As long as the person registering can come up with
a valid reason for an exception to be made an objector won't have a
leg to stand on here.
This actually sounds fine to me since there is at least some slack in
the word "valid." So long as someone is forced to come up with some
sort of explanation, it will prevent at least "blind" registration.
Sure -- the "slack" is entirely intentional.
Note also the explicit provision for marking something as being
of limited use. This is an "out" that basically allows for the
registration of practically anything.
I have a recommendation for an additional line:
Note that character sets registered for the second
reason should be explicitly marked as being of limited or
specialized use
"...and should only be used in Internet messages with prior bilateral
agreement."
I have no problem with adding this.
Or something to that effect. I think we all agree that character set
proliferation is a bad thing. I'd like to be able to say to another
vendor, "You *shouldn't* be sending that character set out to the
Internet in general because it is only registered in the
'not-ready-for-prime-time' category."
Fair enough, although given current experience with getting vendors to correct
things that clearly violate the specifications, I'm unsure as to how much
mileage you'll get...
I really don't want to prevent people from registering character sets
as they wish for their own personal use. And I agree that it's really
up to the IAB to come down with some guidelines about character sets in
general. However, I think there is no harm in making the language
somewhat strong.
Agreed.
Not really. The intended use of the review process is to get accurate
registrations in terms of content, labelling, and sometimes intent.
This has been a real problem for us in the past, and in my opinion it has
actually been a much more serious problem than the purely political
issues we've spent so much time on that have led us to any sort of
useful outcome. Perhaps by divorcing this process from the political side
as much as possible we'll improve in other areas. I certainly hope so.
I think I understand what the "political issues" have been, but I'm not
sure I understand what the problems with regard to "accurate
registrations in terms of content, labelling, and sometimes intent"
are. Could you please explain for the unwashed among us? If you think
that this is getting a little off-topic and would be better taken
off-line, feel free to send me comments privately.
RFC1345 provides one example of what I mean. (I don't mean to pick on RFC1345
in particular, but since it registers more character sets than anything else
its especially vulnerable.) The definition of the DEC-MCS character set is
incorrect: e; should be e> and e. should be i!. (Turns out this "definition"
came from a Falco terminal manual...) OK, inaccuracies can always be corrected,
but given the difficulty getting RFC1345 published to begin with Keld has been
understandably reluctant to try and get an update through the process. So the
error remains, and as it happens it's an important one in some of the
environments I support.
There have also been issues in regards to what constitutes a valid character
set definition and what does not. The current specifications tighten up the
definitions and what you have to provide and will eliminate a lot
of the discussion that's been necessary in the past.
Ned