[21] in Zephyr_Bugs

home help back first fref pref prev next nref lref last post

[pete@ATHENA.MIT.EDU: Zephyr problems]

daemon@ATHENA.MIT.EDU (Jonathan I. Kamens)
Thu May 17 09:43:53 1990

Date: Thu, 17 May 90 09:43:26 -0400
From: "Jonathan I. Kamens" <jik@pit-manager.MIT.EDU>
To: Raeburn@MIT.Edu
Cc: bug-zephyr@ATHENA.MIT.EDU, pete@ATHENA.MIT.EDU
In-Reply-To: Ken Raeburn's message of Thu, 17 May 90 05:14:22 -0400 <9005170914.AA00217@multics.MIT.EDU>
   Date: Thu, 17 May 90 05:14:22 -0400
   From: Ken Raeburn <Raeburn@MIT.Edu>

   Yes, it's due to the changes to support "pretty" messages.  It should
   have been mentioned in the stuff I sent Anne (of which I don't have a
   copy any more), but "-p" is needed for "pretty" messages of class
   "message".  ("Pretty" is still the default for non-"message" classes
   such as "filsrv", otherwise we'd lose big.)

It seems to me that non-"message" classes are one of the big problems
that pretty was put in -- for example, the problem with zmailnotify
messages losing the host address in addresses of the form
foo@[xx.xx.xx.xx].

What's the reason for this design decision, anyway?  Was there any
discussion of it anywhere but in the zephyr team?

   It is debatable which way it should default for class "message".
   Although experienced users will notice a change and may want to use
   "-p" always anyways, the surprise/confusion factor for new users
   should be less if @-strings (such as in Scribe text) don't simply
   disappear.

This seems to be a questionable decision.  It seems to me that, given
a choice over which way to default, we should have defaulted to the
same thing we defaulted to in the previous release.

  jik

home help back first fref pref prev next nref lref last post