"Tony B. Shepherd (Tony B. Shepherd)" wrote:
...
> These days the phrase "low-cost high-volume standards-based" brings to mind
> an image of a desktop PC. If it were up to me, I'd get a few PC's (probably
> Compaq Presario type - middle to high end consumer or SOHO models) and some
> copies of Linux and see what I could come up with that worked. HP3000
> System Programmers are very creative - swimming upstream for something
> worth having makes you strong. Once I got some apps to work on Linux (how
> much more difficult than converting to HP-UX, really?), I'd be a lot less
> concerned about relying on any particular vendor in the future.
>

Can HP take big risks anymore?

If hp were the innovator they and we would like to believe in,
they would grab the opportunity that lays waiting for someone to
deliver a real "ready for business" e-application platform for
the 64 bit world.  This (what Wirt calls TAOS) is what MPE should
have become, with due consideration for migration compatibility,
but without letting the compatibility get in the way of building
the best business platform around.

Business customers don't care about Unix or Windows, they care
about applications that work to do the job they need and they
care about protecting against risks.  They will spend millions
of dollars on the latest ERP/CRM/Buzz-of-the-dsy in the effort
to make their business more competitive.

Lunix is a growing force to be recognized and Windows holds the
huge mind share, but mainframes are *still* depended upon for
really large scale business computing.  The 64 bit world will
turn the cost of hardware topsy turvy over the next several
years.  Yet, business will _still_ require good applications.

With commodity hardware and a quick rate of change, to be
cost effective a new platform would have to hitch it's device
support wagon to another train, such as Linux.  The very high
cost of keeping up with new devices was one of MPE's current
problems that weighed down the train. Innovative, cost
effective approaches are what we ought to expect from HP,
not doing the migration thing the same way as last time
(Spectrum).

It's no longer about the hardware, it's about what value
add you can bring.  Business and application vendors invest
in designs tied to Oracle, in spite of the "open systems"
mantra.  HP could build a highly scalable, cost effective
platform for business applications that was even easier to
run than MPE is now, come with a built in and easy to program
database and all the "built in" business features of MPE today,
utilize Linux for selected device services and provide a Linux
cooperation layer to run Linux applications natively and to
support a select set of the best of the open source tools.
This platform could become wildly successful given good
marketing and good, consistent, really cheap, support for
application vendors of all sizes.  That success won't come
next quarter or even next year, but five years down the road.

Plus, give it an MPE/iX 7.5 compatibility layer to be able to
bring current MPE applications to it. Give it an OpenVMS
compatibility layer to catch those applications.  Charge a
bit extra With these compatibility layers, but you start out
with an existing set of applications and vendors interested in
your success.  Finally, go out and tell the business world loud
and clear that you've built a better way to run their business
and those IT guys can now spend their time making your Windows
machines work instead of making your business run and that the
business innovators can build custom applications, too.

Is is risky, yes of course, so was building MPE and the Laserjet.

How do you make money? Charge the application vendor that uses
it for each sale, and allow customers that what to build their
own to license it directly, like Oracle.

What does it run on, perhaps anything Linux runs on, but HP
selects hardware and devices for which it offers support and
integration.

One of the major features would be integration.  The application
vendor or HP or the hardware vendor could offer integrated
bundled dedicated solutions, in some cases configured for the
customer's enviornment right out of the box.  The goal is to
minimize IT involvement and make it simple, easy, reliable,
and operate at the performance the customer needs.

Is HP innovative enough to pull it off?  Is someone else?
(What do you think all the Enterprise oriented Linux
distributions are about?  What is IBM doing with Linux?)

Richard
--
Richard L Gambrell, Senior Information Technology Consultant and
Director of Computing Systems and Networks
Information Technology Division, University of Tennessee at Chattanooga
Fax: 423-755-4150                Support Help-Desk: 423-755-4000
Direct phone: 423-755-5316       ITD Business Office: 423-757-1755
Mobile (urgent): 423-432-5122    Main UTC: 423-755-4111
Email: [log in to unmask]

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *