3000 NewsWire: Recent Episodes
Ron Seybold
Reprinted from December 2002
Slim majority of customers choose to homestead or study
First of two parts
One year after HP announced its plans to leave the HP 3000, a little more than half of its customers are either still studying their response to the news or choosing to homestead on the platform, according to a 3000 NewsWire poll.
The survey, conducted on the one-year anniversary of HP’s November 2001 announcement that it’s exiting the 3000 marketplace by 2007, showed one-third of companies responding have already chosen to homestead some or all of their 3000 operations beyond the December, 2006 HP end of support date. Nearly one in five companies are still studying their options in response to HP’s 2001 announcement.
The poll was conducted via e-mail messages broadcast to 500 HP 3000 customers selected randomly on November 13. Companies responding totaled 116, for a 23 percent response rate. Customers identified their firms by name, and no anonymous, Web-based replies were used in the poll’s results.
The margin was close between those sites staying with the platform and still studying options, versus companies choosing to leave the 3000 at some point over the next four years. While 52 percent of firms aren’t deciding to leave this year, 48 percent of companies responding to the NewsWire poll report they have begun plans to rewrite programs, replace applications, or follow their packaged app providers onto other platforms.
Homesteaders — those customers sticking with the 3000 beyond 2006 — came in at 34 percent of companies responding to the poll. Of the companies reporting they are planning to leave the 3000 through migration or replacement, 23 percent are following their packaged application vendors. Amisys healthcare customers led the list of those planning to make a move away from the 3000, followed by educational organizations using various applications.
Companies responded over the two-week period following the Nov. 14 anniversary of the HP announcement. Even among those choosing to leave the platform, sentiment about the move ran to regret and disappointment. A few responding firms had already committed to leaving their 3000s before HP’s advice was announced last year.
The poll’s results showed a stark contrast to the HP claims of April, when the vendor said that more than 80 percent of customers would be leaving the platform. HP has recently begun to recognize that a significant part of its customer base cannot justify the expense of migrating from the HP 3000, even in the face of an end to HP’s support. (See our Q&A in this issue with HP's Dave Wilde for statements on HP’s plans to accommodate its homesteading customers.)
Undecided customers on the fence about their plans reported timelines to decide ranging from weeks to years. John Pickering, a consultant serving a North American firm manufacturing wood products, said a recent migration from IBM SAP mainframes to a PowerHouse application on a 3000 has left his client with little budget or time to do anything about its 3000 during 2003.
“We have no real need to do anything yet,” Pickering said, “as we’ve still got several years and [the 3000] is currently meeting our needs just fine.”
Others still looking over their options say HP won’t be winning any new business if they decide to leave the 3000. “If we do migrate it will quite possibly be to a totally non-HP platform,” said Frank Nikoletti of Argyle Diamonds in Australia. “I think that this is where HP got it horribly wrong, because they expected to retain many customers by moving them to other HP platforms — and I don’t think that is what customers will end up doing.”
Nikoletti and others looking at migration are aware that new options are surfacing steadily, however. The company’s diamond sales and sorting application “is our competitive advantage over all other diamond producers, and there is no other comparable product,” he explained, making a sound business transition option difficult. New migration suppliers offer some hope.
“Just in the last three months there have been a number of offerings in the migration market,” he said. “It will be interesting to see who is still there in a year’s time, and also what other new ones arise.”
IBM is getting attention from both companies electing to migrate and those still considering a plan. One such poll respondent said HP’s announcement disrupted IT operations that were running smoothly.
“We may migrate to the IBM iSeries,” said Bob Bonnaci of Leader Services, a school district IT service supplier. “For what it’s worth, we are very disappointed in HP’s decision to do this, and had absolutely no plans to migrate from the platform prior to the announcement.”
Others still deciding have ruled out homesteading as an option. At the Maryland Higher Education Commission, Charles Benil said the organization will follow its application provider off the 3000 as well as rewrite in-house MPE surround apps to migrate its programs.
A stalled economy has many companies predicting a long timeline for making a decision or moving away from the 3000. “We will follow our application provider to an NT environment when it’s cost-justifiable,” said Debra Gauger of the City of Oskosh, Wisc., which is still studying a plan for its internal apps.
Few respondents asked for anonymity with their replies. But one who did said his health organization has already been disappointed by the capabilities of a replacement package.
“We had decided to migrate the business on the HP 3000, as well as on our IBM mainframe systems, to a new vendor purchased product,” said the IT manager. “This plan appears to be in serious trouble, due to system capabilities of that vendor’s product.."
Reprinted from December 2002
Dave Wilde is leading HP through fresh territory in its final year of selling HP 3000s. The business manager for HP’s 3000 operations is overseeing the details of establishing a homesteading practice for customers who choose not to migrate. It’s work that serves as a counterpoint for HP’s assurances that all the pieces are in place for customers who must move their applications onto other HP systems.
After this year’s HP World, customers left Wilde and HP with a host of questions about what else HP can do to assist 3000 owners in staying on the platform. Emotion hadn’t cooled, as HP reported to some publications, but a practical dialogue began between the vendor and its customers about how homesteaders will manage in a world without HP’s resources.
One persistent question began to surface, too: why HP wants to remain a force in a computer community where it will cease sales in 11 months, and end support in four years. It was just one of several queries we wanted to follow up on after the conference, during the months when HP is planning its next set of homesteading announcements. We spoke with Wilde in the weeks after HP World by phone, where he explained HP’s objectives in the 3000 homesteading market and what he’s willing to consider.
Why does HP want to force its customers to use HP hardware while running new MPE licenses on Intel-based systems? Why can’t HP let go of MPE?
The decisions are guided by a couple of factors. One factor is that we clearly want to retain a relationship with customers, to keep them as satisfied HP customers. Clearly as we move forward, we want to structure things in ways that are good for our customers, and continuing to play a part in our customers’ long-term satisfaction is important.
I recognize some of the concerns that people might have — on one hand, walking away from the platform, coupled with this may feel a little bit odd to people. But there’s a fundamental philosophy at play here. It’s important to HP that we continue to play a role in understanding the customers’ needs and in addressing those.
As an example, this licensing question that came up about the emulators: we feel strongly that going forward we’re going to have a market-leading offering in the server space. We would like customers who want to move forward with MPE to do that with HP hardware.
We said HP hardware. In the shorter run that could be HP PA-RISC hardware. It’s not only Intel-based solutions.
What’s going to make the HP solution a better deal for the customers, as well as better for HP’s sake?
We’re going to have a market-leading offering. Where customers are staying with HP-based solutions, it’s in the interest of the customers to have access to market-leading hardware.
Does HP still want these customers?
Yes. For customers who have a need to run things beyond HP’s end of support date, we’re trying to address things so they’re retained as HP customers. The things we announced which HP would try to enable after HP’s end of support date are really targeted at those customers who have more stable environments and smaller environments, and it’s more practical for them to run things beyond HP’s end of support date. It’s not our belief that’s going to be beneficial to or attractive to customers with large organizations, large applications and dynamic environments.
I understand HP’s beliefs haven’t changed. But is it true that you recognize some portion of your customer base won’t be moving away from the system?
Some of our customers have told us that they need more time to plan their transitions, and others have told us they do not intend to transition. They want HP to work on solutions, and work on solutions with our partners, to make that sort of environment more available to them.
Are you willing to recognize there’s a subset of your customer base that can’t afford to do this transition?
Yes. Their drivers are that it’s may not financially the right move for them, either because they can’t afford that transition or they’re deciding it’s not worth it for them to make the investment [in a transition]. We’re trying to have a set of solutions for them to help them address that, without going so far as to say we think that’s the right answer for the majority of the customers. We recognize there’s a significant segment, and we’re trying to be responsive to that.
Nobody has heard from a large enough segment of the customers to have a number that's representative, right?
Yes. I’ve heard different numbers, and I’d rather not get into guessing what the size of that segment is. I’d believe that it’s far from the majority, but it’s a very significant segment.
How soon can HP have a proposed MPE license fee in place, so the companies which are considering building a hardware emulator can gauge their market?
One of the things we said at HP World was that I prefer not to trickle information out one-off. It has a tendency to confuse the message and also tends to be inefficient in terms of getting substantive announcements to our customers. We were able to announce a solid bundling of new information, as well as consolidating that with existing information, and I like that model much better.
The model that seemed to resonate with the folks at HP World was roughly quarterly announcements, while providing updates when available. That would indicate around the end of this calendar year.
I will state, and I think this is important, that with this licensing HP is not looking towards structuring the MPE license to make the emulator financially unattractive to customers. We don’t want to state anything about particulars until we’ve thought the costs through and what some of the customer needs are.
Gavin Scott of Allegro said he thought it would be a lot more attractive to any company offering a hardware emulator if they could ship off a demonstration version of the emulator with enough MPE for a test. This involves someone shipping off MPE other than HP. Is that prospect possible?
I’m very open to different alternatives and approaches within the boundaries we announced at HP World.
The objective is to help make this solution work, right?
Yes. Another objective is to make sure that expectations are clear, so we don’t oversell and overset expectations for customers.
Are you willing to get into a position where HP’s just collecting royalties for MPE?
I’m open to lots of alternatives. There are many issues from a customer and partner perspective: legal issues and constraints in the future that we want to make sure don’t result in an untenable position for anybody. That’s one of the reasons we want to be, not vague, but non-specific at this point. We’re in a gray area we haven’t been in before.
What’s HP’s position on third parties taking over portions of the delivery chain beyond the end of sales date next Oct. 31?
We’re working on partners with different elements of the value chain. That’s something we’re definitely doing in the transition space. Another area is remarketed systems, and we have a very strong relationship with Client Systems’ subsidiary Phoenix Systems. We’re definitely interested in working with them. We’re open to documenting what we’re doing so customers' and partners’ interests are also protected, along with HP’s interest.
Documenting what?
Suppose the emulator runs on HP solutions. We want to make sure partners aren’t left in a bad position in case HP makes future decisions that are inconsistent with that announcement. We’re very open to different arrangements, subject to representing reasonable business needs from HP’s perspective.
The point is that I’m very open to needs not just from HP’s perspective, but also from a customer and partner perspective. We’ll be working with the OpenMPE group and with partners and customers who are expressing concerns.
What’s the thinking behind maintaining the current pricing model for current A-Class performance?
Doing things like releasing all the horsepower that’s available in one chassis has a ripple effect that’s not acceptable for the pricing model. But we’ve heard that customers who have bought into an A-Class chassis may have some needs for more upgradability. One of the questions we asked was if there was a price associated with an additional upgrade within an A-Class chassis, would they be interested? That’s a price consistent with our overall pricing model. There seemed to be some interest in that which we’ll continue to look into.
Is this a way for a customer to avoid having to find an N-Class system to upgrade to after HP’s sales of N-Class systems stop?
Yes. If there’s enough interest and demand, and we can structure it in a way that it fits in well with our established pricing model, it’s something I’d be open to trying to structure.
Why so much attraction to a pricing model that was conceived before HP decided to leave the 3000 market?
It just isn’t practical for us to restructure the pricing model. We feel we structured it in a way that there’s appropriate value in the system for what we’re delivering. It’s a reference to purchases that have already been made.
You recently made a healthy reduction in that cost to customers with the new pricing. Why not continue in that with the A-Class?
Within limits we’ve obviously tried to offer strong value while recognizing the customers’ expenses they’ll be incurring through these transitions.
Is simplifying HP’s support through such an upgrade — which would give the 6.5 customers something to upgrade to in December 2004 — a factor in creating an A-Class upgrade?
There might be some impact there. But there are enough customers in different segments that some customers incrementally going to upgrade probably doesn’t fundamentally change the support picture from HP’s perspective. What’s more meaningful to me is if it gives customers more degrees of freedom.
How long do you think CSY will continue to operate as a virtual entity with HP after 2003?
CSY is the value chain that delivers 3000s to our customers. In that respect, it exists though 2006. How it’s managed internally is a subject that nobody can predict. From a customer perspective, for me the goal is that customers would see a 3000 business orientation through HP’s end of support date.
What’s the contact for that virtual CSY beyond end of sales? It’s looking like the reseller channel will be folding its tents next year.
From a channel perspective, what’s the contact point? We’re working closely with Client Systems in the Americas to identify different ways they can support customers. We’ll be continuing to work with all our partners. We’ll need to stay tuned as things move forward.
From an HP perspective, I’m the business manager, and I’m the go-to guy. We’d like to make sure the virtual CSY value chain is also represented cleanly and consistently to our customers. Exactly how that’s implemented over time may change. It works best where our customers would be able to work with HP in a centralized fashion.
Having said that, I continue to feel like I have an important role in adding value.
What about you? How long do you want to be the 3000 go-to guy?
I basically want to do this as long as two things are true. One is that I feel like I’m continuing to add value, and that HP wants me to continue doing that. Whether that’s months or years, nobody can predict. There’s no timeline for the end of that right now.
I didn’t expect a few years ago to be in the business, let alone in this particular role. Things have worked out in a way that I’m pleased to be in this role, and feeling like I have an opportunity to add value. It’s hard to tell these days what the future holds. Things are changing very quickly, both inside and outside the company.
It was something of a gamble 50 years ago, but Hewlett-Packard rolled out its first HP 3000 servers this month in 1972. "November is a Happening" banners — probably printed out on mainframe greenbar paper — hung on the walls and cubicle dividers of the factory in Cupertino. This was an HP that still put out doughnuts for its engineers (exclusively male) and hosted beer busts on Friday afternoons.
One of the best sources for stories of this era of the 3000 is Bill Foster's TeamFoster website. He's got more than 15,000 words of reporting and commentary about the HP of 1969-1976. The 3000 became known as Omega inside the labs, a more advanced design than the Alpha model preferred by Bill Hewlett.
They had two new computers under development, code-named Alpha and Omega. When completed, Alpha was going to replace HP’s existing minicomputer line, the HP 2116. But Alpha was just Omega’s little sister, and Omega was going to knock the socks off the industry. Everything about Omega was new and state-of-the-art — the iPad of the day. So naturally all of the engineers wanted to work on it -- nobody wanted to touch Alpha.
Omega was just too ambitious for the hardware of the era. The operating system of 50 years ago was the unique flavor that HP added to the minicomputer mix of the seventies. HP canceled the Omega operating system and fell back on Alpha. The computer system was troubled from the start, bad enough that HP recalled those servers it first shipped. As many of them as they could get back, anyway. HP offered the customers a 2116 in exchange, and at no cost.
Foster says the savior of the System 3000, as HP called it in the early days, was Mike Green.
MPE was the most complex part of the computer and it was a disaster. Because of MPE, customers began shipping their 3000’s BACK to HP -- that was definitely the wrong direction.
Mike agreed to save MPE, and after a week or two we were ready to present his plan to Paul Ely. Mike stood up in a room full of important people and gave the pitch. It was a great plan, and Mike said we would be out of the woods in about five months. When he finished his presentation, Ely said, “Are you telling me five months because that’s what I want to hear, or is this really what you think will happen?” Mike looked at Paul in a dismissive manner. “I’m saying this because it’s going to happen. Why would I say anything just to please you?” For once Ely was speechless. There was dead silence as we left the room. And five months later MPE was working.
Foster's reporting is long enough to be a third of a nonfiction book. It's only available on his website, though. With free beer, and nothing but men behind terminals building an OS, the tale might as well be from another planet. Successors of that hardware and software are running today. Gavin Scott, who noted the 50th birthday, tends to an emulator of the original design. It's a turn-key setup "which will let you have your own 1980-vintage HP 3000 system up and running in a couple minutes."
Fifty years is close to a lifetime for human beings. It’s a span in the computer industry that feels like aeons. All along, it’s been MPE to carry that seventies technology into the third decade of the twenty-first.
Considering how an emulator can elevate an elder technology into orbit, MPE might live forever.
From the 1980s onward, one software vendor was able to go toe-to-toe with Hewlett-Packard. A recent verdict from a US federal court shows the vendor used a win at all costs approach: cheating.
The Lund Performance Solutions still include a defragmentation product, Defrag/iX, for HP 3000 servers. That software is now sold by its creators, Allegro Consultants. Defragging a 3000 disk set becomes important when B-tree links clutter. But measuring and adjusting 3000 systems performance was a higher-stakes game starting from the 1980s. Now it looks like the stakes were high for Robert Lund, the founder of the vendor who faces a sentence for nearly 30 years of tax fraud.
The news convicting a 3000 vendor of a federal crime is not unique. During the years of the early Noughts, executives of HardwareHouse and Computech earned jail convictions for illegal transfer of HP 3000 hardware licenses. The cause was so celeb that another 3000 vendor, Phoenix 3000, tried to shoo away business that was going to the illegally licensed systems and their brokers.
The story of the tax fraud reads like an epic out of the excesses of the Eighties. A private landing strip, a 7,000-square-foot home, a trailer park with rentals, a health food store, and Medicaid fraud are among the details in a US Attorney’s press release. “On his food stamp and Medicaid applications, Lund boldly claimed to be a part-time handyman earning just $810 a month. In total, Lund stole approximately $70,000 in public benefits, most of which were paid by the federal government.”
The maximum sentence Lund faces is a 27-year prison term and a $950,000 fine. He’s making restitution payments of $1.7 million to the IRS.
Lock-in brings profiting
Some legacy systems owners might remember heady days when such software seemed to come at soaring prices. Systems were captive to vendor hardware like the 3000 and the rest of the HP Enterprise server line. Excesses are accepted, begrudgingly. Once the hardware gets moved to a non-essential track, though, the pricing from commodity markets starts to settle in. There is still lock in with legacy. It starts and ends with the operating system, though.
Lund was a powerful resource for hardware resellers. Proving a system was maxed out meant replacement systems, or at least memory upgrades. Meta-View from Lund measures 3000 system horsepower. HP has its own entry in that performance derby, Perfview. Many managers were skeptical about any performance reporting software from a hardware vendor selling new systems.
Lund’s clients included large businesses, school districts, and health care companies located throughout the US. After an indictment charged Lund with tax evasion, failure to file personal income tax returns, obstructing or impeding the IRS, and theft of government funds, Lund settled in a plea agreement. He will hear his sentence on Oct. 21.
Legacy was big business compared to Intel-based servers during the 1980s. That was the era when many legacy owners took their systems online. Then the x86 architecture gained its speed as well as commodity advantages. There remain many legacy programs that deserve continued investment from owners. The need for special performance measurement isn’t among those, now that Intel servers can work as if they were HP legacy hardware systems.
A commencement address to those ready to matriculate to faster HP 3000 systems By Mike Hornsby Speed costs money. How fast do you want to go? You want to get all of the speed out of your current HP 3000...
HP’s add-on product provides vital disaster recovery, but you’ll need advice on set-up, disk errors and split-volumes By Andreas Schmidt Mirrored Disk/iX is an optional subsystem for HP 3000 mission critical systems, and it’s vital to guaranteeing high availability of...
HP 3000 hardware networking can be like a trip down a Yellow Brick Road By Curtis Larsen Auntie MAU! Auntie MAU! A Twisted Pair! A Twisted Pair! Once upon a time, networks were as flat as the Kansas prairie, and...
Hewlett-Packard Enterprise has issued dates to terminate support for two releases of its HP-UX Unix environment. Next year will mark the end of HPE’s support for HP-UX 11.11 and 11.12. The final, terminal version of HP-UX, 11.31, is already in...
Call-compatible subroutines, utilities help Unix behave like the MPE/iX environment By Charles Finley Let's look at methods to migrate applications from HP 3000s. One tool is to making subroutines and routines call-compatible, to let existing business logic work on other...
By Neil Armstrong Our development environment at Robelle was quite unusual, in that we had a single job stream, which launched all of the necessary compiling and testing steps for each product. So if I wanted to compile and run...
Included software, tested job offers some disaster recovery By Wirt Atmar We have been using one of our HP 3000 Series 918s to shadow our primary development 918, using the MPE-to-MPE FTP capabilities of MPE. The purpose of doing this...
State-of-the-art security is possible using tokens, agents and RSA secure servers with MPE/iX By Andreas Schmidt This article describes a way to build a security barrier around your HP e3000. It is based on material offered by Security Dynamics (www.rsasecurity.com),...
Back in 2005, November was still a month bleeding in the red ink of memory. You can use shorthand and say "November 2001." Or you can say the day that HP's 3000 music died. The date of November 14, 2001...
He saw nothing, nothing From October, 2001 Just weeks before HP started to brief its vendor partners about the 3000 futures cut-off, customers asked about it. In a public forum of a webinar, the 3000's vendor relations manager, its product...