[vox] OS/2 and Linux, why has IBM changed?

[Michael Wenk]

On Thursday 08 May 2003 10:31 am, Peter Jay Salzman wrote:
>> On Thu 08 May 03, 10:03 AM, Michael Wenk <wenk@praxis.homedns.org> opined:
>
> what exactly was IBM's killer product in the mid 90's?   nobody used
> typewriters.  nobody really used OS/2.  other companies that provided
> cheaper PC's were flourishing because the average person was less
> knowledgable about what "quality components" meant.  the only thing i
> can think of is the high end mini and mainframe market.  but by the mid
> 90's, even that was out of vogue, and can't support a company with such
> a big overhead (as you point out).
>
> i can't think of anything that you could point to and say "IBM is king
> of this market" in the mid 90's.  i think they see linux as providing
> them with the opportunity to have a "killer product" again.

IBM doesn't need to have a "killer product", and I disagree with you in the 
sense that any OS is a killer product.  Its the software that runs on that 
OS, such as MS Office, or Oracle, etc that is the real killer product. All 
the OS needs to do is service these apps and do so in a stable manner.  
Anyways, back to the killer product thing.  Taking a look at the last 
quarterly report via IBM(1-2 quarters ago) IBM had a 83.9% gross profit 
margin in software.  The overall gross profit margin was around 23%, with an 
effective profit margin of around 6.4%.  These #s are incredible considering 
the overall economic climate in the US and worldwide.  These are not the #s 
of a company with very many problems(minus overhead and part of that is 
probably due to necessary expenses.)  

> i agree with your other reasons why they're behind linux, but i think
> this is their primary reason.
>
> > They are doing what IBM customers want, and they
> > will likely do it with quality as well.  Also, if you look at IBM's
> > businesses, they are making money on all of them, but the greatest margin
> > they have is in their software business.
>
> which didn't really take off until the late 90's, after IBM already
> started making overtures to linux.  i can't recall much marketing till
> their "ebusiness solutions' commercials (which is very well done, imho).
>
> > They are also doing well in the
> > hardware and consulting, and financing portions.  The only real problem
> > IBM has is they are so friggin big, their overhead is ungodly.  I was
> > surprised by IBM's quarterly report, especially given the current
> > economic climate.
> >
> > Mike
>
> i actually do see IBM as being confused, even today, and have a few data
> points to back that up.  they have a number of software products that
> could've easily have been something that we all have on our systems, yet
> nobody knows about them because marketing has been so poor.  for
> example...
>
> in my opinion, their techexplorer package could be the answer to the
> entire academic community's dreams.  the ability to put our research on
> the web without the ugly, stupid, and time consuming latex2html.  we can
> put our entire research on the web without modifying anything, and with
> very small downloads.  it could be big.  but every other researcher i
> talk with is like "huh?" when i mention techexplorer.

I'm not really familiar with this one.  I know that when I worked as an admin 
for a univ, it was sometimes difficult to get IBM to come down on price.  I 
know that both HP and Sun were willing to give big discounts where as IBM was 
less willing.  Least that was my impression(I wasn't all that involved with 
cost issues.)  IBM does have problems, and most of it is in the big company 
mentality, and face it, most companies don't really look at academia as a 
large area of growth.  They invest in it as a way to further R&D and 
workforce improvements.     

> they dropped the ball on viavoice, too.  there's currently no mature and
> compelling voice recognition software on linux.  it all sucks right now.
> i actually bought viavoice for linux, and it totally blew chunks.  some
> bonehead made the decision to use java, and the thing is too dog slow to
> be useful.  they eventually discontinued it (i think).  they should've
> realized their product sucked and fixed it rather than hide it under the
> carpet.  it could've been big.

I imagine the sense that Linux is more a server OS vs a workstation OS would 
have much to do with that.  I'm curious as to how things like Tivoli work on 
Linux, and also how well upgrade issues betweeen their old mainframe OS and 
linux.  

> not to mention they didn't even bother writing a decent bug free
> installer.
>
> i see lots of software that had the "right idea" but was executed and
> marketed poorly.
>
> pete

In many ways, all companies out there have this problem.  But looking at IBM, 
it looks as if they are a pretty healthy company.  They have problems, but 
they're big in the black and it appears that they're doing their damndest to 
stay that way.  

-- 
wenk@praxis.homedns.org
Mike Wenk