3 ms·

An issue sidestepped by this article is whether the statement in the second paragraph is correct: > ... tech and ad companies who say they comply with Do Not T

An issue sidestepped by this article is whether the statement in the second paragraph is correct: > ... tech and ad companies who say they comply with Do Not Track could simply ignore the flag set by IE 10 and track those who use that browser. That doesn't quite work. Web servers don't know what browser is on the other end of a connection. Yes, they know the "User-Agent:" line, but that is not the same thing. Claiming to follow a privacy standard, but then ignoring it based on a conclusion reached via fallible means, is a bit scary. It strikes me as the beginning of a slippery slope, regardless of what position Microsoft ultimately takes with IE 10.

Animus7 14y ago
So don't forge your headers. I don't see how this would be slippery at all.
- eli 14y ago
  The parent's point is that some of those IE10 users aren't just stuck with a default, they really did intend to turn DNT on.
  - adgar 14y ago
    Which is why DNT makes zero sense to be on by default.
- ggchappell 14y ago
  Does the DNT standard require that User-Agent strings correctly reflect the browser? [1] Guaranteeing a site follows DNT means that it will follow it, not that it will follow it only if headers are not forged. Also, is "forging" really a dishonest thing in this instance? Browsers have been making themselves look like other browsers for years, in order to deal with stupid servers that make incorrect assumptions about the User-Agent string. [1] Not a rhetorical question. I don't know the answer.
rurounijones 14y ago
Yea, the whole idea of "They are not following part of the spec so we can treat them differently" seems like a sketchy argument. Is there a clearing house for certifying browsers as DNT compliant? Or can the websites just say "Oh, THAT browser? Well I don't think it is DNT compliant so I am going to ignore the DNT headers even though they are perfectly to spec"

Animus7 14y ago

So don't forge your headers. I don't see how this would be slippery at all.

eli 14y ago
The parent's point is that some of those IE10 users aren't just stuck with a default, they really did intend to turn DNT on.
- adgar 14y ago
  Which is why DNT makes zero sense to be on by default.
ggchappell 14y ago
Does the DNT standard require that User-Agent strings correctly reflect the browser? [1] Guaranteeing a site follows DNT means that it will follow it, not that it will follow it only if headers are not forged. Also, is "forging" really a dishonest thing in this instance? Browsers have been making themselves look like other browsers for years, in order to deal with stupid servers that make incorrect assumptions about the User-Agent string. [1] Not a rhetorical question. I don't know the answer.

rurounijones 14y ago

Yea, the whole idea of "They are not following part of the spec so we can treat them differently" seems like a sketchy argument. Is there a clearing house for certifying browsers as DNT compliant? Or can the websites just say "Oh, THAT browser? Well I don't think it is DNT compliant so I am going to ignore the DNT headers even though they are perfectly to spec"