Re: URL work in HTML 5

On 9/25/2012 6:58 AM, Robin Berjon wrote:
>
>> I don't think the problem statement is too difficult. What Anne is after
>> is implementation instructions for browsers. That's a good thing to
>> have. But for somebody creating an URI or IRI, or creating an URI/IRI
>> scheme, browser quirks can and should be irrelevant. It would be
>> hopelessly confusing for them to look at Anne's document.
>
> I am not so certain that the barrier here is so absolute. These things
> leak. For starters, I would expect whatever I use on the server side for a
> Web application to process URLs in the same way that browsers do (as much
> as possible). And once it gets into common development libraries, what will
> prevent it from spreading?

This seems like a classic example of what I view as a sometime breakdown of 
Postel's Law [1]. When you're "liberal" in what you consume, system 
robustness depends equally on others being "conservative" in what they 
send. This is a case where there's a lot of emphasis on documenting the 
appropriate ways of being "liberal"; it's unclear that we have in place the 
social and/or technical mechanisms to ensure that producers will be equally 
diligent in their conservatism.

That doesn't necessarily mean it's a bad thing for Anne or others to 
document the liberal rules for consuming, but we really need some focus on 
ensuring that typical content is reasonably clean.

Note that the above says nothing about claiming that the strings documented 
should be referred to as URLs, or the desirability of "obsoleting" 
important RFCs.

Noah


[1] http://en.wikipedia.org/wiki/Robustness_principle

Received on Tuesday, 25 September 2012 13:13:54 UTC