[SASAG] Proprietary formats on public mailing lists

Jeff Silverman jeffsilverm at gmail.com
Mon Sep 3 01:22:07 PDT 2007


It is always interesting to read a flamefest from SASAG members.  We are 
a very intelligent group and the arguments represent a fascinating 
insight as to how computing should be done.

I am as rabidly anti-Microsoft as they come.  However, the world as we 
know it has embraced Microsoft, for better or worse.  This is not going 
to change for the foreseeable future.  Nevertheless, I think we should 
continue to stick to our views, even though they are divergent.  My 
thinking is as follows:

In our complicated, technologically sophisticated society, we hire 
experts to help us with tasks that are too complicated or arcane to do 
ourselves.  When we are sick, we see a doctor; when our car breaks down, 
we see a mechanic; when we need our taxes done, we see an accountant.  
It is fitting and proper that when people have computer problems, they 
should seek.... us.

Now, just as physicians disagree about the best way to handle certain 
diseases, sysadmins disagree about the best way to solve computing 
problems.  Note that 'best' is itself an ambiguous term.  A certain cure 
might be 90% effective but have a 1% mortality rate, while another cure 
might be 70% effective but have a 0% mortality rate.  Which is "best"?  
You don't know.  You have to explain this to the patient and let him or 
her decide.

There are many issues which have been brought up in this discussion, 
including reliability, safety, avoiding vendor lock-in, convenience,  
acquisition costs and life cycle costs.  Before we can undertake a 
project, we have to explain to the customer what the options are and 
what the relative risks are and then let them decide.   My observation 
is that most customers could not care less about vendor lock-in.  Most 
customers don't care about Digital Responsibilities Management (DRM) - 
at least, not yet (as Vista becomes more ubiquitous, this might 
change).  Most customers already have MS-Windows machines, so they want 
software that runs on Windows.  They want Windows because the software 
they want runs on windows.  That's what they care about.  That windows 
is buggy, unreliable (by design), and insecure means nothing to them.  
Even after they have been hit by a virus which has wiped out their data 
or exposed their bank accounts, they still want Windows.  To most 
people, a computer is something that runs Windows.  Sad but true.

Now, what we can do is make options available to them.  They don't have 
to use windows.  Linux is an option, as is HP-UX, Solaris, and AIX.  
Macintosh is an option, and the Mac has its own rabid fans.  As one 
poster pointed out, they don't need something that looks like a computer 
at all - they can use google and other net based software services 
through a browser.  In order to support these options and make them 
viable, we ought to discourage people from using proprietary formats.  
We should be prepared to fail in this regard.

In order to make options available, we should - no, we must - agree to 
disagree.  It is very difficult to be proficient in more than one 
software ecosystem, so we have become sub specialists.  We must be able 
to do the things that need to be done in the ecosystem we are most 
familiar with.  The debate will continue at many levels - this is the 
way it should be, IMHO.

Somebody suggested that we contact Microsoft and suggest that they 
include native support for ODT, which would be simpler than getting the 
world to embrace a new format.  This sounds like a wonderful idea - how 
do we do that?

Sincerely yours,

Jeff Silverman

More information about the Members mailing list