3 ms·

That's likely a good metric on whether the page developer is exceeding reasonable bounds. If their content looks like an ad as detected by the best, hardened, m

by tomrod 5y ago

That's likely a good metric on whether the page developer is exceeding reasonable bounds. If their content looks like an ad as detected by the best, hardened, most widely used protection software, it's reasonable that the developer should reconsider the implementation or feature altogether.

shadowgovt 5y ago
I've had uBlock reject an image served from my server because it was a particular size and had the substring "ad" in its URL (the image names were just GUIDs, and "ad" is, of course, valid hexadecimal). There's a two-way communication problem with expecting ad blockers to work on arbitrary sites without breaking the sites; they aren't a web standard so developers can't code against them, and there's no way that blockers can predict every valid configuration. False positives will, of course, occur, and since code is changing continuously in both pages and the blocker, making them work with each other is a moving target. When someone reports an issue with my sites, "disable uBlock and try again" is the first thing I recommend. Failing to do so is like trying to debug an issue in MacOS back in the day without asking the user to first disable extensions.
- tomrod 5y ago
  Sounds like we agree in principle and on assessment. Perhaps the daylight is that I believe users should be able to anticipate that sites will work under common configurations like use of ublock, which puts the onus on the dev.
  - shadowgovt 5y ago
    Since uBlock isn't a web standard (and runs on a constantly-shifting set of heuristics), it's unrealistic to expect devs to chase it successfully. To be sure, they will chase it to maximize user eyeballs, but expecting that to be the correct course of action is like expecting devs to adapt their sites to IE6 quirks. I thought we went through a whole standard development process to get away from the need to handle user-agent issues.
    - tomrod 5y ago
      > I thought we went through a whole standard development process to get away from the need to handle user-agent issues. We did, but it didn't address the ad space sufficiently. Hence the need for ublock, which I suspect is sufficient to be the de facto standard.

by tomrod 5y ago

shadowgovt 5y ago

I've had uBlock reject an image served from my server because it was a particular size and had the substring "ad" in its URL (the image names were just GUIDs, and "ad" is, of course, valid hexadecimal). There's a two-way communication problem with expecting ad blockers to work on arbitrary sites without breaking the sites; they aren't a web standard so developers can't code against them, and there's no way that blockers can predict every valid configuration. False positives will, of course, occur, and since code is changing continuously in both pages and the blocker, making them work with each other is a moving target. When someone reports an issue with my sites, "disable uBlock and try again" is the first thing I recommend. Failing to do so is like trying to debug an issue in MacOS back in the day without asking the user to first disable extensions.

tomrod 5y ago
Sounds like we agree in principle and on assessment. Perhaps the daylight is that I believe users should be able to anticipate that sites will work under common configurations like use of ublock, which puts the onus on the dev.
- shadowgovt 5y ago
  Since uBlock isn't a web standard (and runs on a constantly-shifting set of heuristics), it's unrealistic to expect devs to chase it successfully. To be sure, they will chase it to maximize user eyeballs, but expecting that to be the correct course of action is like expecting devs to adapt their sites to IE6 quirks. I thought we went through a whole standard development process to get away from the need to handle user-agent issues.
  - tomrod 5y ago
    > I thought we went through a whole standard development process to get away from the need to handle user-agent issues. We did, but it didn't address the ad space sufficiently. Hence the need for ublock, which I suspect is sufficient to be the de facto standard.