Ken Sletten's open letter on OpenMPE

Sunday 21 December 2003

To all MPE users, especially Homesteaders:

First all please note this important DISCLAIMER:

What follows is the personal position and opinion of the writer, and does NOT reflect the official position of the Board of Directors of OpenMPE Inc.; i.e.: This is a minority report. While I believe other members of the BOD are significantly disappointed with continued foot-dragging by HP on key issues affecting "life for MPE after HP" (L4MAH), they will have to speak for themselves if they choose to do so, on the extent of their disappointment and on what follows.

In any case, the patience of this member of the BOD has been well and thoroughly exhausted: 2+ years and counting after the "end-of-HP-sales" and "end-of-HP-support" announcement on 11-14, appeals to HP by the BOD and other MPE users for some kind of substantive action on MPE licensing and other key issues remain unanswered. Without catalysts only HP can provide, progress on L4MAH is effectively at all stop. Nobody is going to invest significant funds in a potential post-2006 future for MPE as long as there is no concrete indication let alone guarantee that HP will allow and enable that future to happen in the first place. Given the failure to make substantive progress, and due to the nature and circumstances of other information that has come to our attention, I feel I have an overriding obligation to the entire OpenMPE community to bring everyone up to date on the content of recent official communications between HP vCSY and the BOD.

Also note that while this document is written mostly from the OpenMPE perspective since that's what I've been working on, this is NOT meant to apply just to OpenMPE;
i.e.: If OpenMPE is not the organization that should assume oversight responsibility for the care and feeding of MPE after end of HP support in 2006, I've got no problem with that: It could also be one or more or a consortium of established 3rd-party companies that have the technical experience, sufficient staff, and business track record to be custodian / prime contractor / sub-contract manager for MPE after 2006; and that have or can get enough financial resources to sufficiently maintain that position. A list of potential candidates for that task besides OpenMPE Inc is pretty short, and should be fairly easy to generate.

But whoever it is, if MPE is going to avoid being frozen in time at the end of 2006, SOMEBODY needs to be granted formal MPE source code access / patch release / licensing authority; and action by HP to make that happen needs to move forward NOW or it will be too late. HP should have had at least an outline for much of this laid out BEFORE they made their infamous 11-14 announcement; i.e.: They are already a couple years behind the curve. Remember that the overall process of just managing and executing any MPE build / release cycle is a complex task, let alone implementing and testing even minor updates and bug fixes to core MPE routines and subsystems. The in-depth collective knowledge required to reliably accomplish those tasks cannot successfully be carried forward if HP waits until just before they turn out the lights to set MPE free.

I hung on for what is now the 7 weeks after e3000 end-of-HP-sales on 31 October, to see if at long last HP would FINALLY be ready and willing to take some concrete positive action that would immediately be perceived as such by all. The fact that the answer is still no is one of the key reasons that drove me to publish this document as a minority report.

HP will likely maintain they haven't said no, but rather that they're still working on it; i.e.: "Later" (maybe). Crux of the problem is that time is of the essence, and later doesn't cut it anymore. After many months of trying to work constructively with HP and persuade them to get off the dime; after 7 weeks of intensive discussions by the BOD since end-of-HP-sales; and after presentation to HP of a proposed Agreement in Principle (AIP) authored by the BOD that seems completely reasonable and appropriate if HP truly wanted to be a "trusted partner": The only result to date is that HP continues to stall and run out the clock (expect many are shocked; SHOCKED; to hear HP would do such a thing).

Before making further comments, let me cut-and-paste a formal communication from HP to the BOD at the end of October 2003. This email was sent to Jon Backus on behalf of HP by Mike Paivinen Friday evening 31 October, in response to concerns expressed earlier by the BOD to HP.
All members of the BOD got a copy on Sunday 2 November 2003. Note this HP email is bounded by a line of "++++++++" at the start, and by "########" at the end.
Here is the verbatim text of that communication:

++++++++++++++++++++++++++++++++++++++++++

From: PAIVINEN,MICHAEL (HP-Cupertino,ex1) [mailto:michael.paivinen@hp.com]
Sent: Friday, October 31, 2003 8:01 PM
To: 'Jon Backus'
Cc: WILDE,DAVE (HP-Cupertino,ex1); MCDONALD,ROSS (HP-Cupertino,ex1);
VANCE,JEFF (HP-Cupertino,ex1); PAIVINEN,MICHAEL (HP-Cupertino,ex1)
Subject: vCSY Response to the OpenMPE Board

Dear OpenMPE Board of Directors:

Thank you very much for taking the time last week to share your concerns with us regarding the future of OpenMPE and the need for a timely HP response to the critical needs of the e3000 user community.

I have discussed your concerns with other members of the vCSY leadership team (Ross, Jeff, Dave), specifically your request for a publicly communicated timeline covering decisions on long-term community access to hardware (e.g. 9000->3000 conversions), software (e.g. MPE/iX source access), and the other remaining issues that address e3000 community needs after HP's end-of-support date. As Dave Wilde said, we don't want to make decisions on a one-off basis, and we want a chance to assess any changes in community needs after 10/31/03. We also think it's important that we have a general idea of what we are going to be announcing - and have internal alignment around those ideas -- before we establish a timeline for that communication. We don't want to prematurely commit ourselves to a path that limits our ability to meet the varied, and sometimes conflicting, needs of our customers and partners. As a result, we expect to be able to provide a communication timeline to the OpenMPE Board by January 31, 2004.

We continue to believe that OpenMPE plays an important role in advocating for the needs of the segment of the customer community that may require the continued, productive use of their e3000 beyond HP end-of-support. Looking back, in the first year after our November 2001 announcement, OpenMPE played a critical role in both advocating the needs of its user community and working with us in formulating our response. We don't believe our response to our customers at HP World 2002 would have been as complete or responsive without OpenMPE's involvement in the process.

We want to work with you to maintain the viability and credibility of OpenMPE. We acknowledge that this timeline will create a more challenging path towards that goal. So, in my meeting with Jon next week, we can begin to discuss the available options.

Finally, I want to reiterate our commitment to addressing the needs of the OpenMPE community. Although we may not agree on the timeline for doing so, we agree that there are real issues that need to be addressed.

Sincerely,

Mike Paivinen
##########################################


Do the above soothing but completely "commitment-free" words give you a good warm fuzzy, that HP will finally and expeditiously do what they should already have done long ago to give MPE a shot at a viable post-2006 future before the fat lady sings ??...
I think not.

And be sure you catch the full meaning of the one above item of any substance with a date attached to it: "By January 31, 2004" is still NOT when HP proposes to finally tell Homesteaders what actions they will take and commitments they will make to facilitate L4MAH: That is only when they "expect to be able to provide" a SCHEDULE for when AT SOME SUBSEQUENT AND STILL UNKNOWN DATE IN THE FUTURE, HP will tell MPE users what they might (or might not) be willing to do. 2+ years and counting after 11-14, this is in my opinion a fundamentally unacceptable situation.


Not only that: Let the record show that when the BOD told HP we wanted to make the above formal but essentially "content free" communication to the BOD public, HP's response was that they considered the above email to be covered under the informal NDA between the BOD and vCSY, and they did NOT agree that it could be made public.

On balance I find HP's claim that NDA coverage of the above message should prevail neither persuasive nor compelling, for the following reasons:

(1) Consider an approximate analogy: In working with another party you become aware that they are taking action (or NON-action, in this case) that will have a very detrimental effect on a business that you and many others that depend on you have a big stake in. When asked directly about this the other party in effect confirms that, yes: They ARE doing what you think they are doing (running out the clock). BUT: The other party says that you are not allowed to tell your partners, because they have decided their actions are covered by an NDA cloak.
And they made that decision because..... well (apparently), just because... My answer: The obligation of those in positions of responsibility not to conceal material facts from their constituents and to warn the victims that their vital interests are at risk easily trump what was AT BEST a fuzzy, never-formally-defined, unevenly specified and applied NDA. It's mostly a sidebar, but I note in passing that on several occasions in the past two years HP singled out specific items for NDA coverage before getting in to them during discussions with the BOD; they did NOT do so ahead of time for the above email.

(2) The above email presents a strategic message to all MPE users about HP actions to facilitate L4MAH: Other than the limited and heavily qualified emulator statement a long time ago, nothing of substance has happened, and as far as HP is concerned it's going to stay that way for what is still an indeterminate length of time. The OpenMPE community has a right to know that, and I see it as my primary obligation to pass on that important information.

(3) As far as the detailed contents of the above HP message, note that besides reflecting our concerns back at us all it contains is:
[a] A bunch of mushy, feel-good "you're important"
and "we agree that there are real issues" phrases.
[b] Mention of 2 things (9000 -> 3000 and source access)
that have been discussed at great length in public.
[c] Repetition of the "we don't want to make hasty
decisions" litany HP has probably put out in one form
or another at every public meeting since 11-14.
[d] "....we expect to be able to provide a communication
timeline to the OpenMPE Board by January 31, 2004."

THAT's IT !!...: Except for the "communication timeline" sentence, it's either hard-content-free or has been said in various ways in public forums to the point of boring repetition; i.e.: TOTALLY devoid of even a hint of corporate secrets or significant new information. And yet HP *still* did not want the public to see it; and was not willing to agree to so simple a request from the BOD that had no discernable cost to them, except for having users hear what their current plan is for when HP proposes to provide a SCHEDULE for when they are going to tell us something some time later on. Given that unwillingness to cooperate on something so trivial, how in the WORLD can anyone have any confidence that HP will ever be willing to take any SUBSTANTIVE steps to facilitate L4MAH, when those decisions WILL have significant implications both inside and outside HP and for a large number of MPE users;
unless and until those actions are actually seen to happen by all ??

(4) If HP is willing to use release of something as totally bland, mushy, and innocuous as the above email to the BOD as an excuse for stiffing MPE Homesteaders, then I believe it is a foregone conclusion that they're going to look for and find any number of handy excuses to stiff us in the end in any case. Better to know for sure now instead of later.


After getting the above email from HP, the BOD went in to an active drafting phase, where there was general participation in what I think ended up being 9 or 10 revisions of a proposed Agreement in Principle (AIP) between HP and OpenMPE Inc. As per above this was written from the OpenMPE perspective, but as far as I'm concerned it would be O.K. to substitute the names of one or more established 3rd-party MPE businesses, if for whatever reason OpenMPE is not the organization to carry this forward. Note that several members of the BOD including the writer were hampered during this period by being unusually busy with "real work" and etc. (being on the BOD as a volunteer is not paying anyone's light bill); and it was not until just after the US Thanksgiving Holiday that the proposed AIP went to HP. Here is that communication from the BOD to HP:

++++++++++++++++++++++++++++++++++++++++++

From: Birket Foster [mailto:birket@mbfoster.com]

Sent: Tuesday, December 02, 2003 9:58 AM

To: michael.paivinen@hp.com
Cc: OPENMPE-BOARD@RAVEN.UTC.EDU; WILDE,DAVE (HP-Cupertino,ex1)
Subject: Open MPE

Mike,

I am writing to you on behalf of the Open-MPE group to make some changes happen - in order for Open MPE to be able to fund the vLAB, that we have been talking about for many months now, we must be in a position to have enough concrete information to allow a "homesteading company" to be offered something of value downstream.

To that end it is important that, as soon as possible, we have an agreement in principle around OPEN-MPE's role going forward and a public statement from vCSY that will allow OPEN-MPE to get out of the holding pattern we have been in and start taking action to ensure that there is continuity in our advocacy and a smooth transition of the MPE Operating environment.

You said in a recent note to Jon Backus

"We continue to believe that OpenMPE plays an important role in advocating for the needs of the segment of the customer community that may require the continued, productive use of their e3000 beyond HP end-of-support. Looking back, in the first year after our November 2001 announcement, OpenMPE played a critical role in both advocating the needs of its user community and working with us in formulating our response. We don't believe our response to our customers at HP World 2002 would have been as complete or responsive without OpenMPE's involvement in the process.

We want to work with you to maintain the viability and credibility of OpenMPE. We acknowledge that this timeline will create a more challenging path towards that goal. So, in my meeting with Jon next week, we can begin to discuss the available options.

Finally, I want to reiterate our commitment to addressing the needs of the OpenMPE community. Although we may not agree on the timeline for doing so, we agree that there are real issues that need to be addressed."

If vCSY really thinks that Open MPE is the solution to a category of these "real issues" then an agreement in principle (AIP) or memorandum of understanding should be a logical next step ... the timing may be different but the principles should resonate. Here is a draft of the AIP that the OPEN MPE BOD thinks will fill both our needs and allow us to begin a series of steps to put OPEN MPE in a position to be able to address the issues.

========== DRAFT AIP ===========================

On December XX, 2003 HP reached an agreement in principle (AIP) with OpenMPE Incorporated, to grant OpenMPE a non-exclusive license to all source code for MPE/iX and related MPE products, tools, software build/test suites and internal documentation. This license will allow OpenMPE to produce, control, manage and distribute bug fixes and enhancements to these products, and thereby facilitate future support of MPE for sites that continue to run it beyond 2006. OpenMPE will in turn sub-license these products to end users and partners, and collect and distribute appropriate fees for doing so.

This MPE license will be granted to OpenMPE only if certain conditions are met. As part of this AIP, HP recognizes the need for an MPE Lab staffed independently of HP, to allow continued viability of the MPE environment after end-of-HP-support in 2006. HP recognizes that funds and lead time will be needed to bring an effective MPE Lab online. Planning and fund raising for this project will begin immediately. It is expected that OpenMPE will be in a position to begin independent distribution of modifications to MPE after the end of 2006. ===============================================

As we both know there is some considerable "ramp" time even for folks familiar with the MPE bedsheets and the concepts around the build of MPE - OPEN MPE needs to fund vLAB to be able begin the transition soon as December 31 2006 is coming and some overlap will be required as part of a "turnover".

The wording of this AIP was carefully chosen to allow vCSY and OPENMPE to be able to get the details fleshed out later - we need the agreement to present to our membership to be able to move forward.

Is it possible to get this wording approved by December 12th?
I look forward to your early reply,

On behalf of OPEN-MPE,

Birket Foster
Vice Chair
OPEN MPE
##########################################


One week after the above was sent from the BOD to HP, we got HP's answer:
The answer was No.

In the interest of precision, note that HP's "no" in this case was in answer to the last above specific question:
"Is it possible to get this wording approved by December 12th?"
Some might point out that this is not the same as saying "never". My response is (again):
If 2+ years after 11-14 HP's answer is still "not now", that's not good enough.

It appears that HP is willing to "work with us" ONLY if what we want already fits in EXACTLY with what their pre-determined internal plans, schedules, and timelines already say. In fact, looking back now on the entire history since the 11-14 announcement, it's hard not to feel like HP has been playing us for suckers. I don't doubt that the HP Support organization and above all HP Legal are slow-as-molasses-in-January impediments, but at this point for practical purposes I don't know that it matters much anymore how "overt" or laze faire HP's track record has been since 11-14; except for being disappointed that HP managers used their long-term working relationship with MPE users to sell the idea that we could depend on them to do "the right thing" before it was too late. What's left of vCSY may be a footnote within the corporate behemoth, but I don't accept that as a valid excuse for continued inaction and endless delay: HP's internal organizational problems and vCSY's lack of control are THEIR problems: We are dealing with HP as a company, and HP is not delivering.

MPE users were loyal HP customers for more than a quarter century. HP claims they are still putting the interests of MPE users first. But on balance HP's actions and omissions both before and after the 11-14 announcement continue to confirm that their statements along those lines are still just PR spin and fluff without real substance. Remember that at HP World 2001 HP was still saying they were going to deliver on Dr. Harry Sterling's prior firm commitment to port MPE to IA-64. When HP pulled the rug out from under MPE only two months later they left many high-end OLTP sites without a reasonable IT path, and forced many sites to accelerate spending of what will in some cases be millions of dollars on urgent efforts to "migrate" applications to other platforms.
Plus third-party developers had to make huge investments to try and preserve what will likely end up being only a fraction of their prior e3000 installed base.

One could of course make the point that all in vCSY are HP employees; that the official "company line" is to migrate MPE users to other HP platforms with all deliberate speed; and that we shouldn't expect HP employees to do anything other than their best to expedite and implement that company objective while helping to sell migration services. The "trusted partner" thing HP loves to tout rings pretty hollow in that case; especially given recent surveys and information from various 3rd-parties that confirm a larger than expected segment of the MPE user base will NOT be off the e3000 by end-of-HP-support.

With respect to producing a comprehensive business plan for organizing and implementing an MPE "virtual Lab" (vLab) like we have talked about: Sure, OpenMPE or other 3rd-party companies with MPE expertise could do more speculative brain-storming and even detailed "planning" on what a vLab might look like and how it might function, IF HP ever moved on the fundamental road blocks. But without something at least as strong as the proposed OpenMPE AIP on the public record from HP, who would sign up to invest in and be associated with what would be a "virtual-squared" v2Lab, or put in all the time needed to professionally work out the implementation details of same for no pay ?? (Actually, for NEGATIVE pay, due to lost opportunity costs).

Certainly important details would have to be addressed before any functioning vLab external to HP could go online, but spending significant time and effort on any such effort without having the slightest commitment from HP on key licensing and source code access issues is putting the cart way before the horse. For HP to want a comprehensive plan of operations before they are willing to make any substantive commitments is in fact another great delaying tactic, since that process could pretty much go on ad infinitum. HP appears to want a PERFECT arrangement of the deck chairs on the Titanic before they will consider making even the slightest alteration in course to avoid the iceberg; i.e.: They're not ready to let EITHER the chicken or the egg come first. And if endless delays let them avoid any substantive commitments long enough, HP can just turn out the lights on MPE at the end of 2006.

I've had to come around to something a long-time, respected member of the MPE community said to me recently:
That person had for some time shared my thinking; i.e.:
A big part of the reason for all the non-action and wrong actions by HP was likely just old-fashioned organizational paralysis, internal conflicts, and who knows what all similar etc. But finally that person had to conclude that a company like HP and its management ranks just could not be THAT incompetent. Which leaves me thinking that short of a Pavlovian response to PR pain that alters their current path, it doesn't look like HP will "get around" to taking substantive steps to release their death-grip on MPE until after it has ceased to matter (as some MPE "old timers" have already said on the lists); if it's not too late already.

Maybe the fat lady has already sung and maybe not. But if nothing else from the professional self-respect perspective my bottom line is this: Given the statements and attitudes expressed in the above email from HP and lack of any other sign of progress, I for one am not willing to continue to give HP cover and keep smiling while they screw MPE users who need to homestead after 2006. Nothing speaks louder about the attitude of HP management towards MPE users, than their dismissive unwillingness to adjust their plans or respond in any positive way to something so trivial as a request to post Mike Paivinen's above email to the OpenMPE list. Why ??... Because they would be embarrassed to have confirmation of continuing delay come out ??... A word comes to mind that summarize HP's current wish to keep MPE users in the dark: Contempt. I can put up with quite a bit, but I don't respond well to contempt.

Note for the record that shortly after 11-14 I said on 3000-L that if HP did NOT take reasonable action in a timely fashion to free MPE, I would boycott HP. Time for HP to "do the right thing" has expired; we are at the very least well into overtime.

Therefore in the spirit of the famous 1990 "Boston Tea Party" 13+ years ago, I'm delivering on what I said I would do. HP should do the same.

Unless and until the current state of affairs changes, I say:

"Free MPE, or Boycott HP"

If putting this out just before the Holidays makes me a grinch, I'll just have to live with the guilt.
Fundamentally I am not willing to let what we the BOD found out at the beginning of November about HP's continued foot dragging hang in silence into a new year.

Finally, I would like to say that nothing in the above is intended to disparage the work done or positions taken by other members of the BOD. For the reasons I stated above, I have a philosophical difference with respect to the applicability and precedence of HP's claim of NDA on the above 31 October HP email. Since receipt of that email the BOD has in fact come up with a number of good ideas on things that could and should still be done to facilitate L4MAH. Those efforts should continue if there is any indication HP might be willing to move off dead-center....


Ken Sletten
SIGImage/SQL Chair
MPE / IMAGE user for 20+ years


FOOTNOTEs:

If someone wants to forward this to 3000-L or other discussion groups, you're more than welcome to do so..

I regret my above is perhaps not as polished or organized as it could have been if I had put in more time to work on it. Straight story is that not only do I have another life, I expect that even if I retired from my full-time day job I'd still be 100 percent busy. So this will have to be "good enough". Happy Holidays.