> That’s a timeline of 30 seconds of page activity about 5 minutes after the article was opened. To be clear, this timeline should be empty. Nothing should be loading. Maybe one short ping, maybe loading one extra ad. Instead the page requested 2000 resources totalling 5 megabytes in 30 seconds. It will keep making those requests as long as I leave the page open. 14 gigabytes a day.
It's not quite clear if 14GB/day is an extrapolation based off the author's sample of 5 MB/30 seconds or if the author actually left the page open for all 24 hours. Regardless, that's quite a bit of data.
Hi I'm the author. I did the naive extrapolation. 5 * 2 * 1440 = 14,400. Perhaps this number is not entirely accurate. The fact it is any larger than, say, 10 is the problem.
That's a terrible methodology, and makes the headline pretty inaccurate. (Edit: looks like the blog headline isn't inflammatory at the moment: did you change it, or did the HN poster insert the 24-hour figure?)
I'm against bad ads as much as everyone else, but I don't think using a terrible methodology to criticize them is a good idea. Making clear what you measured would be much better, and actually leaving the thing open for 24 hours would be best.
In the time since I read the linked article, I opened the LA Times site and navigated to a story, opened my network console, and read some comments on HN about the article.
Safari says it's been about 12 minutes. It's requested over 1000 resources (it pegged 999+ at ~2 minutes). It's downloaded 88.4 Megabytes of data. It appears to be continually firing pixels from a multitude of places (openX and doubleclick are common). It's also downloading a number of different videos. The article I opened does not have any videos related to it's content. This article is open in an unfocused window; I have not fired any scroll, click, or touch events natively after the few seconds I opened the link, scrolled a bit, then came back to HN.
88.4 megabytes in a few minutes. I remember when my modem couldn't download more than a couple megs in dozens of minutes, and I'm a shitty stupid millennial, not some graybeard. Maybe it could hit a gig, or 14, in a day, but the point really remains set: the LATimes site is absurd.
Oh yeah, like all the extra effort is going to make any actual difference. Sorry, demanding better data or methods on a casual observation is a way to avoid engaging with the problematic behavior. If publishers use bullshit methods to drive revenue then they must expect to receive bullshit critiques of their bullshit methods.
Perhaps the LA Times webmasters could apply your suggested methodology and come up with a more accurate estimate and share it with us here. I've already done my service in pointing how how broken their ad scripts are on a normal page. I don't need to set up a full bandwidth monitoring rig for them, too.
To answer your question about the title, I did not change my blog post. Hacker News' silent moderators seem to have applied their usual policy of rewriting post titles to the article title, no matter how dull it may be.
Someone just posted a comment on my blog they managed to leave it running for 3.2 hours and got 1.4GB of traffic. Then devtools crashed. I realize this is still terrible methodology, but perhaps we literally lack the technology to measure this webpage.
For the reference of others, I opened the LA Times on my computer and left the front page open for 10 minutes, and in that time it downloaded 48.7 MB of data.
It's not quite clear if 14GB/day is an extrapolation based off the author's sample of 5 MB/30 seconds or if the author actually left the page open for all 24 hours. Regardless, that's quite a bit of data.