Re: non-IETF tree URI scheme draft

From: "Larry Masinter" <LMM@acm.org>
> I think that the requirements for the scheme
> registration are:
>
> 1) Bad schemes (poorly defined, insecure) should be
>  discouraged.
>
> 2) If bad schemes can't be discouraged, at least people
>  can find out whether IETF thinks a scheme is good.
>
> 3) Frivolous (vanity, cybersquatting, misleading)
>   scheme registration should be discouraged.
>
> 4) Collisions should be avoided.
>
> If you want to think that (4) is more important than
> (2), that's OK with me. I don't think it matters.
> We need to meet all of the requirements, and so
> the order doesn't matter.


All four are equally important, agreed.  What I was referring to was the
need to explicitly include the qualifier in the string that goes on the
wire. When we had this discussion,  it was pointed out, quite emphatically,
that collision avoidance was the reason. None of the other reasons justify
it.

--Ray

Received on Monday, 24 November 2003 12:27:52 UTC