MathGroup Archive 2009

[Date Index] [Thread Index] [Author Index]

Search the Archive

Version 7.0.1 bug with initial Palette positions

  • To: mathgroup at
  • Subject: [mg97191] Version 7.0.1 bug with initial Palette positions
  • From: Bob F <deepyogurt at>
  • Date: Sat, 7 Mar 2009 02:41:40 -0500 (EST)

I downloaded the 7.0.1 update and started it up and it didn't place
some of the palettes correctly when starting up. I have a dual-screen
Mac setup and I put all my palettes on the left screen, and Notebooks
on the right primary screen so I have visibility of my notebook
windows at all times and not obscurred by the palettes that don't like
to cohabitate with normal windows (something I still find to be a
royal pain -- why cant they be like any other windows - yes I know you
can change their floating attributes, but the whole idea of not
behaving like any other window is so contrary IMO, you can't resize
them, you cant minimize them, etc).

So in this case the WindowMargins parameter in the init.m setup file
had a negative left and top number and Auto for right and bottom. But,
for some strange reason only the ClassroomAssistant palette was
incorrectly placed on the left screen and before with 7.0 and 6.0 and
5.2 things were fine (all the other 4 or 5 palettes came up in the
expected position which is even more bizarre actually - I can only
guess that the code that does the placement of the palettes is
individually isolated to each and every palette and not a general
purpose routine that all share, otherwise all would exhibit the
problem cause they are all on the left screen). The most bizarre
problem is that sometimes one of the other palettes would do this and
not the Classroom assistant, but not very often, and certainly not
repeatable as far as I can tell.

The really ugly thing is that the ClassroomAssistant palette was
positioned so that the title bar was underneath the main Apple/
Mathematica menu along the top of the primary screen and no matter
what I did, I could not move it. It's title bar was underneath and
totally not reachable.

What I ended up doing was using the OptionInspector to change the
WindowMargins parameters so that it was lower on the primary screen so
that I could drag it where I wanted it on the left screen, or just
change it to what it should have been to begin with. But when exiting
Mathematica and then restarting, the same thing occurs all over again.
WHAT A PAIN!!!! And on top of that the init.m file is rewritten on the
fly when Mathematica starts up to change what it considers as an
invalid margin setting and changes them to what it thinks are valid,
but no message ever indicates that it just mucked with the settings in
the init.m on it's own prerogative for this ClassroomAssistant
palette, so you are forced to move it to where you really want to
every time you start up Mathematica as I described with the Option
Inspector and I spend a minute or two every time Mathematica starts up
to fix a problem it caused!

I can't believe this simple user interface issue escaped the QA
process, and the beta testers. The only thing I can guess is that a
two screen setup is not common. When I called Wolfram to see about
this, the person in support was able to locate a dual-screen Mac and
verify that the problem occurs there also. So they have the ability to
test dual screen setups, but guess they either didn't do at all, or
not very extensively. Very disappointed in my first excursion into
7.0.1 land -- but I will get over it. Now I will only have to wait 6
months or maybe a year and the fix will be available. And I pay money
for this??? In the meantime, I know how to get around the problem -
thank goodness for the option inspector.

Why oh why oh why do people put up with this kind of software
problems!! Wolfram get your act together and come up with some way to
get patches to people on a timely basis for when you create bugs and
problems and don't catch trivial bugs you create in releasing new
versions, let alone the hard ones. This is not the first time this has
happened, far from it. I am sure everyone on this list has stories
that would cause us all to shake our heads in familiar disappointment
and belief. If people feel the same, please let your Wolfram sales and
support contacts know how disappointing this kind of problem is and
encourage them to come up with a solution.

That Mathematica is such a powerful tool is it's saving grace, but
Wolfram -- wake up and smell the coffee! Remember a company called
Quark, who was so world famous about not giving a rat's behind about
customers problems -- now Adobe has virtually killed them off because
people gave up on Quark ever getting better. Please, WOLFRAM, do a
better job of QA and do a better job of coming up with a way to solve
bugs, rather than wait for months and years and the next release.
THERE HAS TO BE A BETTER WAY. Solve it and you will have even more
loyal and fanatic customers.

Of course, I realize that the problem is not rampant, but a relatively
rare, and certainly Wolfram is not the only one suffering from this
problem. It only becomes such a frustrating issue when it affects
things in what I am doing or try to do, and I say to myself that this
ought to work, that I get so disappointed. And I am sure that we all
have shared in this frustrating experience before.

But the fact that it happens again and again and again is the part
that completely baffles me and makes me think how could they shoot me
in the foot again and think I wouldn't say ENOUGH IS ENOUGH!!!! I am
so disappointed with bad, careless, and sloppy programming - why after
about 60 or 70 years of the "science" of programming and all the
advancements of computer hardware, do things like this continue to
occur? It is not just Wolfram, but rampant in the entire field of
software. The complexity of code today and the difficulty in solving
trivial as well as difficult problems is hardly any better today than
decades ago. Since the overall complexity of software solutions is so
many orders of magnitude greater, the only saving grace is that the
hardware is so much faster as well. Otherwise we would be in a truly
dismal position.

But I am sure there is a solution for this problem, and Wolfram does
have some of the most talented and resourceful and intelligent people
in the world. My wish is that they solve some really fundamental
problems so that customers don't have to continually have to go thru
this frustration on every release and every bug, but instead remember
and chant to themselves

    Just wait for the next release - the problem will be solved, or at
least moved to some other, hopefully unimportant, spot.

In the meantime, just don't do it that way, no matter how much you
think that it should work or not work that way. Adapt and be happy, as
the alternative is rather miserable.


  • Prev by Date: Re: Re: "Do What I Mean" - a suggestion for improving
  • Next by Date: Re: When a string does not match itself
  • Previous by thread: Re: String substitution system
  • Next by thread: Re: Version 7.0.1 bug with initial Palette positions