[2830] in Release_7.7_team
Re: Zephyr reverse-stacking
daemon@ATHENA.MIT.EDU (John Hawkinson)
Tue Jul 17 12:46:22 2001
Date: Tue, 17 Jul 2001 12:46:19 -0400 (EDT)
Message-Id: <200107171646.MAA10186@multics.mit.edu>
To: Greg Hudson <ghudson@mit.edu>
CC: aui@mit.edu, release-team@mit.edu
In-reply-to: "[2819] in Release_7.7_team"
From: John Hawkinson <jhawk@MIT.EDU>
(I realize that much of the debate may be over)
| * Anecdotal evidence suggests that most people who know about
| it use it.
I would like to suggest that this is incorrect.
The union of cfyi, sipb-office, netusers, watchmakers, source-developers
gives me ~570 unique Athena users, and I would suspect this includes
a lot of clueful Athena users. Of them, 167 have readable .Xresource or
.Xdefaults. Of those, 73 of them set the reverseStack X resource. Of those
who set reverseStack:
41 true
15 True
13 on
1 t
1 false
1 T
1 False
(for the curious, jmorzins and mollfrey are the two falses).
I think it is reasonable to presume that everyone with a readable
.Xresources or .Xdefaults file is clueful enough to be aware of
reverseStack (I understand there may be debate on this issue).
It therefore seems to me that 71/167 (43%), or slightly less than
half, of people who know about it use it.
I think this argues compellingly against the "anecdotal" evidence above,
which I suspect is merely based on people's personal "feel" of it and
not on surveying. As such, there's a lot of self-selection of the population.
Certainly there is self-selection in my methodology, too.
Frankly, I'm surprised the number is as high as it is. What it really
suggests to me is that it is a personal preference, and that reasonable
people will disagree and choose independant of the default. That in mind,
the default should cater to the least experienced people and the common
case. I think that valueing immediecy over some idea of conversational
coherency is correct.
As an aside, I sometimes feel like I'm one of the last holdouts who
use windowgram zephyr for a large quantity of zephyrs ("zctl ret | wc -l"
reports 332 for me right now, though that's probably misleading). I
find the idea of reverseStack repugnant because it either prevents people
from participating in fast-paced conversations (by the time they get to the
last zephyr in a rapidly moving conversation, it has already been
superceded), or worse, it encourages the behavior of multiple respondants
all responding without having seen future replies, leading to needless
duplication and annoyance. I tend to get really annoyed when someone
answers a five-minute old question because they are using reverseStack
and haven't come to all the previous answers yet.
--jhawk
