Please don't try it if you don't know how to remove/disable subscriptions, I'm not 100% sure if the list is ready for "regular" users yet. For everyone else, here is EasyList-based
EasyNext: [url=abp://subscribe/?location=https%3A%2F%2Fares2.org%2Fabp%2Feasynext.txt&title=EasyNext%20-%20Alpha]Subscribe to EasyNext - Alpha[/url] / View
The results are pretty amazing imho, just try a site with ads that you know aren't blocked by the common subscriptions (no matter what language, content, etc) and see for yourself.
Of course there is still a chance the list fails, so please, please tell me the site where this happens so I can try to fix it.Full story: For the last few months or so, I've been trying to find the weaknesses in the existing Adblock Plus subscriptions and here is the most important problem that can be identified: None of the lists, neither on it's own, nor combined, can live up to my and many other user's expectations of blocking all ads. Of course we are trying very hard to get close to that goal, but the reality is that so many ads that slip through, either because they simply never get reported or even just because they do not fit the subscriptions policy, strategy, range of target sites, etc. Now that observation is pretty obvious and most of us know it, the interesting part is to find out what can be done about that - in case such a strategy exists.
The most common filter subscriptions nowadays use the same set of basic strategies (block by "bad" keywords in requests, block adservers, hide "bad" html elements and lastly block/hide per-site). This system as it is used right now, undeniably has some disadvantages so the goal was to solve those problems while not creating any new ones. Also, I tried to limit myself to the possibilities Adblock Plus already offers, rather than requiring to modify it or write a completely new program (I probably couldn't do that anyway
). With that in mind, I started to randomly test some new strategies, checked the results (mostly empirically) - and usually failed to achieve anything. Then I tried to put a little more stucture in the tests (like setting up a list of testsites with desired results, evaluate the results, comparing them to each other, use the best ones to check sites outside of the test array, etc) - and eventually had some success!I will give a detailed explanation of all the new approaches asap, but it could take some time to convert my "notes" to understandable text.
