[CLUG] Desktop Linux Share Overtaking Macintosh

[Mark Twomey]

On Feb 14, 2004, at 5:15 PM, adam beecher wrote:

> I think you're still kind of missing my point Mark.

Probably. We do appear to have some form of communication disconnect on 
this.

>  I'm not suggesting that
> a stat like this should be used to convince a technical chief of the 
> merits
> of Linux on the desktop, more that it would be useful for a technical 
> chief
> /that's already convinced/, to convince the bean counters.

Well like anything when it comes to bean counters it's a matter of if 
it needs to be done and if the price is right. If things appear to be 
"okay" then no matter how cost effective you can prove something to be, 
or how good an idea it might be, bean counters won't go for it as they 
are wary about such a change being destabilising or in someway 
negatively impacting the business. The flipside of that is when they 
believe change has to happen and will force bad ideas even if the body 
of technical opinion believes that such a thing will be disastrous. It 
happens more frequently in large organisations than a person would 
imagine.

> Moreover, I'm not saying that a stat like this is going to be a 
> clincher,
> just that it's a (very) useful one to add to a body of evidence to be 
> used
> to support the rollout of Linux on the desktop in an organisation.

I can honestly say that such a factoid wouldn't matter a jot. Linux 
taking out Windows can be sold as a proposition in part because the 
existing hardware investment can be protected. You can't say the same 
thing about Macintosh. To put OS X on everyone's desk you need to buy 
them a Mac, that's why there's never been and probably never will be a 
mass market conversion to OS X in the same way there will be to Linux 
as time moves on.

> I say "very" because, sadly, bean counters often seem to react more
> favourably to meaningless stats than they do valid arguments that 
> don't have
> a punchline.

Yes they do, and the biggest stats you can throw at them involve a 
large short term retooled support cost as well as employee retraining, 
offset against the cost saved by Windows site licenses. The thing about 
large organisations is that they don't just standardise on a platform, 
they also standardise on applications and for the most part amongst 
those applications is MS:Office. Probably one of the largest hurdle 
Linux faces on the corporate desktop is the fact that it doesn't run 
MS:Office. Sure it has Open/StarOffice but to the bean counters that 
may as well be Lotus Smartsuite or something else they rejected years 
ago.

Mark.