Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!amdcad!ames!elroy!cit-vax!ucla-cs!oahu!frazier
From: frazier@oahu.cs.ucla.edu (Greg Frazier)
Newsgroups: comp.lsi
Subject: Re: VLSI design tools
Message-ID: <9564@shemp.CS.UCLA.EDU>
Date: 18 Feb 88 17:37:49 GMT
References: <154@liutde.UUCP> <5751@eecae.UUCP> <1601@phoenix.Princeton.EDU> <1121@cpocd2.UUCP>
Sender: news@CS.UCLA.EDU
Reply-To: frazier@oahu.UUCP (Greg Frazier)
Organization: UCLA Computer Science Department
Lines: 43

In article <1121@cpocd2.UUCP> howard@cpocd2.UUCP (Howard A. Landman) writes:
>In article <1601@phoenix.Princeton.EDU> buzz@phoenix.Princeton.EDU (Mahboud Zabetian) writes:
>
>Don't confuse "unlike a Mac" with "illogical".  They're not the same thing.
>
>>When you want to draw a rectangle, you click with one button aand then the
>>other.  Why can't we just drag???
>
>This is derived from Caesar, and makes perfect sense with a 3-button mouse.
>The advantage is that, if you have a rectangle that's slightly wrong, you can
>adjust it with a single click.  If dragging is your only alternative, then you
>are forced to redraw (re-drag!) the entire rectangle, which is impossible if the
>rectangle is not all on screen.  Further, suppose you are trying to draw a very
>large rectangle that must be precisely aligned.  With dragging, you must zoom
>
Indeed, going to a one-button-and-drag interface would seriously
reduce the effectiveness of the tool.  The operation described above,
where one is trying to precisely draw a large rectangle is used quite
frequently in large designs (at least, by me).  I often find myself
defining large arrays of structures, and thus have three windows on
my screen - one focused on the lower left hand corner, to make sure
I start the rectangle in the right place, one in the upper right hand
corner of the structure, to make sure I close the rectangle in the correct
place, and a "global" window, so that I can see the results of my
array construction.  Because most of the time you want the structures
to overlap, using an 'f' to put a square around the entire structure
is not effective.

Of course, I have personal prejudices against the single-button mice,
too.  It has always seemed to me that arguing that a mouse should have
a single button was equivalent to saying that a keyboard should have half the
keys - one can generate a 'c' by double-clicking the 'd'.  It seems
to me that any I/O device should have as many different inputs as can
be easily managed - there are obvious problems with putting more than
three buttons on a mouse, but I see no point in having less.  Anyway,
that's just my personal opinion, and really has nothing to do with the
argument above except that I needed enough text here so that rn would
not reject this posting.

.....................................................................
Greg Frazier				 o
CS dept., UCLA				/\
				    ----^/----