I just spent the last two days fighting a fire to get a project
ready for a demo.
Well, I now know why people say "Write once, test everywhere"
If the code you write is not correct, sometimes it still produces
what seems to be correct results on some platforms.
For example, sometimes putting menus into an applet works because
the platform implements menus as just a button with a popup.
(As motif does it) and not like windows, where it is a special
other type of widget. But the applet spec (and the applet frame)
specifically talk about the fact that you can not have menus,
only popups.
There were a number of such issues that were at the root of the
problem for the demo applet on the various platforms. Each one was
traceable to code that was "not to spec" but happened to work on
one platform or another.
As some old Amiga engineers know, if the developer can not tell
that the code was written incorrectly when they run it on their
system, they assume it is correct code. (The Amiga examples are
many, with the earliest one being the case of MEMF_CHIP memory
allocation flag. CHIP memory was memory that the graphics
processors could access and all graphics operations had to be
in this type of memory. 99.9% of all early Amiga systems only
had CHIP memory - few people could afford to buy the rare 'FAST'
memory for the system - thus many programs just allocated memory
to do the graphics and when that program was finally run on a
system which had other types of memory and the allocation did
not specifically ask for MEMF_CHIP memory, the program failed,
sometimes rather drastically, sometimes just visually)
Anyway, the lesson here is that Java needs a way to validate
the correctness of the code. A "pendemic" operating mode that
flags as many not-to-spec operations as possible. This would
address most of the "test everywhere" problems since you would
only need to test in the "strict" environment.
I wonder if such a tool is available or if such a tool would
be worth money :-)
Michael Sinz -- Director of Research & Development, NextBus Inc.
mailto:[EMAIL PROTECTED] --------- http://www.nextbus.com
My place on the web ---> http://www.users.fast.net/~michael_sinz
----------------------------------------------------------------------
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
