Re: Summary of Issues Raised to date

On Wed, Mar 17, 2004 at 10:10:01AM +1000, Laird Breyer wrote:
> On Mar 16 2004, Mark E. Mallett wrote:
> > 
> > Couple that with a recommendation about fallbacks when attempting to
> > compare subjects, similar to the one that says to skip over zero or
> > more "Re:"  strings.  e.g., "when comparing subjects for fall-back
> > threading, initial substrings matching "Re: ", then "\[.*\]", are
> > iteratively skipped until either a positive match is attained, or
> > until the comparison fails."
> > 
> 
> The problem of tag buildup is more insidious than Re: buildup. 
> When you consider an email exchange n messages deep, you can expect
> n - 1 Re: tags, that's about 4*(n-1) characters of subject real estate
> wasted. 

I see what you mean.  OTOH I think one must be very careful when
trodding on "Subject:" text.  If you have a filter that's going to be
filing a message into your private personal mailbox, sure, have at
it.  (This is the mindset I was in when talking earlier in this thread
about guidelines for filter programs when tagging subject lines.)
Otherwise if you are trying to communicate information to some other
downstream filter, I don't think the Subject line is the way to go.  I
realize that (as has been said in this thread) there is some necessity
for subject tagging to help out dumb email clients that can only take
actions based on a limited set of headers.  I'd put that into the
first case, though: the one where a recipient is altering the subject
line for their own personal benefit, and they can pretty much do what
they want, including altering the Subject if their filter program lets
them.


> sample exchange with tags (ok, good):
> 
> 1) Hello
> 2) [ok]Hello           ---> Re:[ok]Hello
> 3) [good]Re:Hello      ---> Re:[good]Re:Hello
> 3) [ok]Re:Re:Hello     ---> Re:[ok]Re:Re:Hello

Perhaps subject-tagging facilities should include algorithms for
cleaning up those sorts of messes :-)

mm