--- Begin Message ---
However, I fail to see what this has to do with the present proposal. Yes,
if you do promotion to flowed text willy-nilly you can have trouble. But this
is already happening -- lots of deployed agents do this sort of promotion
routinely, and it isn't going to stop no matter what the standards say.
Paragraph-oriented text is restricted at the moment to non-compliant MUAs
abusing text/plain. As soon as there is a standard way of labelling paragraph
text, it will probably become much more common. Promotion of format=flowed to
format=fixed will then lead to broken display/reply as in:
Ah, the old "if we legalize it everyone will start abusing it" argument. I'm
sorry, but I don't buy this line of reasoning in the "war on drugs" and
I don't buy it here.
The reality is that non-compliant MUAs are already _very_ common, and they
aren't going away no matter how loudly we shout standards violation. And as a
direct consequence of this even compliant MUAs find it necessary to reformat
text all the time. They don't have a choice. So the net result is that it is
increasingly difficult to get formatted text to stay formatted. I now have
trouble with this on a daily basis and it gets worse all the time.
Again, I don't especially care what we do as long as we do something and do it
soon. But what we do needs to be selected not on the basis of "this could be
abused hence it is bad" sorts of arguments, but instead on the basis of "this
is more likely to be seen as solving a problem for MUA writers and hence is
more likely to be deployed". Any of these schemes can and will be abused.
I can accept an argument along the lines of "this is too complex and either
won't get implemented or won't get implemented correctly". And I definitely
think that what you originally proposed is far too complex. Randy's scheme is
much simpler, but I remain to be convinced it is simple enough. And for that
matter, I can also accept an argument that "we're unlikely to reach consensus
on one of these trailing space thingies any time soon so we'd best go with a
simpler labelling scheme we can reach consensus on".
As for the example you cite, where quoting appears at the beginning of a long
line, I already see it on a regular basis. I'd even say it is fast becoming a
defacto standard way of quoting. In fact what I now see on a regular basis is
the sort of quoting I deliberately used above. And to the extent that _any_ of
these proposals legitimize flowed text, I see them making this sort of quoting
more common. Perhaps I'm missing something, but I don't see this particular
proposal as especially bad in this regard. But I'm prepared to live with some
damage to quoting in the short term if we get a better result in the long term,
especially since I believe damage to quoting scheme in the short term is
inevitable no matter what we do (or don't do).
Ned
--- End Message ---