Nick,
If you use a proprietary system like a Mac running Mac OS X or a Windows
PC, and you aren't a person that has reason to know the semantics of
internal interfaces (a.k.a. APIs) there really is no recourse but to
seek support from the vendors involved, or online support groups.
A second approach is Doug's guerrilla like tactics as with Google. Just
beat them until they give you the answer you want. Most users expect
things not to work, so there's only so far that can take you. There's
only so much outrage that can be generated.
In your case, it sounds like the problem was the Layered Service
Provider interface that Microsoft provides and how it interacts with
some other product trying to intercept that traffic (see the table at
http://support.apple.com/kb/TS4123). Sometimes the developers of one
of these intermediate products will be motivated to debug the problem,
other times you'll need to appeal the the app vendor (here Apple), or
the operating system vendor (here Microsoft). Do a little work with
Google, and the support websites of the likely vendors involved, and
you'll find the answer almost every time.
The third approach is to make it your responsibility. With Linux, there
is source code to the whole thing. Tens of millions of lines of code.
It can all be rationalized. While it is true that few people have the
depth and breadth to understand all of these things, the beauty of the
free software community is that you can almost always find that expert
and someone has likely had the same problem and analyzed it, and _to the
bottom_. Not just in terms of vague phenomenology as with so many
Windows or Mac problems, but the the particular line of code with a
mistake.
I just don't understand how people who use or write software for a
living, especially scientific software, would ever tolerate using a Mac
or Windows box. I won't tolerate being helpless to vendors who make it
hard to understand how their software interacts with other software.
Well, let me qualify that, I won't tolerate being helpless when it
matters and I can get my way. I don't mind using a Mac or Windows box
for entertainment, for example. And I'll use Microsoft Word or
Powerpoint if collaborators want to use that. Those things don't
involve `real' problem solving -- at worst solving problems
with`presentation' issues can become an annoying distraction. Any
interesting program will have bugs, and any interesting program runs in
concern with a lot of other programs. If bugs exist in `secret'
components, we'll you're often completely powerless to do anything about
it.
Could I tolerate some obstacles and `secret' components? That is,
tolerate intellectual property of software companies? I often could,
but I think it is better if I don't, and will also try to persuade more
people not to tolerate it either!
(I mostly sit in front of Macs, but I do all my work on Linux machines
either over the network or in Linux systems in virtual machines.)
Marcus
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
