On 6/15/11 5:07 AM, Mykyta Yevstifeyev wrote:
Applications SHOULD resolve unrecognized "about" URIs in the same way as "about:blank".
...
I don't think MAY is fine here, as this is a recommendation.
I'm questioning it being a recommendation, is the point. Why is this behavior recommended, exactly? Given lack of existing interop and lack of a MUST-level requirement here, the only reason for a SHOULD would be if the behavior is believed to be better than other alternatives, right? Is it? I don't see why.
The point of this comment is to propose abandoning normalization of 'about' URIs because of some ad hoc behavior of an only application - Gecko.
No, it's to propose abandoning normalization because it's not necessarily compatible with existing deployed uses of about:, not clearly compatible with the web security model, and because the normalization is not defined in the spec. The Gecko behavior is just an illustration of the first point.
The purpose of our draft is to give a stable specification of the scheme
Yes, this is fine.
and normalize all existing types of behavior with regard to handling 'about' URIs. It will be easier for Gecko to change its behavior rather than for other apps to do this.
That's not clear to me given the security implications. Do you have data to back this up?
Boris, could you please let me know whether you have some strong opinion regarding your January comments/insist on incorporating them in the draft.
See above. -Boris _______________________________________________ Ietf mailing list Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf