Re: [Last-Call] New Version Notification for draft-crocker-inreply-react-07.txt

[Ned Freed <[e-mail]>]

> >> On Fri, 26 Feb 2021, Dave Crocker wrote:
> >> I'd flip it around.  What reason do we have to believe that any particular
> >> restricted vocabulary that we might define would be useful to users we
> >> don't know and who may not even speak any language we speak?
> >
> > cf, the reference to established practice, which is distinguished from
> > free-form text, which is what you now seem to be proposing

> I see a rule allowing a string of emoji, which we've heard is problematic,

With precious little evidence to back it up, and no suggestions at all as to a
useful alternative.

Here, why don't you try it yourself? Can you come up with a set of reaction
indicators and a way to structure them which:

(0) are fully and openly specified,

(1) enjoy widespread support in existing UI toolkits, both for display and
    composition, and

(2) can leverage considerable experience using them in similar applications?

I'm betting the answer is no.

Frankly, at this point I'm having a serious case of deja vu. The similarly of
this discussion and the charset discussion during the development and
deployment of MIME is nothing short of uncanny. And we ended up getting it
wrong - mind you, not wrong enough to refuse to accept Unicode and UTF-8 when
they became available, but not after wasting a ton of time, and wrong enough to
encumber our solutions with worthless crap like language indicators that AFAIK
nobody has ever used.

I am not going to make that mistake again.

What we're doing - again - is something the IETF really does have experience in
doing: Letting the best be the enemy of the good. (And this case I happen to
think the best is pure fantasy.)

> and a base-emoji rule which has an unupported assertion that it's five
> emoji developed from existing practice, although I'm not aware of any
> existing application that uses that set.  Do you have a reference?  In the
> apps I use, the set of emoji responses differs from one device to the next
> and is invariably very large, hundreds at least.

It's a very weak may. I think it's useful to have such an example in order to
make it clear that subsetting is allowed. Although a reference for the subset
would not hurt.

> > (which is odd, > given what stage of processing this draft is in.)

> I agree that it was extremely premature to last call this draft.

Sometimes a last call is what's needed to get people to engage. Dave
asked for comments several times, and got U+1F997. What were
we supposed to do?

				Ned