Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

They key thing to understand is that if IE10 did not have DNT enabled, that the default setting would be _just as arbitrary_ and would still therefore not "map to user intent" in their words. There has to be a default in one direction or the other.

That, and many users will use IE10 knowing that it ships with DNT pre-enabled. To ignore this is totally immoral and unethical. This is totally shameful.



>They key thing to understand is that if IE10 did not have DNT enabled, that the default setting would be _just as arbitrary_ and would still therefore not "map to user intent" in their words. There has to be a default in one direction or the other.

The default is to send a null value in the header meaning that no intent has been expressed, that's not arbitrary.

> That, and many users will use IE10 knowing that it ships with DNT pre-enabled. To ignore this is totally immoral and unethical. This is totally shameful.

Most users will have no idea what DNT is, nor will they bother to switch the flag. MS setting it to do not track by default undermines the effort being put into the standard. No one is going to comply with a voluntary standard if one of the largest browser vendors turns it on by default.


> The default is to send a null value in the header meaning that no intent has been expressed, that's not arbitrary.

My statement that there has to be a default in "one direction or the other" is clearly wrong, as you've indicated. I would still argue that a null value is arbitrary and does not map to the user's wishes. My opinion is that having it 'on' is no more or less arbitrary than having no wishes expressed.


> I would still argue that a null value is arbitrary and does not map to the user's wishes.

It's not arbitrary though, it clearly spells out that the browser vendor doesn't know what the users intent it.


Again, fair point - but then by definition the second clause of my statement is correct - the user's wishes are unknown, so any direction taken is then... arbitrary? :)

At any rate, you seem well-versed in this area, so let me ask you a question: what is the difference the website's behavior between a "null" and a "track me, please" value in the header?


It's not binary, it's ternary. 1 opt out, 0 opt in, null unknown choice.

In practice the difference between null and don't track me please are probably non existent. However, in the future if this were to take off it's possible that someone would come up with creative benefits/uses for tracking that provide incentive for users to be tracked.

edit: Don't think I addressed your question. The difference between null or tracking ok and don't track are likely just to be generic advertisements shown to a user instead of targeted ads, the prevention of some back end selling of user interaction data and some other things that most people have no idea is going on. If you can't track users across sites you lose some of your ability to build up profiles for them. Advertisers will argue that the ads will have significantly less value without those secondary or tertiary ways to monetize eyeballs and I suppose you could see some decline in what advertisers are willing to pay for impressions. Ultimately I've been removed from advertising for a few years so I'm not entirely sure how much it will make a difference.


This is really helpful - thanks.


Nope. For the past 10-20 years, browsing has worked as if there were not a DNT signal. So clearly "default off" is not "just as arbitrary". Yahoo is right.


My opinion is that _any_ setting that has not been chosen by the user is arbitrary. I understand your opinion, however.


Another way to look at it: this is the current status quo, and Yahoo is not going to make a radical change to its business because Microsoft changed a default setting in its web browser.

I had no reasonable expectation that Yahoo would handle this differently.


Yeah. Anyone who reads the default settings summary in the OOBE (out-of-box experience) in Windows 8 will be aware it's enabled.


In other words, almost no one will know. The thing is most people don't take the time to read those things they just click through blindly.


Yet somehow software developers are quick to screech that users agreed to 'license agreements' that have the same weight, right? (Words on a screen that everyone just 'clicks through blindly'.)

You can't have it both ways. Either having something show up and clicking next manifests consent or it doesn't. IE is extremely clear that DNT will be turned on and lets you opt out of doing so. Users are expressing intent by using the default settings; it isn't like it's a hidden default that they don't tell anyone about.


Somehow you replaced "lawyers" with "software developers," right?


I don't know any developers that believe that however I've met many managers and attorneys that believe it. Common industry practices should be laid at the feet of those responsible.

Beyond licensing agreements though, people won't read instructions, disclaimers, warnings, etc. It's not just one thing, and when designing things at work one of our common points to try and make things more clear includes "people don't read" so how can we improve this.


Almost no one? It's a short summary and I expect, since people have to choose between defaults and custom setup, people will read it.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: