Web tracking puts lead in your saddlebags, finds Mozilla study You already know that too many tracking cookies will slow Web page loading down to a crawl. Now, a study by Mozilla and Columbia University quantifies the problem. According to Columbia's Georgios Kontaxis and Mozilla's Monica Chew, spiking the excessive load of extraneous connections on the Alexa top 200 news sites can …
'Eliminating third-party content improves user experience'
What a revelation! Since I use NoScript, I guess I never noticed this before! I never knew!
[/s]
Wow. Guess those idiots at Mozilla never try out the add-ons for their own products. No wonder you can never get Mozilla to actually address real user problems, they are apparently living life in their own private bubble.
Also, use the Request Policy plugin for Firefox. This blocks any attempt to download third party content whatsoever. A few simple mouse clicks via its status bar icon will let you have fine detail control over whether to allow such content on a temporary or permanent basis. It's easy to use and understand.
So Mozilla finally implemented tracking protection, half a decade after IE did. And why now? Because their deal with Google has been not renewed, so they're not interested in enabling third parties to track their users anymore.
Actually, they are now counterinterested in enabling that. Instead, they want only themselves be able to track user behavior, so their own advertisements (called "Sponsored Tiles") stay relevant, while others' do not. That's the sole and only reason why they've implemented tracking protection.
So much about Firefox and a "better, open web", and keeping their "users first".
Thank god Firefox's market share is already down to barely above 10%, and most likely will sink to a few percent in the years coming.
Yup. I remember being disgusted well over a decade ago after indulging in a brief perusal of bugzilla entries relating to exactly this. I'd hoped to find out why the progress was so slow on implementing proper granular preferences for cookies, content classes, etc. with strict second/third party differentiation across the board: as it SO OBVIOUSLY should be. What I found was the utterly depressing spectacle of a procession of Mozilla's Google sponsored finest all professing to be unable to even comprehend what "third party" even means and openly & aggressively vandalising that little nascent privacy protection they found. I really was genuinely disgusted. It was Opera from there for me.
Mozilla: Utter, shameless, shit. Sponsored/shafted by Google Inc.
"Essential, but too much for the masses."
True. I've tried putting the AdBlock/NoScript etc on Mrs IPs computers several times. After a few days I have to take them off because it is "too difficult" to do what she wants to do, or she starts using IE instead (call me old-fashioned, but that just isn't acceptable). She does not understand how I can use the internet by setting permissions individually for each site every time (very few blanket exceptions for me - mainly temporary).
Essential, but too much for the masses. I has a sad
Have a look at Bluhell Firewall instead of Adblock Plus. Totally idiot-proof. Install, forget, enjoy.
The one site that years ago broke the camel's back & drove me to permanently install an adblocker (and later NoScript & a cookie manager) was none other than theregister.co.uk.
I got sick of waiting quite literally for 2 or 3 minutes (on an ADSL link!) for DoubleClick to wake up and serve their shit just so the page would render. Installed AdBlock, and I haven't looked back since.
Sorry guys - love your content, but that's when & why you lost me as a revenue-generating reader...
And I note a recent trend of scripts being served from cloud servers *which change* on a regular basis, which is a bit annoying when one has to enable a random URL for a page to work.
(That's nearly, but not quite, as annoying as purchase pages which fail to check that you're executing scripts from the money handling site until you hit 'submit'.)
"And I note a recent trend of scripts being served from cloud servers *which change* on a regular basis, which is a bit annoying when one has to enable a random URL for a page to work."
A cynical person might conclude this is a deliberate attempt to make things like NoScript more inconvenient, because you have to fiddle with it every sodding time you visit a page so affected.
Such a person might therefore be glad they have the facility to temporarily enable scripts for the session on any given page, in case one of those random URLs ever gets reused for a script used for annoying adverts (or worse) rather than something necessary to make a page work.
You can count how many favours the people behind such nonsense are doing themselves on the fingers of no hands.
Do you know, that idea hadn't even begun to consider the possibility of starting to think about crossing my mind... or perhaps it had.
What I don't know is how long a temporary noscript permission lasts. The wiki page states 'while the browser session is open'; as I (and I suspect many others) might have the browser open for weeks or months, with hibernation/suspension of the OS, I can't help feeling that it should expire as soon as the broswer tab containing the initial script request is closed. Perhaps it does - I don't know enough about it to know.
Do you know, that idea hadn't even begun to consider the possibility of starting to think about crossing my mind... or perhaps it had.
You're not being cynical enough... or perhaps you are. ;)
What I don't know is how long a temporary noscript permission lasts. The wiki page states 'while the browser session is open'; as I (and I suspect many others) might have the browser open for weeks or months, with hibernation/suspension of the OS, I can't help feeling that it should expire as soon as the broswer tab containing the initial script request is closed. Perhaps it does - I don't know enough about it to know.
I suspect it'll be until the browser is closed. It should be easy enough to test, though - you need to load a page that uses some javascript (from a static location) and isn't already whitelisted (permanently or temporarily). Temporarily allow the script(s). Close the tab. Visit the page again. If the Javascript is still working, it's for the duration of the browser session - but if it's no longer working, it's the duration the page is open in the tab.
"I got sick of waiting quite literally for 2 or 3 minutes (on an ADSL link!) for DoubleClick to wake up and serve their shit just so the page would render. "
THIS, so very much. It's not specific to El Reg - but my problem isn't the "44% bandwidth saving" but the interminable (literally - the damn progress indicator keeps spinning until I close Firefox later that day if I let it) loading times of some third-party fluff that will never arrive (long after the actual site is loaded), without which the bloody damn page either won't render at all or will simply neither let me scroll down nor at least let me browse any OTHER TAB (because hey we only have a single thread here and everything down to the last stupid IP lookup blocks it indefinitely). Firefox is currently in so many different ways so far away from "adequate" it's not even funny...
browsers with a black-list function that can deny content by subdomain (content from ads.somesite.com is denied whilst allowing other somesite.com stuff) are the browsers of choice. Adblock, Ghostery, and Noscript are fine, but they can not catch everything and "search and destroy" against the source code should always be an option. Sad to see the interwebs are becoming an arms race between throughput speed and ad-/track-/spy-/other mal- ware bloat, and that the self-defense moves learned during dial-up days are still required. (Honestly, does anyone BUY the stupid crap that gets advertised?)
They left out the part where entire sites become utterly useless when e.g. ajax.googleapis.com isn't whitelisted. In other news, Phoronix forums have just recently gotten a face lift, and stopped loading if JS is off-- white screen for you, thank you, drive through. Mickens was right.
what Reg left out-- with this soft, academically acceptable variation:
"One challenge of this approach is that blocking content can break a site’s functionality or appearance. For example, elements on the page may depend on a script being loaded from a tracking domain. To minimize unanticipated side-effects of Tracking Protection, we rely on a curated blocklist. Previous approaches have used heuristics or a default-deny policy at the expense of usability." (refers to PrivacyBadger)
One challenge, they say. Kick me if that ain't the central challenge of any approach. Full disclosure: I only followed the link to skim for that and stopped reading when I found it.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
