Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!sharkey!itivax!umich!dip.eecs.umich.edu!anon
From: anon@dip.eecs.umich.edu (Omar S. Juma)
Newsgroups: comp.windows.x
Subject: Re: Gadgets in Motif
Message-ID: <232@zip.eecs.umich.edu>
Date: 16 Aug 89 19:39:40 GMT
References: <322.8908161255@jupiter.cmi.no>
Sender: news@zippy.eecs.umich.edu
Reply-To: anon@dip.eecs.umich.edu.UUCP (Omar S. Juma)
Organization: University of Michigan EECS Dept., Ann Arbor, MI
Lines: 45

In article <322.8908161255@jupiter.cmi.no> pererik@KHEOPS.CMI.NO (Per-Erik Nordbo) writes:
>Gadgets versus Widgets in Motif
>-------------------------------
>
>Motif has introduced the user interface object "Gadget" as an alternative to
>"windowless" Widgets ( arrow, button, label, separator etc).
>
>Is the advantage converting a Widget into a functionally equilavalent Gadget
>primarily speed, memory savings or both? Is this a Motif-only toolkit extension?
>
>============================
>Per Erik Nordboe
>Chr. Michelsens Institute
>Norway
>email: pererik@kheops.cmi.no
>============================

I believe the trend in UIMSen [the new unofficial plural of UIMS] is towards
windowless interface objects.  You're not likely to save memory; the allocation
burden is now moved to the client, not the server.  You are, however, relieving
the server from tracking hundreds of tiny little objects all over the place.

Giving the client full tracking "rights" enables a variety of highlighting
methods (maybe even none).  It leaves the server more time to do what it does
best: drawing text and graphics in windows, and handling device input.  All
other higher-level functionality is the domain of clients.  That is the correct
abstraction, until somebody decides to write an interface server.

That's one small step forward.  Unfortunately, we are now entering the tail-fin 
era in X user interfaces, and for a while, there will be too many garish,
eminently forgettable UIMSen.  But remember, even though some of us look back
at pink Cadillacs with fondness, it's very unlikely than anyone will want to
exhibit any emotion over the demise of the xedit-xmh school of user interface
design.

By the way, could someone from OSF provide information on the choice of "Motif"
for a title?  How did OSF get away with not naming it "Open Motive"?  :-)


Omar S. Juma
Computer Aided Engineering Network (CAEN)
University of Michigan, Ann Arbor MI 48109

phone: 313-936-2486
email: anon@caen.engin.umich.edu