so 11. 1. 2025 v 21:47 odesílatel Rahul Gupta <cxres+ietf@protonmail.com>
napsal:
> Tim,
>
> Though they have certainly served us well, the "brutally simple semantics"
> as you describe them, are proving in a sense inadequate for media-type
> registrations that have been proposed recently. This is plain for anyone
> who has been at the mediaman meetings in the last couple of years and have
> seen the painful discussion on the suffixes draft. Something is gotta give,
> unless we find another approach (mine is but one possible idea). I am open
> to suggestions.
>
Can you spell out what’s been “painful” in the suffixes
discussions—especially for those of us who weren’t at the meetings? A bit
more detail could help us see where things are stuck and how to move
forward.
>
> BR/Rahul
>
> On Sunday, January 12th, 2025 at 1:32 AM, Tim Bray <tbray@textuality.com>
> wrote:
>
> On Jan 11, 2025 at 10:50:01 AM, Rahul Gupta <cxres@protonmail.com> wrote:
>
>> What if we could instead specify something like "Content-Type-Novelle:
>> wt(json(ld(vc+sd)))" with some form of standardized content arithmetic,
>> each string token being an registry reference.
>>
>
> Thinking about all the code that web clients would have to write to take
> advantage of this and currently unconvinced that the upside justifies the
> effort. Media types are one of the nice things about the Web, and I suspect
> the brutally simple semantics have generally encouraged good design. -T
>
>
>
>