« Overview compares emulation strategies | Main | Fine-Tune Friday: One 3000 and Two Factors »

February 28, 2018

An Opening for MPE's Closing Numbers

OpenMPEHeaderTen years ago this month OpenMPE was holding one of its last contested board elections. Spring meant new questions and sometimes new people asking them. The advocacy group launched in the wake of HP's 3000 pullout notice never had a staff beyond its directors. Once in awhile a 3000 expert like Martin Gorfinkel would be contracted for a project, but most of the time the six to nine volunteers from the 3000 community debated with HP on every request, desire and crazy dream from a customer base stunned into transition.

Directors operated on two-year terms, and those elected in 2008 were the last to serve theirs in the era when HP operated an MPE/iX lab. While the lights were still on at HP, there was some hope homesteaders might receive more help. OpenMPE was in charge of asking, but could report nothing without HP's okay.

OpenMPE never pulled from a deep base of volunteers. Most elections featured the likes of nine people running for seven seats, so a long list of losing candidates was never part of the results. The ballots came through from members of the group via emails. Here at the NewsWire I counted them and had them verified by a board committee. In the biggest of elections we saw 111 votes cast.

OpenMPE Links
OpenMPE's 2009 patch proposal

The 2008 election revolved around source code licensing for MPE/iX. The vendor would report the last of its end-game strategies by November, and the end of HP's 3000 lab activity was to follow soon after. Support with fixes to security problems and patches would be over at the end of 2008, and that made the election important as well. There was no promise yet of when the source code might be released, to who, or under what conditions. It looked like HP was going to shut down the MPE/iX lab with beta test patches still not moved to general release.

Getting candidate positions on the record was my role. I dreamed up the kinds of questions I wanted HP to answer for me. HP only promised that by 2010 customers would know whether source would be available and to who. A question to candidates was, "Is it acceptable for the vendor to wait until the start of 2010, as  it plans to do now?"

The question was more confrontational than the volunteers could ask. HP controlled the terms of the discussion as well as the content. Hewlett-Packard held conference calls with volunteers for the first five years of the group's existence, but the calls stopped in 2008. 

Fewer than 30 people ever served on an OpenMPE board. The list was impressive, but keeping good talent like consultant and columnist John Burke or Pivital Solutions' Steve Suraci was tough. Directors like that always wanted more out of HP than the vendor would provide: more transparency, more resources. Of the final three to volunteer, Keith Wadsworth was the only one ever to ask the board to consider if OpenMPE should continue to exist. His election in 2010 gave him the platform to demand an answer. During the next year the directors dwindled to three. 

The group was finally granted a license for MPE source, but it had little else as an asset. There was a $10,000 fee due to HP in 2010 for that license which the group couldn't pay. It also didn't have developers who'd polish it toward any productive use among 3000 customers. That source code was like metal stock without a drill press to shape it, or even an operator.

Staffing resource was always an issue. The group did its best work when it pointed out the holes HP had left behind in its migration strategy for customers. A license transfer process would've never surfaced without OpenMPE's efforts. As of this year, license transfers are the only remaining HP 3000 service a customer can get from HP Enterprise.

Springtime in the 3000's Transition Era replaced the summers of Interex advocacy. During the spring hard questions would be asked in public of people whose mission was getting answers and making promises. HP always gave a nod of thanks to OpenMPE but never portrayed the group as the keystone to the homesteading experience. After awhile the group began to hope HP would find a gracious way to contribute source code to the community.

A limited number of licensees was all the community would ever get. As he ran for his board seat in 2008, Tracy Johnson said that it was fruitless to tell HP a 2010 deadline for source news was unacceptable. Even while customers continued to drift away, HP held all the cards, he said.

It is apparent HP cares not one wit whether OpenMPE declares any decision "acceptable" or not, and making such declarations isn't going to gain any friends at HP.  We're more like a public TV station that needs a telethon every once in a while to keep us going.  But there's only one donor with the currency (MPE) to make it worthwhile, and that is HP. The one accomplishment that OpenMPE needs to put under its belt is to get HP to work with us, and not be at odds with each other.  Everything else hinges on this.

Third parties were supposed to negotiate their own separate contracts for support tool licensing. Testing the beta patches thoroughly using OpenMPE volunteers was proposed and Johnson signed off on it. The volunteer group couldn't line up testing resources, which didn't matter because HP didn't release the beta patches for tests beyond HP's support customers. Nobody even knew for sure if patches were something customers desired.

"Addressing the question of testing," Wadsworth said 10 years ago this spring, "although the OpenMPE board members and  members at large command considerable expertise, it does not seem apparent that OpenMPE as a whole has the ability, let alone the infrastructure, to  conduct such testing."

Higher-order proposals—like un-crippling the final generation of 3000 boxes—went unpursued, too. It seemed like HP was shutting down this product line. Why give customers the rights to full use of their servers at shutdown? "This type of change  would not only add new breath to the e3000, it would add new life to a platform that is being shut down," Wadsworth said in answering a candidate question. "So because of the unlikelihood of this happening I do not think it is a direction that OpenMPE should concentrate resources on at this time."

Source was released and licensed in the spring of 2010, and the group did the advocacy for the transfer of licenses, and to include an emulator clause in that license transfer. By the end of the effort, funding came from loans and contributions from the board members." The group's chairman Jack Connor, the next to last leader, said "the contributions showed their commitment to the OpenMPE concept."

That concept changed constantly over the almost nine years of the group's existence. It began with the desire to get HP to open up source code and technology about a server it was discontinuing. Then the mission was the creation of a 3000 emulator. MPE/iX licensing issues, as well as necessary but overlooked HP procedures for migration, because the longest-term mission.

HP's Dave Wilde, the penultimate HP 3000 business manager, said that OpenMPE was an important part of HP’s planning for a post-2008 ecosystem for the platform. Springtime elections offered hope that the ecosystem would flourish. 

 

 

08:19 PM in History, Homesteading | Permalink

Bookmark and Share

Use our search engine to find 20 years
of HP 3000 news and articles

Comments

Comments

The comments to this entry are closed.