Re: Mini-ITX PC's a the future of HA

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

"Dean Roddey" <droddey@xxxxxxxxxxxxxxxx> wrote in message

<stuff snipped>

> Just for the record...
>
> One big reason our product is so good is because we are software people
and
> not automation people. In many cases the best products come from people
are
> really into something and then implement a great product based on their
own
> knowledge as a user of such things.

I think that was true in the beginning when getting an appliance module to
turn ON and OFF by PC control was a novel idea.  I watched lots of subject
matter experts who were sharp business pros in their own field thrash into
bankruptcy because that wasn't enough.  Super sharp PC skills were needed to
stay competitive as soon as the Pentium class chips appeared and PCs began
blowing the doors off minicomputers in the business environment. Now people
were beginning to have serious expectations regarding ease of use,
interoperability and reliability in software.  That took skill.  It took
constantly evolving skills as the HW platform evolved at lightning speed,
often changing the entire playing field in less than a year.

> But when you are talking about software on the scale required
> for this type of product, if you don't have the hard core software
> engineering skills, no amount of knowledge of the problem
> domain will help.

Well put.  I enjoy our discussions because I never doubt for a minute you've
given great thought to the all the issues facing the modern software
designer.  At first I thought you were too software-centric but I've come to
realize your model is entirely appropriate for an ever-shifting hardware,
firmware and OS landscape.  A high level of abstraction allows for a lot
flexibility.

> You obviously do need someone who knows the problem
> domain, but your customers are of those types of people
> and they can tell you what needs to be done. What you need
> as the implementer is very strong software engineering skills in
> order to make it so. OTOH, if you already know a lot
> about the problem domain and less about software engineering,
> your customers cannot tell you how to write the software.

Well, in my experience, customers will tell you anyway, but I get your
point!  The best software engineer is like the best film maker or attorney
or CPA.  Give them a good client that can explain the intricacies of the
subject at hand and they can then use their detailed professional knowledge
to secure the best outcome.  It's also why I laugh a little when every so
often some newcomer announces the latest, greatest HA program ever written
looking for front money from beta alpha and beta testers, as if it were a
privilege and not a Herculean effort to design, code and debug application
software.

> I guess it also has the benefit of us never telling our customers,
> no that's wrong and we know more about automation than you
> do and know we are right.

That is an interesting side benefit.

> We don't know more about automation than our customers
> and we know that and therefore we have no ego tied up in
> doing it this way or that. We are happy to do what can be
> done within the practical limitations of what we can
> accomplish, if that's what our customers need.

Hmmm.  I could make a pretty strong case that most progress came from people
who believed that there were NO limitations to what they could accomplish if
they kept at it.  There was a great program on just recently about Newton, a
man who just figured things out and saw answers to things that people didn't
even understand as questions.  I realize you have a business to run, so
dreaming's probably left to the idle rich, but I want to know when the real
smart house is coming.  Not this loose collection of electronic fiefdoms
just barely out of the Dark Ages.

> OTOH, we know very well what we can and cannot do (or should not do for
> strategic software reasons), and we know that the software is the only
> reason we are here and won't risk its long term stability and robustness
for
> a quick buck, and end up with a piece of junk in the long term. I think
that
> this is too often the case when the people running the company just know
the
> problem domain, or business, and consider the actual product something to
be
> jerked around as required to make money as fast as possible.) They seem to
> forget that the only reason people are coming to them is for the product
and
> that it must remain high quality.

That's what all you "quality first" guys say until you get rich!  :-)

I'm having trouble even thinking of three manufacturers that still put
quality over profits.  There's this guy who makes a couple of pianos a year.
My friends just bought one for $80K so I assume he's a quality-oriented kind
of guy.  Jeff Volp's a quality guy or a soldering machine - can't tell
which!  :-)  But big companies?  The ones I would have called "quality" 10
years ago have eroded mightily.  The micro-Benzes are but one example.

--
Bobby G.