[OreoCookie]

Originally Posted by subego 

I didn't cover every work of every link, but the Verge one seemed to be the only one discussing tracker impact on bandwidth.

Please look at the other links more closely:
- MondayNote 1 has one table which compares the download size and number of connections with and without ad blocker. It contains a second table that compares the sizes of desktop and mobile pages without ad blockers.
- MondayNote 2 contains a graph comparing load times.
- The Brook's Review's conclusion is right in the title: they conclude you can save about 30 % of bandwidth on average by using ad blockers. But instead of comparing different sites, they compare different ad blockers.

If you follow the links within the articles, you will find other examples. If you do not think the links I have included are representative, feel free to Google for them yourself or better, to try it yourself. You can also use web dev tools for your favorite browser to get more detailed analyses. Open Safari's Web Inspector from the Develop menu, and reload pages. Make sure to delete the cache, though, otherwise that would invalidate your results. Every single review of ad blockers (desktop and mobile) I have seen has come to the same conclusion: you end up saving massive amounts of data and CPU cycles (= battery) if you use ad blockers.

Virtually every ad comes with a tracker, and besides, tracking is part of the deal between publisher and their ad networks. So it's really quite easy to understand what the tests I've referred to mean: ads and trackers on average double the size of web pages, there is persistent JavaScript activity for over a minute in many cases (which means you'll churn through your battery much faster) and pages take much longer to load. They initiate a lot of persistent connections whose only purpose it to track your browsing activities.

To give you an idea of the impact: if you have a 1 GB data cap (like I had in Canada) on my iPhone, then assuming a page size of ~6 MB without a blocker*, it means you can view less than 6 pages per day (!) until you exceed your data allowance. No wonder that even when being frugal my data cap ran out before the end of the month. These figures alone should be alarming, because they imply that people will actually surf your site less on mobile!

* I'm basing this figure off of the size of the desktop site.

[subego]

Originally Posted by OreoCookie 

Please look at the other links more closely:
- MondayNote 1 has one table which compares the download size and number of connections with and without ad blocker. It contains a second table that compares the sizes of desktop and mobile pages without ad blockers.
- MondayNote 2 contains a graph comparing load times.
- The Brook's Review's conclusion is right in the title: they conclude you can save about 30 % of bandwidth on average by using ad blockers. But instead of comparing different sites, they compare different ad blockers.

Am I missing the part where these examples are splitting the size of the ads versus the trackers? In your admonition here you're not giving me tracker-specific examples, or if you are, you've chosen to eliminate tracker specificity from the description of the data you claim is relevant.

Telling me an article claims you'll save 30% on bandwidth doesn't help me find the tracker specific data in the article.

No one is questioning the ads themselves take up enormous bandwidth.

[OreoCookie]

Originally Posted by subego 

Am I missing the part where these examples are splitting the size of the ads versus the trackers?

No, you're missing that virtually all ads contain trackers, and hence, it is not ads vs. trackers, but rather the only meaningful distinction here is between trackers with ads and trackers without ads. Note that one single ad may contain several trackers.

According to Frédéric Filloux from MondayNote, about 60 % of trackers in his tests were ad related. (I do not think somebody has computed that in relative download sizes, though.) Given the fact that Filloux had to wade through 500 trackers he encountered on the 20 pages he had chosen, and classify them by hand, you're in for quite a bit of detective work. But if you make the sensible conclusion that trackers associated to ads are comparable in size to trackers without ads, then roughly 60 % x 50 % = 30 % of the download volume is due to trackers with ads, the other 40 % x 50 % = 20 % is due to trackers without ads.

[subego]

I'm honestly trying to understand this.

The reason I feel value in differentiating trackers from ads is trackers with ads take more bandwidth, no? The trackers we can assume are the same size across both, but the trackers with ads have an extra payload.

It the split in trackers with ads to trackers without ads is 60/40, how can the total download volume also be split 60/40? This can only occur if trackers are equivalent in size to trackers plus ads. They're not. Trackers plus ads are larger because they're plus ads.

Likewise, I see the 60/40 split in the article, but don't see where you're getting the 50% to multiply them and arrive at a total percentage of download volume.


Edit: I apologize if the above appears obstinate. I really am trying to understand how you're arriving at your conclusion.

[subego]

Okay... I think I get how you're using the 60/40, but I'm still lost on where we get trackers being 50% of total download volume.

[OreoCookie]

Originally Posted by subego 

I'm honestly trying to understand this.

The reason I feel value in differentiating trackers from ads is trackers with ads take more bandwidth, no?

No, because that is not a meaningful distinction to make: almost all ads come with trackers, some with multiple trackers, and there is no way to separate the ad from the trackers that come with the ad.

If you would like to argue that trackers associated to ads plus the ads take up more bandwidth than stand-alone trackers, you have an argument, but in my opinion the difference is irrelevant for the present argument anyway: since you cannot separate ads from trackers, you cannot selectively block trackers only but leave the ads in place. Moreover, the presence of the trackers is a crucial revenue stream to publishers, and if we assume it were possible (which I think is not), then selectively cutting off the revenue stream from tracking also hurts them. (In your language: the reader is not holding up his or her end of the bargain by blocking trackers, he or she is just a slightly better citizen.)

So why do you think is the distinction you are trying to make relevant? The extra payload from ads and trackers combined can be measured easily and reliably, and ads cannot be distinguished well from trackers. And especially if you stick to your perspective of implied contracts, tracker blocking is no different from ad blocking (if it were possible to separate the two).

[OreoCookie]

Originally Posted by subego 

Okay... I think I get how you're using the 60/40, but I'm still lost on where we get trackers being 50% of total download volume.

It is in the tables I've quoted. To make it really simple, allow me to repost them here:
From MondayNote 1:
Home pages with and without ad blocking:



If you take the average across the selected sites, you see that ads contribute ~40 % on the selected sites. Strangely enough, two sites showed no differences, and if you exclude the two, the difference grows to 54 %.

Desktop vs. mobile site:



From MondayNote 2:
Difference in load times:



The Books Review compares the effectiveness of different ad blockers on iOS.



The Ghostery-powered, but deprecated blocker by Marco Arment, Peace, fares best and blocks almost 40 %. Note that this graph measures the amount of blocked content vs. the total size of the page. As the article explains, for certain small sites (he gives daringfireball as an example), this vastly overestimates the impact (50 % of 0,5 MB is much less significant than 30 % of 5 MB).

[subego]

Originally Posted by OreoCookie 

If you would like to argue that trackers associated to ads plus the ads take up more bandwidth than stand-alone trackers, you have an argument...

This is exactly what I've been arguing.

The reply of mine which started this tangent was specifically in regards to the assertion trackers are unfair because of the extent to which trackers chew your bandwidth.

I took great care to preface the statement explaining I was talking about bandwidth used by trackers and only bandwidth used by trackers.

In the Verge example, more than half of what you download was js. Totally agree that's an example of irresponsibly eating bandwidth, and it's essentially being done in secret, so I see arguments why that would be a problem. OTOH, the example given for The Daily Mail would have to max out at 10%. You're free to be against trackers there, but if you assert the problem with it is bandwidth usage, I'm going to challenge the assertion because it's false.

[OreoCookie]

Originally Posted by subego 

I took great care to preface the statement explaining I was talking about bandwidth used by trackers and only bandwidth used by trackers.

You continue to ignore the fact that you cannot separate trackers from ads, because almost all ads come with trackers — you can't separate the two. Now matter how much care you take, your distinction makes no sense in this debate.

Originally Posted by subego 

The Daily Mail would have to max out at 10%.

One point does not a statistic make. I think I've provided you with a lot of information, including statistics across a multitude of sites. There is more out there, but since it took you a while to have a look at the links I provided, I leave it up to you to find others.

Originally Posted by subego 

You're free to be against trackers there, but if you assert the problem with it is bandwidth usage, I'm going to challenge the assertion because it's false.

On what basis? Please provide evidence.

[mindwaves]

I started using Crystal the day it came out and got it for free. It worked great until I read that the developer will have certain companies pay him to allow ads using his app. A few days later, I noticed some of my websites suddenly having ads when they had none before. That is when I switched to Purify where I paid $0.99 for it and blocked the ads that Crystal started having.

[subego]

Originally Posted by OreoCookie 

One point does not a statistic make. I think I've provided you with a lot of information, including statistics across a multitude of sites. There is more out there, but since it took you a while to have a look at the links I provided, I leave it up to you to find others.

If I was making an argument based on that one data point why did I include a data point in the same paragraph which highlights the exact opposite?

I find your admonishments ironic in the face of glossing over the content of my posts. Multiple times now.

If you desire the last word, you may have it.

[starman]

Radio Survivor Won't Abuse Your Trust - Radio Survivor

[Mike Wuerthele]

Four newsposts a week from the Radio Survivor people.

Not the same.

[ghporter]

There are so many (at least somewhat) legitimate sources of information out there that have rampant advertising content, that ANY means of reducing the on-screen chaos will improve those sources' reader experiences immensely.

What's "rampant advertising content?" When pop ups pop up OVER EACH OTHER, when a pop-over takes control of the screen but doesn't bother to supply that welcome "X" that lets you kill that mother$%^$%^, and so on. Huffpost has some of this now and then, and I've even seen it on LATimes, but I see it a lot on "softer" info sites (often through links friends have provided).

When the medium is overwhelmed by the advertising that "supports" the message, you just do not get the message. Instead of ad-blockers, I think some sort of specification for where, how big, and how intrusive online ads can be is a real answer. Until then, we'll continue to play the same sort of cat and mouse game that virus and antivirus writers are engaged in, except the advertising'blocker game is (supposedly) legitimate and above-board.

[OAW]

I swapped out Crystal for Purify because it has a whitelist feature. Everything is blocked by default but it's really simple to whitelist those sites that aren't obnoxious with their advertising. The sites I frequent such as this one which do NOT serve up pop-up ads and auto-play videos get on my whitelist. As for all the rest ... screw them. At the end of the day however I don't think it's going to matter that much. I prefer the News app to Safari for my regular sites. Safari gets used these days for forums such as this (though a proper app would be much nicer IMO) and when I need to perform some Google Fu about something.

OAW

[starman]

Originally Posted by Mike Wuerthele 

Four newsposts a week from the Radio Survivor people.

Not the same.

And you know their numbers?

Also - the point of the post was that there are other ways to make $ on the net, which is what I said a few days ago.

[Mike Wuerthele]

Originally Posted by starman 

And you know their numbers?

Also - the point of the post was that there are other ways to make $ on the net, which is what I said a few days ago.

I know ours, and its not four news posts a week. We also crank out more than one podcast episode a week.

I am aware of other alternatives. I said earlier in the post that I wasn't confirming nor denying anything regarding a Patreon or other similar venues.

[OreoCookie]

The New York Times has published an article yesterday on the effectiveness of mobile ad blockers. They have included pretty neat bubble graphs that give you an idea how much data was actually used by trackers and ads. Pretty interestingly, they also include a cost per page computation (for US data plans). With some sites you apparently pay ~$0.10 per page view just for the ads (the average is ~$0.05). That adds up quickly (no pun intended).

The slowest page, boston.com, is insane: it takes 30 seconds (15.4 MB) to load the ads alone (compared to an already crazy 8.1 (4.0 MB) seconds after blocking)!

PS You need to pause your ad blockers to correctly view the page

[The Final Dakar]

[osiris]

Geez Louise, If you're a mobile browser you're not only watching the ad but paying (through the nose) for the bandwidth it takes to load.

[starman]

I was at NYCC today. Had to look something up on a Star Trek wiki on my iPhone.

Ads. Through. The. Effing. Nose.

You find the page, scroll a little, BAM!...ad.

You can't close it because it keeps redirecting.

Eventually it stops. Ok, maybe that was a one-shot.

Nope...scroll again...BAM! ads.

I don't know what wiki it was. Memory Beta, maybe?

[Mike Wuerthele]

Ooh, Memory Beta got really, really bad in the last year or so.

[The Final Dakar]

Just installed an ad blocker on my 6s after getting constant frequent redirects this place all week. It's your own ****ing fault.

[Mike Wuerthele]

Redirects here? From what?

[The Final Dakar]

The ads, I imagine. I've known subego to complain about this from time to time.

[Mike Wuerthele]

Huh. Shouldn't be any - we refuse them when proposed and kill them when we catch them. I'll see what I can find.

[reader50]

I haven't seen redirects lately. Just a lot of oversized top banners, mostly shutterfly.

btw, the shutterfly people aren't getting their money's worth. I'd assume from the name they have something to do with photography, but the banner suggests knick-knacks, cookies, or milk. If people can't figure out what you sell, there's something wrong with your ad.