From: J.B. Moreno (planb@newsreaders.com)
Date: Mon Jun 16 2003 - 11:05:15 CDT
On 6/16/03 8:07 AM, Bruce Lilly at <blilly@erols.com> wrote:
> Seth Breidbart wrote:
>> If Re: were being used correctly, in Case 2, the child article would
>> have "Subject: bar" or perhaps "Subject: bar (was: Re: foo)" or
>> "Subject: bar (was: foo)".
>
> "used correctly" doesn't apply to unstructured fields which can be edited
> by the user. You're assuming that no user will ever edit the Subject, and
> that's not a valid assumption. You can try to mandate that UAs prevent
> users from editing the subject on follow-ups, but such a requirement would
> almost certainly be ignored by software authors.
I am a user, and the only headers I can't edit are the ones added by the
server (client restrictions are irrelevant as it's not all that infrequent
for me to access a news server via telnet, and I probably POST via telnet
once or twice a year). So if your argument is that because a user (myself
for instance) can disregard the requirement then we shouldn't make it, then
that means that we have to require the server to create all headers.
As for mandating that the clients UA prevent the user from editing the
Subject -- that's not what we are saying, and you know it. At *most* we are
saying that the UA should prevent the user from entering a leading "Re: "
(which is not actually that hard to do, and since the UA should be doing
some pre-posting comparisons to the original [because of *charset issues],
it's not even that unreasonable).
More realistically we are requiring the UA (in particular the Followup
Agent) to do some pre-processing before handing it over to the user, after
that it's up to the user to follow our directions. News isn't just programs
talking to each other, humans are an integral part -- and our restrictions
apply just as much to them as to programs.
*When posting non-ascii characters the "previous" Subject should be used to
help determine what, if any encoding, should be used.
-- J.B. Moreno