Re: FVWM: Re: "active focus" applications (MetaCard)

From: Barry A. Warsaw <bwarsaw_at_CNRI.Reston.VA.US>
Date: Wed, 23 Oct 1996 15:46:03 -0400

>>>>> "SR" == Scott Raney <raney_at_metacard.com> writes:

    SR> Perhaps it was a bit strong, but we've been pointing out the
    SR> inherent and insurmountable problems of using pointer focus
    SR> with real GUI applications

I've never seen such a list posted to this forum. Why don't you post
your list of problems and let us debate them point by point? I'd
especially like to know what it is about your application that doesn't
work with pointer-focus. I really don't buy the argument that
pointer-focus is inherently flawed. And what is your definition of a
`real GUI application'?

    SR> And the statement is true: explicit focus is the de facto
    SR> industry standard

Hmm.
    
    SR> *You* know how to set up fvwm and will probably be changing
    SR> the configuration anyway to suit your needs, so why not just
    SR> make the default the one that's best for the X and UNIX
    SR> community as a whole?

Double hmm.

    SR> The MetaCard installation instructions include details about
    SR> how to change the focus model for mwm and olwm (which together
    SR> account for about 98% of our installed base).

Are you saying that if I wanted to use MetaCard I'd have to buy into
click-to-focus for my entire desktop?! I have no idea what MetaCard
is or does, but not cooperating with pointer-focus would a huge
disincentive for me ever using a product -- especially if it were GUI
product! (or at least I'd complain loudly and often on the bug list
:-) I'm sorry but that just seems very arrogant for an application to
force this global change on its users. I certainly don't begrudge
those who like click-to-focus since its inherently a religious issue,
but remember, the key initial in GUI is `U'! :-)

But maybe I'm jumping to conclusions about the interoperability
problems of MetaCard and pointer-focus. Please clarify!

    SR> Unless you care to dispute the fact that most people are
    SR> familiar with explicit focus, or that explicit focus is easier
    SR> to learn for novices, the correct default setting seems clear.

Do you have statistics. I'd really like to know how many of users
that are aware of having a choice between the two models, actually do
prefer to use click-to-focus, as opposed to just having no idea they
had a choice and using whatever mwm or dtwm gives them as a default.
You might be right, but I'd still like to see the results of a study.

> This is encouraging, as I think it shows signs of independent thinking on
> the part of Microsoft users (;^)

Thank goodness I found Xmouse on Microsoft's web pages. Win95
suddenly because moderately useful for me.

    SR> I actually find it confusing, unless these Windows users are
    SR> former workstation users.

As I (unashamedly) admit I am. And anyway, since I've begun using
Xmouse, I haven't had any problems with any of the Win95 applications
I use.
    
    SR> Windows is exactly the kind of environment that I was refering
    SR> to that doesn't work very well with pointer focus. Actually
    SR> the problems aren't as bad in Windows applications as they are
    SR> with X applications because Windows has floating palettes that
    SR> stay on top even when they lose the focus, and because there
    SR> is no way to do "active focus" applications in Windows
    SR> (clicking in a window always activates that window in
    SR> Windows).

Well, I'm still unconvinced that point-focus can't work just fine,
even for applications with lots of windows. Take Grail
<http://monty.cnri.reston.va.us/grail/>, a free web browser we've
written that's implemented in Python/Tk. In the next release, you'll
be able to drive the thing almost entirely from the keyboard, and even
as separate top level windows pop up and down (e.g. to interact with
the Bookmarks, History, or File dialogs), focus is managed correctly
to the subwindows without ever having to move the mouse (as long as
its one of the application's top level windows), even with
pointer-focus. Okay, I'm not an Xpert so maybe there's some inherent
design flaw in X I'm not aware of. I've just never seen it as a real,
unsurmountable problem with all the apps I've used over the years.
Even XEmacs (although it has some bugs) handles focus management in
subordinate top-level windows that do not contain the pointer.

    SR> It's also confusing because pointer focus really only has
    SR> advantages when you're dealing primarily with xterm-type
    SR> windows (as I invariably discover is the standard tool of
    SR> pointer focus afficianados). When your rarely or never use
    SR> them, pointer focus is just a pain in the ass because you're
    SR> constantly having to move the mouse before you can type in
    SR> dialogs and such.

Nope, its just that not all applications manage active focus
correctly. Its one of the main reasons why I can't stand Netscape on
X. I hit Meta-O and the dialog that pops up doesn't get focus. It
was nearly trivial for me to add this to Grail, *and* to move focus
back to the original window when finished interacting with the dialog.

>> Which only goes to show that the problem is MetaCard not
>> supporting fvwm, and not fvwm's lack of support for MetaCard.

    SR> This is really putting the cart before the horse: you buy a
    SR> computer to run applications, not a window manager. If your
    SR> applications work with other window managers but not fvwm,
    SR> it's fvwm that's broken, not the applications.

Sorry, but this is backwards, unless I bought my computer to only run
MetaCard. My applications should honor my environmental preferences.
They should not dictate to me what's for my own good especially for
such fundamentally religious issues as focus policy.

-Barry
--
Visit the official FVWM web page at <URL:http://www.hpc.uh.edu/fvwm/>.
To unsubscribe from the list, send "unsubscribe fvwm" in the body of a
message to majordomo_at_hpc.uh.edu.
To report problems, send mail to fvwm-owner_at_hpc.uh.edu.
Received on Wed Oct 23 1996 - 14:47:17 BST

This archive was generated by hypermail 2.3.0 : Mon Aug 29 2016 - 19:37:59 BST