Exactly, we've been through 6 systems in 20 years, with no recompiles for
old stuff, no database conversions etc., all on our in-house code.  This
always allowed us to move forward, at a constant orderly revision cycle with
out constantly revisiting old finished stuff until business practices
required, and definitely not on the schedule of our hardware/Software
vendor.

The main reason we would never have considered any other vendor.   Like the
old days at ibm, dec and so on each transition caused these massively
painful migrations, even within a family ie system 34, 36, 38, as400, or the
pdp.   Both IBM and DEC finally learned HP's lesson now with IBM's AS400 and
dec's VAX family, and HP is forgetting it :(.   I agree the right thing for
HP to do at a minimum to make this last jump just as painless, we paid them
the premium they deserved for all this time, and we would still pay a
premium for this to happen one more time.

We have is that we've build the infrastructure of our company on the solid
base that was the HP3000.  Not the on current fly by night fads that last a
couple of years only to be replaced by the next silver bullet candidate.
i.e.  RDMBS, ODBMS, ISO/MAP, RPC, EDI, DCE, OLE, COBRA, PASCAL, C++,  Visual
Basic, JAVA, J2EE, XML, XML/RPC, SOAP, .NET and so on.  Anyone chasing these
fads and ignoring the fundamentals will just follow the other lemmings over
the cliff, ever converting, never completing never competing.

If you examine all these trendy paradigms, they end up working good the the
simple example cases, but in the real, complex world end up looking more and
more similar to the same old code, with extreme overhead, and in some cases
even more extreme effort to produce exactly the same results.



"Michael Anderson" <[log in to unmask]> wrote in message
news:a373eq0i8v@enews3.newsguy.com...
> I haven't been reading much on the list lately (great listserv, just been
> busy), so I may touch on a thing or two that has already been discussed. I
> just want to spill my guts here, and maybe get some good ideas or
> direction from others on the list. Main reference to HP-e3000 users,
> businesses that have invested 10, 20, or more years into developing their
> own business applications on the MPE/IMAGE/VPLUS/COBOL platform. Are they
> looking at a complete re-write within the next 10 years?
> HP has claimed to offer a sound migration path to HPUX, and/or NT. HP is
> advising its customers to begin evaluating and planning transitions from
> the HP e3000 platform to other HP servers. To keep customers informed
> about options available to them, Hewlett-Packard has developed something
> called a 'HP e3000 transition web site'. I am admittedly bitter about
> 11/14/01, and may make some statements that are a little less than polite.
> However, my statements are valid and true for any and all companies that
> have invested many years into custom application software development, and
> I'll refer to them as 'These Companies' while I'm writing.
> On this 'Transition' web site HP states that the range of applications,
> tools, and resources available on HPUX is extensive enough for nearly
> every business need. What this means for 'These Companies' is, if you have
> NEVER purchased third party software, then you need to start doing so now
> to properly plan for your forced migration from the HP-e3000. HP has also
> developed something that is referred to as the 'HP-e3000 transitions
> partners program' Again, for 'These Companies' this means, don't call us,
> but for your convenience, here is a list of other companies to call
> instead. I think HP is making an assumption that all the HP-e3000 users
> have purchased application software, and that companies like Ecometry,
> eXegeSys, Mitchell Humphrey, and other software vendors will provide the
> migration path for HP-e3000 users. This could just be my own twisted way
> of thinking, but is true to me anyway. I think that the NEW Hewlett
> Packard is using carefully thought-out wording to avoid doing the right
> thing. I don't think HP is doing the right thing at all. What has HP
> actually done for 'These Companies', Hp-e3000 users who don't buy third
> party application software? What about 'These Companies' that have been
> confident enough, and relied upon HP, enough that they have spent years
> developing custom software, specifically for their own use, on the
> MPE/COBOL/IMAGE/VPLUS platform. I know of a Mfg Company that has twenty
> years of development invested into this platform. If HP is going to put an
> end to the HP-e3000, then 'why O why' don't they port Image, Vplus, and
> all the various compilers used on MPE, including Intrinsics (SL, XL), and
> CI.PUB.SYS, to HPUX, and/or NT. Then a step further would be to port some
> of the MPE (FOS) to HPUX, then, maybe the migration could get back to
> being as simple as a STORE/RESTORE, only this time to HPUX. It would be
> kind of like the OLD HP, remember when HP customers had no reason to do
> business with another computer mfg.
> I know some software engineers will say that porting the above list of
> software is not possible. However, when it comes to programming a
> computer, I know nothing is impossible, all the above software can be
> ported to other platforms, starting with the creation of a PA-RISC
> Compatibility mode for the IA-64 & IA-32 Instruction set, and maybe even
> SPL. Any of 'These Companies' using CISC Compatibility mode will be able
> to justify re-compiling into PA-RISC. But how is a company going to
> justify re-writing twenty years of development, not to mention the fact
> that some companies in this spot may not be able to survive such a
> daunting task? Such a daunting task that even HP refers its customers to
> other software companies. Why won't HP allow PA-RISC compatibility mode
> for the next designated instruction set for computing. I think its because
> HP doesn't want to make the investment required to do the right thing.
> Come on HP, no matter what platform you're using, it is still just a
> continuation of one's and zero's. I don't care if it's called MPE, or
> e3000, or NT, or HP-UX. What-ever you call it, the functionality that has
> been designed to full maturity and called MPE/IMAGE/COBOL/VPLUS needs to
> run on the latest designated platform. Call it whatever, let the marketing
> people name it, but MPE and the Products around it have a quality and
> functionality that is far and above anything on the market today. Just
> think, get it, rename it, remarket it, no other platform has products that
> can compete with it, IMAGE/VIEW/COBOL/MPE, no one can touch it. Not only
> will you sell more widgets with the above list of software available on
> them, but you'll also be 'Doing the Right Thing'.
> So what is the NEW HP doing now, why isn't the next migration as simple as
> the last, a restore from tape?
> Why has HP decided to treat its customers with such disregard?
> Will 'These Companies' (Once loyal to HP) continue to invest in HP
> Products?
> Enough said, I'll probably just start rambling on now, but I'd just like
> to say a few more things. In the past, when I first came to the
understanding of what Hewlett Packard 'The Corporation' was all about I was
most
> impressed. HP was about innovation, success, and solutions. They were
> proud of each and every one of the HP products, and most of all, product
> support was second to none. For example, back in the day when an HP SE,
> and/or CE would arrive on-site to help you out of a tight spot, you knew
> that they were the experts. You just knew that they would make things
> right, and do the right thing, no matter the cost, because it was about
> something they used to call pride and honor. I remember one particular
> Saturday, years ago, when we were converting from Image to Turbo-Image. HP
> had it all planned out for us, all the utilities, detailed instructions.
> The conversion was a piece-a-cake, but the HP SE showed up that Saturday
> anyway, because he (Bob Barker) knew what customer service was all about.
> He didn't need to be there, HP's written and programmatic instructions
> were more than enough for us. The SE was there for support, support (not
> to mention the lunch we bought him), on a Saturday, free of charge, part
> of the HP support package. Most everyone on this list should remember the
> conversion from CISC (16 bit MPE/V) to PA-RISC (32 bit MPE/XL and MPE/Ix).
> Another great example of HP making sure all Hewlett Packard customer
> investments were preserved. For example: Take this HP-e3000/969, MPE/iX
> 6.5 that I know of, running all the applications for an entire
manufacturing company, several applications that have not been recompile
since being
> developed on MPE/V using a HP-3000 Series 70 in the 1980's. They run in
> CISC Compatibility mode on the PA-RISC platform.  The OLD HP would never
> assume its customers would just simply re-write everything to stay on
> board with HP. This type of thinking that HP has today is nothing less
> than stupidity, only looking at the dollars. Sucking too much money from
> the bottom up, will only cause the foundation the crumble.
>
> --
> Michael Anderson
> Spring Independent School District
> 16717 Ella Boulevard
> Houston, Texas 77090-4299
> office: 281.586.1105
> fax: 281.586.1187
> -
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>

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