MOTM

At 10:58 PM 2/10/2004 -0600, Paul Schreiber wrote:
>h) Looking ahead: I've learned my lesson. I'm not going to announce *anything*
>until it's ready to ship. I mean, sitting here, with manuals, etc. ready to go
>out the door. So, it may be 8 months from now until something 'new' shows up.
>But, I think that's the best way to set expectations going forward.

Paul,

Thank you for moving to this model.  I know it is very hard and every 
engineer loves to share about the cool thing(s) they are working on but it 
really comes down to setting proper expectations to get the best customer 
satisfaction.  Folks will wait a long time for stuff they want, witness 
your customer base and backlog.  The problem is that schedules are 
notoriously hard to keep accurate the further out they get.  I know, I live 
it every day managing SW engineers and large engineering projects.  So, the 
only thing you can do is to announce stuff when it is ready or clearly on a 
schedule that you have 95% or greater confidence in.  This will also mean 
perhaps a bit more speculation on your part on what your customers want, 
although there are ways to address it through careful surveys.  We will 
lose the opportunity to comment more on new designs and panel layouts but 
IMO, that is not a large loss in order to have greater predictability.

         Eric