[3233] in SIPB bug reports
Suggestions for XRN 6.17
daemon@ATHENA.MIT.EDU (mrhagger@Athena.MIT.EDU)
Tue Oct 6 19:05:48 1992
From: mrhagger@Athena.MIT.EDU
Date: Tue, 6 Oct 92 19:05:31 -0400
To: bug-sipb@Athena.MIT.EDU
This isn't a gripe but some suggestions for xrn.
I like to read the news by thread, as I suppose most people do. The most
convenient format to do this would be to have the subject list list each
subject only once, with perhaps a count of the number of articles in the
thread. Then reading one of these objects would start the thread, and
session killing it would delete the whole thread. It might even be convenient
for xrn to remember the threads that you were reading, and the next time it
was run to highlight any new articles in those threads.
Short of such a reorganization, a simple but useful improvement would be to allow
session kill to operate on a marked block of articles. That is, I could highlight
a chunk of articles which are uninteresting, and analogous to "mark read" which
passes them by, I could indicate "session kill" and any articles with subjects
matching the subjects in the marked block would be killed. As if I had executed
"session kill" on each of the articles in the block. That way future articles
with subjects that I had already indicated that I am not interested in would
already be gone.
Further, it would be convenient to have the option of making killed articles
vanish from the subject list. Again, it would help one only have to decide once
per thread whether to read or not.
The most useful improvement would be to increase the speed of the newsreader.
It would be wonderful if the reader would, for example, allow you to start
reading articles in a newsgroup before all of the subject lines are available.
(I assume that the real time drag is communication with the mail server.)
Furthermore, the next few articles in a thread (or in chronological order)
following the article currently being read could be being prepared in the
background so that when the user hits "next" or "subject next" the next article
would be ready immediately. Lengthy operations such as "session kill" could
also be done off-line to improve the response time of the program.
I realize that all this would be a lot of work. But the second and third
paragraphs describe changes that sound relatively straightforward, and which
alone would save much time for the users of xrn.
Yours,
Michael Haggerty
mrhagger@athena.mit.edu
