[net.cog-eng] windows...

weamc@pyuxa.UUCP (12/07/83)

I don't have anything against windows, per se. I am aware of the analogy
between over-lapping windows on a CRT and the little piles of notes
on a desk. What irritates me is the head-long rush to announce software
with "windows." I am not convinced that the analogy is an apt one, because
no one has done any research (that I've seen) to indicate that windows do
help the user. In a time-sharing system, bit-mapped displays would impose
an huge load on the system. 
An example of this is the infamous BLIT terminal,
(Bell Labs Intelligent Terminal) which requires a special software front end
to run, and reportedly drastically slows system response.
I suspect that there are cheaper ways to provide some of the things that
windows provide.
   As far as using a mouse goes, I have used a Lisa, and found the mouse
irritating because I had to keep moving my hand off the keyboard. I was 
doing text input, which was part of the problem.
   I don't have anything against new ideas, but it bothers me that they
are picked up and used before the paint on the old stuff is barely dry.
The poor slob of a user suffers as he is told, time after time, that 
the framistan is the cure for all his or her ills. (Replace framistan
with data base, user-friendly, windows, mouse, UNIX, bit-mapped, etc...).
    
                         Andy Cohill

grunwald@uiuccsb.UUCP (12/16/83)

#R:pyuxa:-42000:uiuccsb:19000005:000:1017
uiuccsb!grunwald    Dec 15 22:23:00 1983

  Certainly no software package should FORCE you to use a mouse, but several
reports do demonstrate that mice are faster for the average user than step
keys (e.g. VI) for positioning. If you're not one of those average uses, fine.
However, I like it and I think that a properly designed editor using a mouse
would be much more useful than a step-key based editor. Before people had cheap
computers, interfaces really sucked because the basic resource was expensive.
At the current time, I think editors are a little less than thrilling because
of the "expensive resource" -- cursor motion and reverse bandwidth.
   I can specify much more information to the machine using a mouse than
step keys.


As for analogies of windows with papers on your desk -- maybe for people in 
offices. A better analogy as far as I am concerned is that windows provide
a way to have 7 terminals on your desk without the clutter.

Dirk Grunwald
University of Illinois
USENET	: ihnp4 ! uiucdcs ! grunwald
CSNET	: grunwald.uiuc@Rand-Relay

tay@ssc-vax.UUCP (Thomas A Yap) (12/18/83)

  Certainly no software package should FORCE you to use a mouse, but several
reports do demonstrate that mice are faster for the average user than step
keys (e.g. VI) for positioning. If you're not one of those average uses, fine.
However, I like it and I think that a properly designed editor using a mouse
would be much more useful than a step-key based editor. Before people had cheap
computers, interfaces really sucked because the basic resource was expensive.
At the current time, I think editors are a little less than thrilling because
of the "expensive resource" -- cursor motion and reverse bandwidth.
   I can specify much more information to the machine using a mouse than
step keys.


As for analogies of windows with papers on your desk -- maybe for people in 
offices. A better analogy as far as I am concerned is that windows provide
a way to have 7 terminals on your desk without the clutter.

Dirk Grunwald
University of Illinois
USENET	: ihnp4 ! uiucdcs ! grunwald
CSNET	: grunwald.uiuc@Rand-Relay

ucbesvax.turner@ucbcad.UUCP (12/18/83)

#R:pyuxa:-42000:ucbesvax:25800005:000:1379
ucbesvax!turner    Dec 11 16:03:00 1983

Re: windows, mice, framistans...

I have much the same complaint about mice--the touch-typist's bias
against moving hands away from home keys.  Another related problem
is having to move my eyes from the screen.  A common sequence of
events is

	look for mouse
	take hand from keys and grab mouse
	look on screen for target
	find cursor on screen, establish eye-hand interlock
	move to target and point
	(then, assuming that keyboard interaction is needed)
	put hand back on keyboard--and type the wrong thing

After a while, I wise up, and those last steps become four (overlapped)
steps

	look down at keyboard
	put hand on *right* keys, look up and type the right thing

But that's too many eye and hand motions for me.  My shoulders start to
hurt, and my eyes start to burn out.  I suspect that if you gave me a text-
editor with a mouse interface, I would rarely use the mouse, and would
deeply resent any operations that *forced* me to use it.

I think I will make the move away from keyboards when there is finally
adequate support for recognition of hand-written characters.  Even at
that, I type about 3 times faster than I write.  The interface would
have to be heavily oriented around graphically-invocable textual trans-
formations, to be worthwhile in terms of speed--lots of short-hand
macros for building up structure...
---
Michael Turner (ucbvax!ucbesvax.turner)

dave@rocksvax.UUCP (Dave Sewhuk) (12/24/83)

Seems that you have a poorly written program if you have to think of all those
motions.  The only time I ever have to "look" for the mouse on our editor is
when when you are not typing anyways, I never seem to have a hand/mouse
"context" problem with any of the mouse oriented programs that I have
used here at Xerox.

If done correctly those changes of context should occur infrequently
so they incur little penalty.   For pointing tasks it seems to me that any mouse
motion pays back so fast in terms of ease of getting cursor to correct place
that the re-indexing of fingers is not a problem.  Besides you could arrange
all those interactive commands to use the opposite hand as the mouse hand,
it would not have to leave the keyboard!

As for figure drawing stuff, I find that having a "line straightener" function
a must.  Plus you need a mouse with enough resolution to make a useful picture.
If you are drawing a picture into a 72 pixel per inch screen then you need
a mouse that has a small enough "gain" that you can draw into that screen
without seeing all the quantization noises and mechanical "noises" of the
mouse system.  After a small learning period you can draw some pretty
good drawings/graphs using the mouse as a pencil.
-- 
Dave

Arpa: Sewhuk.HENR@PARC-MAXC.ARPA
uucp: {allegra, rochester, ritcv, ritvp, amd70, sunybcs}!rocksvax!dave