The counter on the disconnect icon eventually ticks up to 936. It's easy to see the counter does not represent the number of distinct tracking sources (domain names), because there's easily less than 100 of them. Disconnect is just showing a useless measurement, which is unfortunate for it's credibility.
It looks as if it represents the total number of requests by trackers. This is clearly unrepresentative when you have multiple requests to a single "content" 3rd party, generated naturally by a list of "content" items. I think it's happening with "gravatar", which may be the single best example of this.
Because of the large number, the ticking on the counter also appears bogus. The rate is completely steady, and there's no network traffic going on that it's representing. The page loads fully in a few seconds, but the counter still keeps ticking for tens of seconds.
Try this test page:
https://parahumans.wordpress.com/2013/10/12/interlude-29/
The counter on the disconnect icon eventually ticks up to 936. It's easy to see the counter does not represent the number of distinct tracking sources (domain names), because there's easily less than 100 of them. Disconnect is just showing a useless measurement, which is unfortunate for it's credibility.
It looks as if it represents the total number of requests by trackers. This is clearly unrepresentative when you have multiple requests to a single "content" 3rd party, generated naturally by a list of "content" items. I think it's happening with "gravatar", which may be the single best example of this.
Because of the large number, the ticking on the counter also appears bogus. The rate is completely steady, and there's no network traffic going on that it's representing. The page loads fully in a few seconds, but the counter still keeps ticking for tens of seconds.