: AutoDesk's practices have basically "dumped"
: licenses on the US market, and anybody who can't see that is an idiot or
: just likes to kid themselves.
Seems like they are in the business of kidding the RP people. They're the ones
study is aimed at and the ones who use the figures for making decisions on
expansion. 'Oh, yeah, guys, go out and buy that half million dollar Super SLA
machine. Look at these numbers, there's a ton of work in the pipeline, judging by
this jump in solid modelling licenses [points to chart with spike on it]. No, no,
those machines won't be sitting idle for months at a time, sucking the life out
your bottom line. We're absolutely, positively almost certain of that, judging by
the numbers.' Everyone loves optimistic predictions and when they don't come
who remembers the ones that made big bucks selling you on the "next big thing".
On the validity of the job figures for judging installed base, consider a couple
things. First, the numbers (50, 100, 300, whatever) represent a very, very small
portion of licenses in use, like .02%. Even a small fluctuation in something like
retention rate (job hopping, layoffs) can dramatically change the numbers, easily
doubling or halving them. Second, comparing the numbers of jobs available assumes
that one seat of brand x is the same as a seat of brand y. But what if licenses
among SolidWorks users (or Inventor or Alibre) were mostly by individuals; what
the vast majority of Pro/e users worked for companies like Caterpillar with 3000
licenses. Which of these is going to turn up on Monster? Not the oneseys and
twoseys. They don't show up on Monster or in the unemployment figures either. It
also effects what proportion of licenses have been handed to agencies where the
highest turnover is and where the bulk of jobs appearing on Monster, Yahoo Jobs,
Net-temps come from. So, the numbers are probably skewed, but not in a way that
favors Inventor over SW. They're probably skewed in a way that makes Pro/e look
like it's more than it is.