[rfc-i] Data point [Re: Fwd: I-DACTION:draft-hoffman-utf8-rfcs-03.txt]
dhc2 at dcrocker.net
Wed Oct 8 05:44:24 PDT 2008
Martin Duerst wrote:
> At 02:32 08/10/08, Dave CROCKER wrote:
>> Anything that disrupts that utility for the human reader constitutes fragility.
> Seems to make sense. As the main author of the IRI spec (RFC 3987),
> it is my experience that the current ASCII-only restriction created
> fragility, and that removing that restriction for examples,... would
> significantly enhance end-to-end robustness for those who actually
> need it (implementers,...).
If something has never worked before, then its not working now is not
"fragility". Adding support for something new is an "enhancement". Breaking
something that has always worked is fragility.
The issue is about preserving the installed base of user practice. Adding
something new should not disrupt existing practice.
More information about the rfc-interest