home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gatech!usenet.ins.cwru.edu!agate!spool.mu.edu!sdd.hp.com!cs.utexas.edu!gateway
- From: ophof@SERVER.uwindsor.ca (Scott Ophof)
- Newsgroups: comp.mail.elm
- Subject: More enhancements?
- Date: 28 Dec 1992 18:35:31 -0600
- Organization: CS Dept, University of Texas at Austin
- Lines: 100
- Sender: daemon@cs.utexas.edu
- Message-ID: <9212290038.AA04210@SERVER.uwindsor.ca>
- NNTP-Posting-Host: cs.utexas.edu
-
-
- More items for the wishlist (maybe already there for 3.x?):
- And may I invite discussion for those points not already on the
- wishlist for 3.x?
-
- - Reading and/or writing multiple items at the same time, in any
- case from the same folder, but preferably also from different
- folders.
-
- - And while writing one item, being able to "flip" to some item
- being read (and back, of course)?
-
- These points actually mean implementing the concept of different
- "screens", and maybe levels of screens. Like this example:
-
- Entry point
- +-- Help screens
- +-- Aliases (**)
- | +-- Related Options
- +-- List-of-Folders
- +-- Related Options
- +-- Index of folder p
- | +-- Options (for THIS level)
- | +-- Item FROM person, date, subject
- | +-- Item FROM person, date, subject
- | +-- Item TO person, date, subject
- | +-- Headers-screen (*)
- +-- Index of folder q
- | +-- Options (for THIS level)
- | +-- Item FROM person, date, subject
- <*> | +-- Item TO person, date, subject
- | | +-- Headers-screen (*)
- | +-- Item TO person, date, subject
- | +-- Headers-screen (*)
- ....
-
- The "<*>" would show where one was before displaying this
- navigation screen (for those of us (incl. me) who easily forget
- where we were...).
- Of course a screen showing a tree as above to assist in
- navigation through those screens would be of great help. :-)
- Of course I don't mean that *all* indices of *all* folders
- would be built; only the one(s) specifically requested... :-)
- NOTE (*): See further on for comments re headers screen(s).
- NOTE (**): IMHO the aliases facility should be implemented as a
- totally separate thing; it's very useful for so many
- other applications (like "talk", "ftp", "telnet").
-
- - Another thing is that the headers of an outgoing item are not
- visible while writing the outgoing item. One needs to drop out
- of edit mode to see the headers (and be very careful *not* to
- accidentally hit RETURN, otherwise it will be sent off...), and
- then re-invoke edit mode to continue. A key that allows one to
- just "flip" to the headers screen and back is good enough
- already. :-)
- ("Pine" has the headers in a semi-protected area at the
- beginning of the item being written, and so does RiceMAIL.)
- An option to toggle between having the headers visible (or not)
- in the item being written would be nice. In any case in
- ".elmrc", but preferably also in the options screen.
-
- - May I seriously suggest the implementation of a "command line"?
- Where one can type in a whole (set of) command(s) withOUT the
- screen being updated at every key-press? Where the commands
- would be executed only after RETURN is hit, thus causing only
- *one* screen-update.
-
- - To go forwards or backwards n screens on the index, one needs
- to either repeatedly hit the relevant cursor (and wait for the
- screen updates), or type an item number. IMHO a cursor-key
- preceded by a number should indicate how many screens to go
- forwards or backwards.
-
- - In read mode, "[-]number SPACEBAR" could function as command to
- go forwards or backwards "number" pages, and "[-]number RETURN"
- to go forward/backwards that number of lines. This seems like
- what's normal in Unix useage, imho...
-
- - At index level, it would be very handy to be able to see whether
- an incoming item is personal mail, or from some discussion list.
- Most items from ((ex)BITnet) discussion lists are distributed
- with the "Sender:" field containing the name/address of that
- discussion list.
- For those people pulling netnews items into mail and reading/
- replying there, the "Newsgroups:" header performs the same
- function.
- But "elm" as it stands doesn't make use of these headers, and
- thus the user can't discriminate between the two, let alone sort
- on such fields.
-
- - BTW, being able to (temporarily) "exclude" from display (resp.
- "include") on the index items conforming to some mask would be
- handy too. Of course one would need a "display all" command to
- restore the display to normal...
-
-
- Regards.
- $$/ (F. Scott Ophof <ophof@server.uwindsor.ca>)
-
-
-