Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

The iOS 9 Content Blocker thread

Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
gaanee said:
@avatar-adg: Just a suggestion: Is it possible to notify the user that ads/tracking elements are unblocked similar to accessing https: sites on unsecured connection? This way, it will be transparent to the user when ads/tracking are active and the user can decide whether to continue on that website or move on.
Click to expand...

I like this idea.
 
The question is: does unblocking lets ads/trackers through thus negating the purpose of content blocker? This may be important for privacy concerned users.
After it's unbocked, does Adguard hides unblocked ads/trackers? Is there any visible clue to know something was unblocked intentionally?

avatar-adg said:
There is no simple answer, these things are always different and site-specific.
Click to expand...
 
gaanee said:
The question is: does unblocking lets ads/trackers through thus negating the purpose of content blocker? This may be important for privacy concerned users.
Click to expand...

Content blocker does not have any "purpose", it is just a tool that allows you to block requests or hide page elements. Filter lists have a "purpose".

I'll just copy/paste here the policy we use.

https://blog.adguard.com/en/adguard-news/adguard-filter-policy.html said:
When discussing Adguard ad filters, there is often one topic that causes disputes – whether shall Adguard block this or that ad. By framing the rules we stick to special criteria. Now it’s time to publish them. Our criteria are much similar to EasyList Policy, which we find correct and appropriate. Nevertheless, we have put changes to some points.

Ad filters
Those are actually language filters, including: Russian, English, German, Dutch, Spanish, Portuguese, Japanese.

What shall these filters block?
  • The filters shall block ads everywhere possible.
    You can find the definition of “ad” right here.
  • Ads shall be blocked regardless of reasons and goals of such ad.
  • We will block ads caused by malicious apps or extensions that inject ads. Please note that we do it only on condition that you will specify how to install this app or extension.
Limitations and Exceptions
If a rule is subject to the list of limitations described below, then it won’t be added to main filters.

  • Websites’ own ads are not to be removed on purpose. On the other side it shouldn’t be unblocked, because such unblocking can cause displaying of third-party ads.
  • Rules that are specific to the particular website will be added only if the website has sufficient traffic.
    Traffic is determined by open statistics (if it has such) or in other ways.
    Website’s traffic is considered sufficient when it has 30 thousands visitors a month.
  • Anti-adblock scripts will be blocked only if they limit functionality of the website or
    interfere in its use.
  • Rules that often cause problems with work of some websites will be deleted.
Spyware filter
What shall be blocked with this filter?
  • This filter shall block all counters that collect personal data of users.
Limitations and Exceptions
If a rule is subject to the list of limitations described below, then it won’t be added to this filter.

  • Rules that cause problems with functioning of websites will be removed.
  • Rules that are specific to the particular website will be added only if the website has sufficient traffic.
    Traffic is determined by open statistics (if it has such) or in other ways.
    Website’s traffic is considered sufficient when it has 30 thousands visitors a month.
Social media filter
What shall be blocked with this filter?
  • Filter shall block social media widgets (on third-party websites) such as: “Like” and “Share” buttons, recommendation widgets and more.
Limitations and Exceptions
If a rule is subject to the list of limitations described below, then it won’t be added to this filter.

  • Widgets that are a part of website’s functionality are not blocked.
    Such as “Comments”, “Embedded Post”, “Surveys” widgets or authorization via social networks.
  • Links to website’s communities in social networks are not blocked.
  • Rules that cause problems with functioning of websites will be removed.
  • Rules that are specific to the particular website will be added only if the website has sufficient traffic.
    Traffic is determined by open statistics (if it has such) or in other ways.
    Website’s traffic is considered sufficient when it has 30 thousands visitors a month.
Quality requirements for filtering rules
  • The rules should be as efficient as possible in terms of performance.
  • Exception rules shall be specified at most, in order to avoid unnecessary unblocking.
Click to expand...

gaanee said:
After you unbock, does Adguard hides unblocked ads/trackers?
Click to expand...

Again, this is always site-specific. Usually the ads are blocked or hidden, so you won't see it.

A few real life examples.

1. macworld.com - site javascript is broken by blocking two requests to a known tracking servers.
Unblocking these two requests does not allow any ads through.

2. info.singtel.com - site is broken by blocking request to visualwebsiteoptimizer.com. This is a known A/B testing tool. Unblocking it does not allow any ads through.

3. hulu.com (i am not sure if it's possible to watch hulu from Safari on iOS, just telling how it is on desktop) - blocking ads breaks the video player, so the ads are allowed.
 
avatar-adg said:
A few real life examples.

1. macworld.com - site javascript is broken by blocking two requests to a known tracking servers.
Unblocking these two requests does not allow any ads through.
Click to expand...
I just visited Macworld using another content blocker and none of the links work, likely due to the broken JavaScript you mentioned. The site is effectively useless. The content blocker is working exactly as intended.

I would prefer the site to be useless, or fail to load, rather than unblocking requests to known tracking servers. The purpose of a content blocker is to block undesired content, correct?

So, the best way for a webmaster to force developers to unblock requests is for them to load JavaScript that breaks due to the behavior of the content blocker? You may not realize it, but you're teaching webmasters how to circumvent the purpose of a content blocker.

A broken website is not the fault of the content blocker.. and it is not your problem.

I realize that you want to provide the best web experience for your users, but it is my opinion that crippling your content blocker is not in our best interests.
 
Last edited:
tosbsas said:
Yes with autoupdate - awesome
Click to expand...

Yeah so speaking of that I did speak with the dev on Twitter and there's a known issue where if you go in and try to update via the Magic app itself (i.e. go into the app, go to Advanced Settings and select Update Blocklist) it will sit on the updating... screen forever.

Also, you have to uninstall the current version of Magic completely and then install the new version that's in Test Flight; otherwise the app will force close every time you try to start it. Both are known and already resolved in upcoming updates as I understand it.
 
gaanee said:
@avatar-adg when the next Adguard update will be released and which new features are coming?
Click to expand...

We've delayed it a bit, version 1.0.3 will be pushed to review tomorrow.
Review usually takes about a week.

Changelog:

[Added] Edit mode on filters screen
https://github.com/AdguardTeam/AdguardForiOS/issues/30

[Added] Video tutorial on how to manage Adguard filtering from Safari
https://github.com/AdguardTeam/AdguardForiOS/issues/39

[Added] Simplified Filters mode. This is an experimental feature. For now it just disables all-sites-wide CSS rules that improves rendering a bit. In future we will change that.
https://github.com/AdguardTeam/AdguardForiOS/issues/45
https://github.com/AdguardTeam/AdguardForiOS/issues/44

[Changed] Improved content blocker rules, ~30% performance boost.
https://github.com/AdguardTeam/AdguardForiOS/issues/49

[Fixed] A bug responsible with filters getting out of sync with the server while doing a background update.
https://github.com/AdguardTeam/AdguardForiOS/issues/51

[Fixed] A bug in rules converter which didn't allow to enable some filters (like RUAdList):
https://github.com/AdguardTeam/AdguardForiOS/issues/57

Also some minor changes and bugfixes:
https://github.com/AdguardTeam/AdguardForiOS/issues/34
https://github.com/AdguardTeam/AdguardForiOS/issues/26
https://github.com/AdguardTeam/AdguardForiOS/issues/38
https://github.com/AdguardTeam/AdguardForiOS/issues/46
https://github.com/AdguardTeam/AdguardForiOS/issues/54
 
Found this review site.

https://brooksreview.net/2015/12/content-blockers/#more-19547

Photo-2015-12-02-20-02-PX2UCVfzqeP.jpg
 
This chart alone is misleading. If you read the review - Eluo broke most sites, so being fastest is useless if the site is broken. Same for Script Scrap broke many sites.
I think this chart should include only the content blockers which worked on all test sites, that would be a fair comparison. Otherwise, these numbers don't mean anything if the site failed to load.
What's the point in saving 75% data and 64% faster if the webpage didn't load? Faster at loading a blank page or broken site? You will end up spending more time reloading the page or disable/switch content blockers.

I think the best content blocker is the one which loads most of the websites and notifies you if it had to unblock any content for the page to load and giving you an option to block.


funkymonkeyboy said:
Found this review site.

https://brooksreview.net/2015/12/content-blockers/#more-19547

Photo-2015-12-02-20-02-PX2UCVfzqeP.jpg
Click to expand...
 
Last edited:
gaanee said:
This chart alone is misleading. If you read the review - Eluo broke most sites, so being fastest is useless if the site is broken. Same for Script Scrap broke many sites.
I think this chart should include only the content blockers which worked on all test sites, that would be a fair comparison. Otherwise, these numbers don't mean anything if the site failed to load.
What's the point in saving 75% data and 64% faster if the webpage didn't load? Faster at loading a blank page or broken site? You will end up spending more time reloading the page or disable/switch content blockers.

I think the best content blocker is the one which loads most of the websites and notifies you if it had to unblock any content for the page to load and giving you an option to block.
Click to expand...
I tent to agree. As I have noted before, I will freely admit that Purify feels a lot faster than many of the other content blockers I have tried. However, it broke functionality in far too many of the websites I visited for it to be of value to me.

Also, we may eventually reach a point of diminishing returns when it comes to speed. Let's say that I was able to have a website loaded up in roughly 5 seconds; using the graph above, that means I would maybe have the website loaded up in AdMop by let's say 3.2 seconds (roughly 36%ish percent if eye-balling/guessing from the graph above) and Purify would have it loaded up in 2 seconds (roughly 60%ish using the same metric.) Yes, Purify would indeed be faster but would most people notice a difference of 1.2 seconds? Maybe some would, but I feel like most wouldn't unless they were timing it. If the difference were more pronounced, like 10 seconds vs 5 seconds, but realistically we probably would not see differences that large unless the blocker was utterly incompetent.

I will totally admit that speed is less important to me, personally, than some others and this is a subjective issue. I would much rather a site work properly and be a little slower than have it load fast but broken, which is what I found happened a lot while using Purify. I ended up having to whitelist a lot of sites I regularly frequented just in order to have the content work properly. That's how I settled on my current blocker of choice, as it is fast enough for my needs while also working on the sites I visit the most.
 
Max(IT) said:
no one else ?
Click to expand...
Way back when I tried doing that for a little while. Think I mixed Adblock Fast and Adblocker (now Umbra) together and I didn't notice any kind of appreciable hit to performance. Haven't felt the need to use two at the same time anymore though.
 
I was using Purify and had many issues: pages did not load fully. Somebody suggested Refine and it is working perfectly!
 
Is Purify actually still supported? I am not talking about filter updates etc. but the application itself.
The last update was total crap and breaks tons of pages.

Was a great App but died to soon.
 
Eyeo effect? Purify dev was active here, and then there was a news about Eyeo offering money and gmail rendering issues for the amount being offered, all that came at the cost of Purify app itself.
 
Sparxxx said:
Is Purify actually still supported? I am not talking about filter updates etc. but the application itself.
The last update was total crap and breaks tons of pages.

Was a great App but died to soon.
Click to expand...
Last update to Purify was just a few days ago, so I think it is still supported
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back