Skip to content
This repository has been archived by the owner on Sep 9, 2022. It is now read-only.

Myth: This other extension blocks more stuff!

Jump to bottom
Gitoffthelawn edited this page Jun 19, 2015 · 1 revision

TL;DR: Do not rely on a number shown on an extension badge to judge the amount being blocked, as that is not an effective measure.

For both Adblock Plus (ABP) and uBlock (and many other filtering extensions), the badge on the icon reports the number of net requests blocked by the extension.

Sometimes, for the same page, one extension can report more stuff blocked than the other, while the reality could be the opposite.

The less a blocker blocks, the more initial requests go through. The more initial requests, the more likely some of the sub-requests will need to be blocked. Consequently, sometimes you end up with more requests blocked (as shown on an extension badge), while internally, more requests were actually allowed on the web page.

Ultimately, it is the benchmarks run that report the amount being blocked, which tell the real story.

If you do not want to run a benchmark, there is this little online tool with which you can find out the requests which were not prevented from leaving your browser. To use it:

  1. Open the dev console for the page for which you want a report.
  2. Go to the Network tab.
  3. Clear the browser cache by right-clicking somewhere in the Network tab console.
  4. Force a reload of the web page.
  5. Right-click in the Network tab console.
  6. Select "Copy all as HAR".
  7. Paste the result in the text area of this online tool, and click Parse.

You will be shown the hostnames which were hit by the browser for the particular page you loaded.

For example, for the front page of http://www.cnet.com/, as of when this was written, uBlock showed 10 request blocked, while ABP showed 16 request blocked (both with a lot of filter lists). However here is what actually happened internally:

Remote servers reached:

Adblock Plus

  • dw.cbsi.com
  • cnet3.cbsistatic.com
  • cnet4.cbsistatic.com
  • fonts.cnet.com
  • urs.cnet.com
  • www.cnet.com
  • 1ab45d4854fe036a37ff-6643978b1699ef52a80b7f45a7bcfe3d.r85.cf2.rackcdn.com
  • www.googletagservices.com
  • zor.livefyre.com
  • platform.twitter.com
  • s.yimg.com
  • dw.cbsimg.net
  • geo.query.yahoo.com

uBlock:

  • dw.cbsi.com
  • cnet3.cbsistatic.com
  • cnet4.cbsistatic.com
  • fonts.cnet.com
  • urs.cnet.com
  • www.cnet.com

So uBlock caused the browser to hit significantly fewer remote servers, meaning it blocked more, and yet its badge displayed a lower number of requests blocked.

In summary, do not rely on the badge to judge how much is being blocked.

On the other hand, blocking too much may "break" some aspects of a web site, so in the end, as long as you understand what is really happening under the hood, you can make informed decisions.

Update:

To be objective and unbiased, there are only two ways one filtering tool will block more than another. The first is if it designed to actually block different content. The second is if is equipped with more filters. The above tests likely are biased in that the results are probably reflective of the tester configuring uBlock with more filtering lists than ABP. The only validity of the above tests is to demonstrate that a higher number on an extension badge does not indicate more blocking.

That said, uBlock has two distinct blocking advantages over ABP. The first is Strict Blocking. Strict Blocking allows uBlock to block a site at the highest level. ABP does not have this ability. The second is that uBlock is designed to be much more efficient than ABP; as a result, uBlock can handle many more filters than ABP while bogging down the system less.

Clone this wiki locally