Loading…
µBlock 0.6.6.0-rc.0 is phenomenal (Not an issue) #264
5 participants
How much is the "page load" decreased with this on?
if you use ublock is it necessary to have disconnect also or can I remove it ?
@ferronrsmith Nothing is a necessity for uBlock to "work".
@kurtextrem I don't know really. Not pulling and waiting for resources from the network will surely contribute to reduce page load time (and bandwidth), that's just what I can tell. I suppose on bloated page this won't even be noticeable.
@ferronrsmith Keep it. To find out whether it's worth keeping your other blocker, try a diff of benchmark data to find out what is not blocked by one but blocked by the other. Here is what I get for uBlock/Disconnect with latest benchmark (which I need to update once 0.6.6.0 is out): https://www.diffchecker.com/y5gj3ed5. The diff shows what was not blocked.
As you can see, uBlock blocked things Disconnect didn't, and Disconnect blocked things uBlock didn't. Selecting "Fanboy's Anti-Facebook" in uBlock will take care of facebook.com/net that was taken care by Disconnect. We can see uBlock "blocked" ajax.googleapis.com (not really block, it's the local mirroring), whereas these were not blocked by Disconnect. Etc.
The "206 lines added. 91 lines deleted" at the top means uBlock blocked 206 things that Disconnect didn't block, and Disconnect blocked 91 things that uBlock didn't block.
Maybe I will take time today to create these diffs for reference. One thing I expect this will show is that keeping ABP with uBlock is pointless.
Update: uBlock/ABP: https://www.diffchecker.com/4c5aox70. I have to say I was surprised to find out googletagservices.com is not blocked by ABP-EasyList-EasyPrivacy. This was taken care in uBlock by local mirroring. Also, google-analytics.com is not blocked by ABP-EasyList-EasyPrivacy. This is taken care by "Peter Lowe's" in uBlock.
@ferronrsmith Page loading time has decreased a lot, there's a huge difference between the last Stable version of uBlock and this RC0 version. At least my Chromium opens pages a lot faster.
Great job indeed gorhill.
I don't recommend Disconnect, even Chrome Web Store starts to block it in the future:
@ferronrsmith Actually, I need to correct my earlier "keep it". If I use a diff of uBlock (gorhill)/Disconnect, suddenly this shows there is no point keeping Disconnect: https://www.diffchecker.com/cc8o7ymk.
"uBlock (gorhill)" is uBlock with my personal selection of lists. Frankly these extra lists do not cause much site breakage in my experience. I know people with absolutely zero tolerance for site breakage, and yet they find these more aggressive settings all fine.
Great Idea with Local Mirroring concept!
I have a question regarding how it works with below scenarios -
- How often the resources cached with local mirroring are flushed? I mean if the website offers an updated version of the resource, shouldn't user be downloading the latest resource?
- Would it add any benefit if a user browses under Incognito mode? Would the resources be still cached?
Thanks!!
@harshanvn Currently they are flushed when the cumulated size reaches 6MB, then least recently used resources are flushed until 5MB is reached. I have a TTL (time-to-live) parameter in there, but not used yet. I have to study more about this (hence the experimental). Note that local mirroring targets resources which are deemed reasonably immutable. So when querying ajax.googleapis.com for jquery 1.7.1, it's safe to assume the result won't change, ever. Re. incognito, no special treatment, so yes, what would normally cache in non-incognito still will be cached (or else one would actually end up being less incognito from the external internet point-of-view, i.e. you IP would show up in more server logs).
I ran a new benchmark with uBlock 0.6.6.0-rc.0 and ABP 1.8.5 with the same lists (except Peter Lowe's because there is no easy convenient way to install it in Adblock Plus -- so I assume users are likely to not bother).
Here is the resulting diff: https://www.diffchecker.com/5z91i47m
In red what ABP blocked which was not blocked by uBlock.
In green what uBlock blocked which was not blocked by ABP.
I suspect the only two requests not blocked by uBlock which were blocked by ABP is just the result of the page content changing between the time I benchmarked uBlock and ABP.
Filter lists:
- ABP: Out of box settings + EasyPrivacy, Malware domains, Fanboy's Social Blocking List, Anti-ThirdpartySocial + "Acceptable ads" disabled
- uBlock: Out of box settings + Anti-ThirdpartySocial + Local mirroring enabled and primed (as it would be through normal use).
What local mirroring accomplished in the above benchmark... Prevented connections to:
2mdn.netajax.googleapis.comcloudflare.comfonts.googleapis.comgoogletagservices.comgstatic.comjanrain.comjquery.comtwitter.com
And since the remote resources were available locally, preventing the connection didn't cause page breakage.
I wonder how that will work out with "jquery-latest" and ajax.googleapis.com/ajax/libs/jquery/2/jquery.min.js (= loads the latest 2.x.x version available). But nevertheless great feature. Really looking forward to it.
That comparison was impressive, very interesting to see how adblocker's are blocking stuff and how much they let ads through. Keep up your great work gorhill.
As the title says, phenomenal work gorhill.