Adblock Plus

i think it's time we start working on a community based filterset.. first we need to come up with a name, and get the list gathered, we should start fresh , i think this would be a good idea

There are already a couple things like this, like adshare. Moderation would be good, though, because that one has had some filters that are way to broad and it was once vandalized.

we'll submit any filters you like, you wanna be mod for the coolaberative filterset?

btw

a while ago i proprosed an open project at g's site butit was rejected

i think this is a good idea and might go pretty far, if you disagree.. i'll just leave.

I don't think I'd be a very good mod for it. I'm not good enough with regular expressions to understand the craziness used in Filterset.G and FiltersetP and filtersetX and stuff, and I don't want to block every ad, just the ones that annoy me.

Edit: Thinking about it a little, what's the real point of such an open filterset? We have skilled people already working on filtersets, and they're doing a good job. G has his forum where you can point out false positives and negatives. Okay, so Filterset.G is ugly and huge and all, but that approach is still good.

well that's fine.. i don't know regular expressions all that well either..

Heck, I edited something into my post.

I'm good enough with regular expressions to do whatever I've tried to, but it would take some work to understand those filtersets.

(BTW, it's not filtersetX I was thinking of. It looks like it's Filter von Dr.Evil, or fanboy, or others.)

Edit: Because my style is just to block things that annoy me, I mostly just block specific URLs, so I don't have a use for the crazy complex regular expressions.

@ubergame: A community based filterset is actually what I plan to do. Now I only need to find time

Oh, neat.

This french "community" filter has some 3k filters... (yes, I know, ABP is fast, but shorter lists are always faster. )

I also thought about something, but this is difficult, very difficult.

For the best results a database is needed, which can store the filter, ad URLs, site URLs containing these ads, additional comments and tags. Tags can contain language or filtertype (eg. "en", "nl" or "tracker" or something) so that customized filtersets can be generated.

Only when these informations are kept together, you can easily avoid false positives and block ads effectivly.
Programming such application is not very easy. But when this is developed, it is scalable and multi-editor-friendly. So more authors can make new filters; when one has no time, some other can do it. All information are in one place, so problems can be solved successfully. But the amount of editors should be limited, to keep the list clear and plain.

i mean my proposition for a coolabertive on filterset.g was set on the back burner by g himself.. i dont really know if it was rejected.. he could change his mind..

It's an interesting idea, but it would be a huge challenge need good moderation because it'd be subject to vandalism by advertisers themselves if it ever got too popular. And clear, defined goals would help too.

That said, I'd still be the first in line to help out if something like this ever started up.

@ubergame: I don't care what G does, I am certainly going to do this.

@VF: Yes, moderation is an issue. At least in the start people will only be able to suggest filters while the actual adding will be limited to several administrators. We might switch to a more open scheme later if we can find some good solution.

I don't care what G does, I am certainly going to do this.

huh?? i didn't say i was doing this because of his filterset i said i want to do this because i want to make a filterset that people can contribute too..


wlad, what do you mean by "I" am certainly going to do this? are you the guy who hijacked this thread?

No, I didn't hijack anything - the plans to do this existed for a while and have been discussed in other threads here. But I'm sorry, I shouldn't bother you with this in your thread

This sounds really interesting, Wladimir. I don't mean to rush you, but I have to ask what sort of time frame you see yourself doing this in. I mean, is it just a far-off idea for the future, or something you see yourself setting up fairly soon?