Text 1166, 58 rader
Skriven 2008-03-21 07:41:26 av andrew clarke (3:633/267)
   Kommentar till text 1128 av Lars Kellogg-Stedman (1:322/761)
Ärende: Questions re: format of packed messages
===============================================
On Tue, 18 Mar 2008 23:47:16 +0000 "Lars Kellogg-Stedman -> All" <1:322/761>
wrote:

> Some more technical questions for y'all...and I'm asking here because I
> think they're FTSC related, but if you think I should be asking
> elsewhere just let me know.

There is an echo called NET_DEV for these sorts of questions but I
haven't seen any traffic on it since 2006-02-01.  Actually, now that I
think about it I notice that my NET_DEV feed was through Bob Juge at
1:106/2000, whose system went down around that time and never
resurfaced.  It looks like it took www.juge.com with it.

Is NET_DEV still active anywhere?

Is Bob Juge still alive?

I guess posting here is fine.  :-)

> (1) Is the *order* of kludge lines in a message important?

Your question was adequately answered by others, but it reminded me of
a proposal from Paul Edwards about 10 years ago.  I can't find it on
the FTSC web site so perhaps it was never accepted, but basically it
was a suggestion to put ^aSOT at the beginning of the message body and
^aEOT at the end, before the tearline (if any) and origin line (if
any).  He was not satisfied with the way messages had to be parsed
backwards to extract a message body, ignoring the tearline and origin.

  @MSGID: 3:666/666 12345678
  @SOT:
  Hello world.
  @EOT
  -+-
   + Origin: X (3:666/666)

I don't think it was really a serious problem to require kludge lines
to "fix", so it never really caught on, but it was an interesting
proposal. I think the only software that supports SOT/EOT (as an
option) is Msged (which Paul wrote code for, at the time).

Anyway, one issue was with JAM message bases where after a message
from Paul was tossed, both the SOT & EOT kludges would be stored in
the JAM message header instead of the message body, which meant that
software honouring the SOT/EOT kludges may just display an empty
message.  This was also a problem if you "rescan" a JAM message base,
exporting it to a node would mean the SOT & EOT kludges would be at
the top of the message and not where they should be.

One could argue that the JAM message base format should never have
allowed kludge lines to become detached from the message body, but (1)
it's a bit late now and (2) I'm not sure the FTSC specs at the time
JAM became popular were very clear on how kludge lines should be
parsed, stored and so on.  For example, the Via kludge line was not
documented as an FTSC standard until 2003!

--- Sylpheed 2.5.0beta1 (GTK+ 2.10.14; i686-pc-mingw32)
 * Origin: Blizzard of Ozz, Melbourne, Victoria, Australia (3:633/267)