On Tue, 3 Oct 2000, Kenton R. Wilson wrote:
> Firstly, I am not a Motif expert. [...]
Ditto...
> Lastly, and ironically, I have been assigned to a project in which I
> am required to do some programming using Qt :(.
:-)
> [...] I believe Motif is still a viable GUI technology we need to be
> active in making sure of it. If the consensus is that there isn't any
> future then we may as well just close up shop today and call it quits
> because there really isn't any point.
I strongly believe that Motif is still a viable technology. I tried to
use use other toolkits before and gave up because they were either too
slow, weak, poorly documented or dependent on non-portable technology
(mainly C++).
> Those in the OpenMotif and Lesstif projects, and others such as myself
> who feel that Motif is still the best, *on the whole*, X11 GUI toolkit
> available need to come together and reinvigorate interest in Motif and
> move forward with what I would call the Motif Enabled Desktop. For all
> the focus on KDE/Qt and Gnome/GTK in OpenSource circles I get the
> distinct feeling that there are a considerable number of people that
> are pleased with neither KDE/Qt or Gnome/GTK and these are the people
> that are potential users for the Motif Enabled Desktop.
There was an effort to create an open source (GPL-licensed) desktop
environment based on Motif. It was named eXode and was in the Lesstif's
applications list for a long time. Unfortunately, the project seems to be
dead now.
> What is the Motif Enabled Desktop? [...] I do see at its basis a
> framework which consist of the following.
>
> 1) Based on Motif :)
Oh, surprise!
> 2) A cooperative framework/standard in which applications can
> cooperate with one another to accomplish some larger task (e.g.
> passing documents and data from a word-processor to a browser). And
> no, I'm not advocating a "component" model such as Gnome project
> where you can gee-wiz embed you html widget into your word processor
> widget, into your ...; I don't envision something on that level of
> complexity. It has certainly been the bane for Windows users and it
> probably will be for Gnome users as well.
I advocate for a component model. Actually, lacking such thing is a
weakness of Xt. There is already a rather primitive component support in X
in the sense that two or more applications can share drawables and data
(by means of properties) and send messages each other. There are
applications that use such features (e.g. Ghostscript/Gostview) but this
was never explored by Xt, whose "component model" is restricted to
widgets.
> 3) Provide a minimum set of apps that users need on a daily basis.
> I think it pretty well established (at least for now) that users
> want:
Most of them already exist, but are not integrated.
> a) Word-processor
Ted, SIAG's Pathetic Writer, Maxwell, Papyrus (seem to be dead), Star
Office :-P
> b) Spreadsheet
SIAG, GNU Oleo
> c) Mail
ML, IshMail
> d) Browser
Netscape, Amaya, Mosaic, XmHTML (may be a starting point)
> e) File-manager
MOXFM, LLNL-XDIR, etc
> These seem to be the *must have* applications that *most* people
> want on a day-to-day basis. If you can provide these you go a
> long way in making you desktop appealing. To a lesser degree
> people also want goodies such as:
> e) CD player
XmCD. I'm working on a lightweight one.
> f) Mpeg/Avi player.
Timidity
> g) Paint/Drawing/Graphics
XPaint and XFig are based on Xt/Xaw. May be ported to Motif. There are
other drawing tools.
> h) ???
PPP dialer, calendar, calculator, personal accounting system, etc.
> One would also need a window manager and I would suggest mwm as
> a starting point. Personally I think its a pretty good window
> manager though I'm sure it could be made better. And, yes you
> could replace mwm with some else if you preferred...
I think FVWM is much more customizable than MWM, but there are some
religious issues regarding window managers :-).
> 4) An efficient environment that is responsive to users. I myself
> dumped KDE and switched to mwm because I was tired of waiting 2-3
> seconds for a kvt terminal to come up. Under mwm my mxterm terminals
> come up immediately and it makes for a much more pleasant experience
> (and I like mxterm better than kvt too).
Xterm is still much better that any of the proposed replacements, most of
them based on RXVT.
> 5) An easy way for the user to interactively customize the desktop
> environment. Mostly this would be the ability to set color schemes
> in an interactive way. Perhaps this could based on the Resource
> Configuration Management (RCM) scheme (See the X11 release notes).
> I'm *not* suggesting abandoning the concept of X11 resources. I
> think they are powerful and useful and I believe RCM is based on the
> use of X11 resources.
X resources are good enough, but we would need some standardization on the
resource namespace to provide the necessary configurability for all
applications.
> Okay, that's a start. Does it sound a little like CDE?
Yes, it sounds like CDE.
> As for the future of Lesstif... Unless the project team feels that it is
> very close to having a compatible CDE/Motif 2.1
Lesstif is still far from Motif 2.1 compatibility and CDE compatibility.
> then I think that the best thing to do would be for the project to
> refocus its efforts on projects which would help Motif in general
> become more widespread. The Lesstif project has a lot of talented
> people working for it and I think that expertise should be focused on
> making Motif (via OpenMotif) better through advocacy, bug-fixing,
> application development, development of the Motif Enabled Desktop,
> etc...
I like your idea. Taking Gtk and Qt as examples, their developmet was very
improved by by the existence of GNOME and KDE. Ordinary people want
applications. Only developpers need toolkits.
> I'm not suggesting that all work on Lesstif stop. I'm sure you could make a
> case where having Lesstif in addition to OpenMotif would be useful, but at
> this point in time I think the important thing is that Motif in general be
> moved forward.
Humm, more religious issues here.
--
Carlos A. M. dos Santos
Federal University of Pelotas Meteorological Research Center
Av. Ildefonso Simoes Lopes 2791 Pelotas, RS, Brasil, CEP 96060-290
WWW: http://www.cpmet.ufpel.tche.br RENPAC (X.25): 153231641
Phone: +55 53 277-6767 FAX: +55 53 277-6722
