many a boardroom of companies that see these ad blocker figures ever on the rise. The problem is that this 'solution' is not really a solution. It degrades the web to a fraction of its original use case and it would indicate a drastic departure from the open standards that got us here in the first place. In a war - if you want to end it - the idea is to de-escalate. A reasonable first step from a property such as wired would be to demand from their ad-tech providers that they get rid of any tracking code and other trickery in their ad tags and that the ads served would be simple imagery rather than flash and/or javascript ads that leak an enormous pile of sensitive data about you to the 'supply side platforms'. But since that would go hand-in-hand with a reduction in effectiveness they would then have to de-rate their ad prices or they would have to live with less income. It appears that wired would rather self destruct or at a minimum force users to convert directly at rate card prices (a win for them if they do in signfificant numbers) rather than go down this route. Time will tell if it is the winning move.
dingo_bat 11y ago
How would that be different from all-Flash websites of the past? It would be quite ironic to kill off Flash and return to the same with HTML5.
njharman 11y ago
There is more value in tracking, web bugs, etc. then actually showing you the ad.
quotemstr 11y ago
Computer vision has made great strides lately; human-level face and object recognition can run on a laptop's GPU in real time. Segmenting a rendered webpage, classifying regions as ad- or non-ad, and overwriting (in a way invisible to your canvas renderer the former, by comparison, is a walk in the park. The client will always win. Browsers are user agents, not server agents. https://www.youtube.com/watch?v=1C-2dKe9NjI https://www.youtube.com/watch?v=1C-2dKe9NjI
pdkl95 11y ago
As I've said in the past here on HN, the key feature for this will be webassembly. When it becomes possible to easily compile a "custom" version of freetype and do all the rendering into a canvas element, and send the entire "webapp" (really: document with ads) down in a binary blob, everybody that currently depends on advertising will switch to it.

by msoad 11y ago

SquareWheel 11y ago

As a user, that would certainly turn me off. The entire page being in canvas fundamentally breaks native functionality. Text selection, copy/paste, navigation keys. Most of these could be simulated but it would never feel truly native. Heck, I get annoyed when on Snopes simply because I can't select any text (which I use to keep my place in the article).

jacquesm 11y ago

It's interesting that you hit on this solution which most likely has been debated in many a boardroom of companies that see these ad blocker figures ever on the rise. The problem is that this 'solution' is not really a solution. It degrades the web to a fraction of its original use case and it would indicate a drastic departure from the open standards that got us here in the first place. In a war - if you want to end it - the idea is to de-escalate. A reasonable first step from a property such as wired would be to demand from their ad-tech providers that they get rid of any tracking code and other trickery in their ad tags and that the ads served would be simple imagery rather than flash and/or javascript ads that leak an enormous pile of sensitive data about you to the 'supply side platforms'. But since that would go hand-in-hand with a reduction in effectiveness they would then have to de-rate their ad prices or they would have to live with less income. It appears that wired would rather self destruct or at a minimum force users to convert directly at rate card prices (a win for them if they do in signfificant numbers) rather than go down this route. Time will tell if it is the winning move.

dingo_bat 11y ago

How would that be different from all-Flash websites of the past? It would be quite ironic to kill off Flash and return to the same with HTML5.

njharman 11y ago

There is more value in tracking, web bugs, etc. then actually showing you the ad.

quotemstr 11y ago

Computer vision has made great strides lately; human-level face and object recognition can run on a laptop's GPU in real time. Segmenting a rendered webpage, classifying regions as ad- or non-ad, and overwriting (in a way invisible to your canvas renderer the former, by comparison, is a walk in the park. The client will always win. Browsers are user agents, not server agents. https://www.youtube.com/watch?v=1C-2dKe9NjI https://www.youtube.com/watch?v=1C-2dKe9NjI

pdkl95 11y ago

As I've said in the past here on HN, the key feature for this will be webassembly. When it becomes possible to easily compile a "custom" version of freetype and do all the rendering into a canvas element, and send the entire "webapp" (really: document with ads) down in a binary blob, everybody that currently depends on advertising will switch to it.