Re: HTML forms

nemo/Joel N. Weber II (devnull@gnu.ai.mit.edu)
Mon, 3 Mar 1997 22:17:40 -0500


Date: Mon, 3 Mar 1997 22:17:40 -0500
Message-Id: <199703040317.WAA13469@churchy.gnu.ai.mit.edu>
From: "nemo/Joel N. Weber II" <devnull@gnu.ai.mit.edu>
To: walter@natural-innovations.com
CC: www-html@w3.org, scotti@microsoft.com
In-reply-to: <Pine.SGI.3.95.970303174306.11842D-100000@shellx.best.com>
Subject: Re: HTML forms

   Date: Mon, 3 Mar 1997 18:11:52 -0800 (PST)
   From: Walter Ian Kaye <walter@natural-innovations.com>

   On Mon, 3 Mar 1997, nemo/Joel N. Weber II wrote:

   > I don't really want to see a hierarchy of keystrokes.  It probably will get
   > too hard to navigate.

   "Hierarchy" probably needed qualification as "two-level". Either you're in
   "menu mode" or "window mode". I think the problem comes in indicating to the
   user which mode they're in. The user must *never* be unsure which mode is
   the current one. Perhaps disable menus to the right of "Edit" while the Edit
   menu contains the mode toggle command for getting out of window mode?

It's too hard to intuitively handle more than one level.  The conflict
about which mode you're in demostrates the problem.

   Gee, I set up Ctrl-Shift-Q to move the active window of the frontmost app
   to the upper-left corner of my screen. I use Ctrl-Cmd-E to launch Eudora.
   I use Ctrl-singleclick on a window titlebar to roll up the window (using
   the WindowShade system extension). Sure I have function keys across the
   top of my keyboard, but they're hidden behind lots of 2x1.5" Post-It(tm)
   notes! <G> However, all my Ctrl+[alphanumeric] assignments do have an
   additional modifier, so that I can still use the Control key to enter
   ASCII 0-31 characters in those programs which can make use of 'em.

You're unique.  Most people don't spend that much time customizing their
machines.  I can't even be bothered to look in my .fvwmrc at home to
find the rgb values of the background color I like and then figure out
how to tell OpenLossage to use that color for my background at school...

It's important that browsers make this customizable, so that everyone,
including people like you, can be happy.  But we need to concentrate
on finding sane defaults that the majority of people will like.

   > If you click on the button, something happens.  If you click on its label,
   > I don't think anything happens.  So why would keystrokes make more sense
   > on the label than the button?

   I think this is more an issue of perception and programming rather than an
   issue of logic. Is it easier for software to determine what label precedes
   an object or what object follows a label? Does it really matter?

Does anyone follow my logic?  My suggestion seems much more logical...