Re: measuring linux usage

From: Justin Mason (jm@jmason.org)
Date: Thu Mar 29 2001 - 03:21:45 PST


Josh Cohen said:

> oh please.
> If we're talking about "bonafide" real companies using it,
> the license to machine ratio is low, since the support
> from redhat (or other) is important to them.

I don't agree. I've never felt the need to buy Linux support from Red
Hat, and, judging by my local LUG, I'd say most of the installed base of
Linux machines (in Ireland at least) are in the same boat.

(This probably reflects more on the support I'd expect to get from Red
Hat, though. We did try RH support on 1 occasion, but got no useful
reply, and eventually resolved the problem ourselves using net searches, a
well-directed question to the LUG mailing list, and some of our own
debugging. There's no point paying for support, unless, in your
judgement, they'll provide help -- so we didn't.)

Most of the cases I'm talking about there are reasonably small companies,
though. For an example of a very large-scale, company-wide deployment,
Google would be a good example; as far as I know they get their Linux
support from VA Linux, not Red Hat. I'm sure there's others using
LinuxCare, too.

The point being: in the free OS field, OS support is not necessarily tied
to the vendor, and in some cases is better done on a DIY basis, so it's
not going to show up on *any* company's books apart from your own. So
it's not a good way to measure Linux usage.

--j.



This archive was generated by hypermail 2b29 : Fri Apr 27 2001 - 23:15:09 PDT