Firefox removes "do not track" feature support • Hacker News Board

Re: Firefox removes "do not track" feature support

#171

Post by WWLink » Tue, Dec 10, 2024, 8:45 PM UTC

darkhorse222 wrote: post #102

coldpie wrote: post #91

Earlier quoted context omitted.

> I would love if the cookie modals on each site became browser-level They are, in a roundabout way. Hop into your uBlock Origin settings and enable the Cookie Banners and Annoyances filters. The modal gets silently nuked in the background and you can carry on with your browsing. Since you never consented, this ought to be functionally the same as Declining the banner. The Kill Sticky bookmark works similarly, for cr…

As long as someone who does this is prepared to pay for every site they do it on (or forgo the site in the future), since targeted advertising often pays for the site they're visiting. Personally I would like to see ads improved, not removed, as I am unwilling to have 40 different subscriptions to 40 different websites all because every user disabled targeting.

Most of the websites I visit are selling something or sold me something. For the ones that aren't, I would consider a subscription service. If we're talking about something like an inexpensive web search or a subscription for a family of news websites or something. I'd totally pay for a good set of general/car/entertainment news sites. Definitely entertained that. But they'd have to nuke this nonsense about trying to get metrics and tracking and stuff.

Re: Firefox removes "do not track" feature support

#172

Post by registeredcorn » Tue, Dec 10, 2024, 8:45 PM UTC

recursive wrote: post #151

registeredcorn wrote: post #56

Earlier quoted context omitted.

>The percentage of websites that honor DNT is a rounding error Referring to my previous comment: "This is spurious reasoning. 'Many' is neither a percentage, or a basis for justification. Many people ignore speed limits - so what?" Privacy configurations are not a popularity vote. Instead, they are an implementation of software design. E.g. "No one" bothers with PGP, therefore it should be supported. A more charitabl…

> poorly enforced The problem is that there is no, and can be no enforcement mechanism. Unless you live behind a great firewall of [country] or are only accessing the web through your employer's corporate network. Operating a motor vehicle on public roads requires a license and vehicle registration in most jurisdictions. These rules are not always observed, but there are enforcement mechanisms in place. There is no w…

You raise some good points! I was going back and fourth about the speed limit analogy since cops do pull people over for speeding. I was trying to think of some thing that is supposed to be followed, but is commonly ignored without consequence. Perhaps the 6' social distancing recommendations during lockdown would work better? There was an expectation that people follow it, but actually adhering to it differed wildly depending on where you were.

Returning to the point, I suppose the question might be what enforcement would even look like, as you mentioned. Some governmental organization ala DMV? Seems excessive. A professional organization of some sort, similar to IEEE? Probably not. Perhaps just a "watch dog" organization like a sister organization to the EFF, specifically focused on reporting which sites honor, and dishonor DNT.

Personally, I would prefer the last, with some kind of check that could help indicate in the browser visually to the user whether the site honors DNT. Similar to (or tied into) the lock symbol for HTTPS would be nice. Outside of that, it seems like such a niche and fringe thing that, unless enabled by default, would fall on deaf ears. I don't see myself endorsing some kind of financial penalty, simply because I think it would hurt small websites. And also because website stuff is confusing and stressful enough already. I personally think a kind of visual "name and shame" to the end user, along side improved support to support DNT would be optimal.

I.e. Hot dog on a stick in one hand, dog poop on a stick in the other hand.

Reading through the Wikipedia article, it sounds this is largely what was already done with DNT, and failed. https://en.wikipedia.org/wiki/Do_Not_Track#History

...And subsequently, GPC began to act in a way that I'm not thrilled about. https://en.wikipedia.org/wiki/Do_Not_Track#Global_Privacy_Co... It comes off reeking of a pretext to confine the nature of how websites, and the internet at large, are allowed to operate.

Re: Firefox removes "do not track" feature support

#173

Post by kevin_thibedeau » Tue, Dec 10, 2024, 8:48 PM UTC

godshatter wrote: post #161

darkhorse222 wrote: post #102

Earlier quoted context omitted.

As long as someone who does this is prepared to pay for every site they do it on (or forgo the site in the future), since targeted advertising often pays for the site they're visiting. Personally I would like to see ads improved, not removed, as I am unwilling to have 40 different subscriptions to 40 different websites all because every user disabled targeting.

I'm not responsible for making a web site owners business model work for them. I'm just filtering data that I downloaded from a publicly-available because I don't want advertisers tracking me around the web and profiling me. Advertisers are welcome to treat me like a magazine from days gone by with lots of static images of their products and no code involved. If they don't like that, then I'm okay with going back to…

You're just following the FBI guidance and not allowing untrusted software to run on your computer. If that is a critical part of a business model, the business needs to change it's behavior, not me.

Re: Firefox removes "do not track" feature support

#174

Post by photonthug » Tue, Dec 10, 2024, 8:57 PM UTC

gkoberger wrote: post #156

miki123211 wrote: post #150

Earlier quoted context omitted.

I think a new browser could genuinely gain massive popularity if it was really good at this, and advertised the feature heavily, particularly in the EU. Google will never touch it with a hundred-foot pole due to antitrust concerns, they're effectively banned from making any significant, user-experience-affecting changes to Chrome at this point. Many people would immediately switch to a browser with 1) reliable Youtub…

I somewhat agree... but browsers aren't a profitable business. In 30 years of browsers being mainstream, nobody has built one that's sustainable –– only works if it's subsidized by a larger company. There's been a few attempts (Brave wants to monetize via crypto, Arc is pivoting away), but it's really hard. People don't want to pray for a browser – 99% of people are apathetic, and the 1% that cares aren't known for p…

After 30 years, isn’t it weird that the conversation is still about whether building a browser is profitable or if users are willing to pay for one? One would think that the technology would be so mature and ubiquitous that this is not a major issue 30 years later. If the core technology is still changing so fast that browsers need to be in constant development for the entire duration of their useful lifespan, maybe that is the problem, and the web is just doomed to be a shit show until corporations are distracted by enshittification of alternative platforms like VR.

It’s just weird that a few hobbyists can generally throw together a database in a weekend, fork kubernetes and probably run with it forever if they really wanted to, create a free operating system that takes over the world, etc. And yet for browsers, we’re shaking our heads and saying the situation is impossible, we kind of always have done this, and it looks like we always will.

Re: Firefox removes "do not track" feature support

#175

Post by registeredcorn » Tue, Dec 10, 2024, 9:03 PM UTC

tehjoker wrote: post #66

registeredcorn wrote: post #39

Earlier quoted context omitted.

Sounds good to me. When (major advertiser/website detected): Prompt user, "Warning: (Advertiser/website) insists on tracking you, and have made public statements affirming this position. Your privacy is not enforceable on this website."

more like class action or regulatory step in. in theory market transactions are based on mutual consent. if consent isn't respected, then that's a problem.

I actually ended up expanding on this a bit in a different comment thread, if you're interested. :) [1]

It's a tough position to be in because the thing that really gets heads turning is regulation, licensing, and fines, but...when it comes to website design (assuming we're not talking about illicit material) I get queasy at the idea of (the/any) government saying, "You're not running your website the right way! Pay us money!" Perhaps the few exceptions being something like: PII storage, or payment processing.

I dread the idea of anyone saying you have to, say, use a specific font type or whatever, you know? I don't want to put that burden on website owners, or complicate my own life.

I'd rather inform the end user, point out the biggest offenders, and leave them be. "Detect, inform, and move on." Big scary message[2] then leave it to the user to decide what they want to do. "Oh, they're going to track me? Let me look up that VPN-thingy I keep hearing my nephew talk about."

[1] https://news.ycombinator.com/threads?id=registeredcorn#42380...

[2] https://i.imgur.com/W885J9X.png

Re: Firefox removes "do not track" feature support

#176

Post by BobaFloutist » Tue, Dec 10, 2024, 9:05 PM UTC

nemomarx wrote: post #160

darkhorse222 wrote: post #124

Earlier quoted context omitted.

I think your approach is reasonably ethical. I'm not all that concerned with ethics; rather the continued funding of the websites we like. One important element of your first hypothetical is that a donation model probably won't suffice, websites will need to be able to set their price based on their costs, so it can't be evenly split between the sites you visit. So I think the experience would be like you add $5 to y…

the problem with ads targeted to what I might like is they lead to more purchasing and spending, which is not without cost. something you won't buy without seeing an ad for it is probably an impulse purchase, and those are somewhat predatory. it seems more reasonable to me to only advertise like that when a user signals they're interested in a particular type of product, not passively all the time.

Yeah people think of advertising as a "free" way to fund things, but companies wouldn't pay the advertisers if they weren't making a profit off of it, so in theory the customers are still paying for the website, but it's only some of them paying for the vast majority that don't pay a dime, and everyone else is just getting irritated by the ads for no benefit to the website.

Re: Firefox removes "do not track" feature support

#177

Post by araes » Tue, Dec 10, 2024, 9:06 PM UTC

wlesieutre wrote: post #100

gkoberger wrote: post #67

I was at Mozilla when this was implemented. It was completely optional for websites to support this. A few did at first. A lot of people internally wanted it to be on-by-default, but the argument was that if it was, nobody would respect it – after all, what tracking platform would willingly only track the 0.1% of people who went into the settings to enable it? (Internet Explorer did eventually enable it by default, w…

> IE did enable it by default, which got them good press but ultimately killed the feature since everyone stopped respecting it. That's why RFC 35140 "Do-Not-Stab" specifies that the user agent MUST NOT enable it by default. https://www.5snb.club/posts/2023/do-not-stab/

> most stabbings are not done by malicious actors, they are simply law-abiding companies which will gladly stop stabbing you if you ask.

> The header has only one form, Do-Not-Stab: 1. This is because the lack of a header indicates a clear preference that the user wants to be stabbed.

> Exceptions to the Do-Not-Stab header are accepted when commercial interests outweigh safety concerns. These include: Stabbings requested by a government. Websites SHOULD NOT try to challenge the legality of any stabbings requested, the user probably deserved it.

Re: Firefox removes "do not track" feature support

#178

Post by jmward01 » Tue, Dec 10, 2024, 9:08 PM UTC

What I want is to have a 'pay me for ads' option that links payment info so that I get $ for viewing the ads if enabled or doesn't show the ads if not enabled or the ad service isn't willing to pay up. At least this way I would get a cut of the money they are making on me. Expanding on this idea you could even set your rates 'cross-site tracking is XX, video is YY, sound is 1$/second, etc etc'

Re: Firefox removes "do not track" feature support

#179

Post by dmattia » Tue, Dec 10, 2024, 9:10 PM UTC

lukeschlather wrote: post #139

JohnFen wrote: post #50

Earlier quoted context omitted.

> Companies that never respected DNT aren't going to respect GPC. It depends. While I agree that GPC is technically just a more complicated form of DNT, the major difference is that DNT is 100% optional for websites to honor, which is why they don't, but GPC becomes mandatory for nations that have reasonable laws around tracking. Companies operating in those nations will honor it because there are legal penalties if…

Does this mean that if I set GPC, companies are not allowed to show me cookie banners under GDPR but just assume I hit whatever their "decline all tracking" button says?

In California under CCPA, it actually goes beyond just frontend regulation and cookie banners, and into the realm of backend tracking management: https://oag.ca.gov/news/press-releases/attorney-general-bont...

The California Attorney General ruled that if a user presents a GPC signal, the company should update all of their backend systems to opt out of tracking in the same way as if the user clicked a "Do Not Sell My Personal Information" button.

Re: Firefox removes "do not track" feature support

#180

Post by kuschku » Tue, Dec 10, 2024, 9:15 PM UTC

gkoberger wrote: post #67

I was at Mozilla when this was implemented. It was completely optional for websites to support this. A few did at first. A lot of people internally wanted it to be on-by-default, but the argument was that if it was, nobody would respect it – after all, what tracking platform would willingly only track the 0.1% of people who went into the settings to enable it? (Internet Explorer did eventually enable it by default, w…

> Overall, I'm happy to see this sunsetted. I don't think it actually did anything – in fact, I think it implies that it did way more than it did, so it was just a faux feeling of security.

I'm sad to see this, as many sites actually used it.

Geizhals.de, a major european price comparison site, uses DNT as cookie opt out.

My personal sites, but also the official websites from a few companies I worked at used umami or plausible metrics, configured to obey the DNT header for opt out handling.

And only recently German courts have ruled that the DNT header is legally considered rejection of tracking (Az.: 16 O 420/19)

It's actively used across the web, and Mozilla just decides to kill it? What the heck?