I just returned from
HPWorld 2002 in Los Angeles. This was, of course, the first HPWorld since Black
Wednesday, November 14, 2001, making it the most important HPWorld for the
future of the HP 3000 since at least the Boston Tea Party meeting over ten
years ago. Why? HP had promised to open the door at least a little on its exit
strategy for the HP 3000, telling us whether there is a place for homesteaders
and more than one choice for what I like to call fence sitters. This was also
probably our last chance as a group to influence HP about what the landscape
and ecosystem will look like in the 2003 to 2006 time period and beyond. I was
there partly as Chairman of SIG MPE, partly as reporter for the 3000 NewWire,
partly to determine whether to stay with HP for our next round of Wintel server
purchases and partly as prospective rabble-rouser, though even I was surprised
at the unexpected turn this took.
This month I am going to
depart from my normal format to report on HPWorld 2002: my perceptions of,
along with some facts, fantasies and oddities. Usually here I’d mention some
wacky threads from HP3000-L as an introduction. This isn’t wacky, but it is
ironic. While Carly Fiorina was giving her keynote address and saying, “…great
organizations are defined not just by their capabilities, but by their
character – by the company they are, and by the company they keep”, HP was
announcing to Wall Street an additional 1,800 layoffs above and beyond the
15,000 pegged as part of the merger integration process. Later in the speech
she said, “I’m proud to say that so far we have hit all of our integration
milestones and targets”. Not exactly wacky, but … Normal programming will
resume next month.
As always, I would like to hear from readers of net.digest and Hidden Value. Even negative comments are welcome. If you think I’m full of it or goofed, or a horse's behind, let me know. If something from these columns helped you, let me know. If you’ve got an idea for something you think I missed, let me know. If you spot something on HP3000-L and would like someone to elaborate on what was discussed, let me know. Are you seeing a pattern here? You can reach me at john@burke-consulting.com.
Speaking
of Carly
I was at HPWorld 1999 when Lew Platt introduced Carly Fiorina as his replacement. I was prepared to not like her because I felt HP was making a mistake bringing in an outsider, and a non-engineer at that. I had to grudgingly admit I was impressed. She strode out on stage with a wireless microphone and spoke passionately, without referring to any notes, about her vision of our technological future. She worked the crowd, roaming the stage making frequent eye contact with the audience. Whether you agreed with her or not, it was quite a bravura performance.
Contrast that with HPWorld 2002, where she hunkered down behind a massive podium reading from a prepared speech, sounding to all the world like she first read it that morning. Scheduled for only a half hour, she droned on for over 50 minutes. It was as if we were watching a different person. Question: which is the real Carly Fiorina?
You
can read the whole speech for yourself, including, “Let me repeat what I said last year: we have
never stranded our customers on legacy technology. We have always provided a
migration path to the future and to new business benefits. That has been our
hallmark in the systems business, and it will continue to be our hallmark in
the systems business.” at http://www.hp.com/hpinfo/execteam/speeches/fiorina/hpworld02.html.
“Disingenuous” is the word that comes to my mind to describe that last quote.
SIG Sysman is dead; long live the new SIG MPE
With the support of
the Chair of SIG Sysman, Donna Garverick and the Chair of SIG MPE, John Burke,
the support of the executive committees of both SIGs, and by vote of those
members of each SIG in attendance, SIG Sysman was merged into SIG MPE. The new
SIG MPE will be co-chaired by John Burke and Donna Garverick. John and Donna
will be picking an executive committee this month. This merger was first
proposed 2 1/2 years ago. Others and I were adamantly opposed to it at the
time, believing that while there was natural overlap in the issues of concern
for the two SIGs, there were also areas where the SIGs had non-intersecting
agendas. Combining at that time, it was felt, would unnecessarily dilute the
influence of the members. November 14, 2001 changed everything and I grudgingly
agreed that merging now was in the best interest of our members. Donna and I
are committed to keeping SIG MPE active as long as there is interest, but
particularly over the next year we want to ensure the SIG does everything
possible to represent MPE user's desires as we wind down to end-of-sales.
As part of the SIG MPE meeting, Jeff Vance reported on the status of the top ten items from the last two SIBs. In response to a question from the floor, CSY promised PowerPatch 2 for MPE/iX 7.0 at the beginning of 2003. SIB 2001:
|
1. Investigate making IMAGE thread safe |
Done. Large project. Will not implement |
|
2. System CI variables |
No. Use the wrappers on jazz as a workaround |
|
3. LDEV 1 disk utilization beyond 4 GB |
Done in 7.5. Not available for earlier releases |
|
4. Sendmail in FOS |
Done in 7.5. Patch for 7.0. Maybe for 6.5. |
|
5. Shutdown command |
Done in 7.5. Patch for 7.0. |
|
6. Samba password encryption |
Expected in late ’02 patch |
|
7. Improved Java runtime |
Done in 7.0. |
|
8. UPS integration |
Done in 7.5. Patch for 7.0. |
|
9. PHP, OpenSSL and NTP as HP freeware |
OpenSSL and NTP on Jazz. PHP on Invent3k. |
|
10a. Perl in FOS |
5.6.1 binaries available on Jazz |
|
10b. Perl in Apache |
Sort of. Webwise in 7.5 FOS |
SIB 2002:
|
1. Enable MPE to survive beyond 2006 |
Working with OpenMPE, etc. |
|
2. Port MPE to Intel platform or provide emulator |
Allowing the development of an emulator by 3rd parties |
|
3. Support for very large disks |
Ability to use up to 300 GB per disk is WIP |
|
4. Port IMAGE to another platform |
No. Consider Eloquence. |
|
5. Port CI |
See 3rd party CI emulators |
|
6. Samba password encryption |
Expected in late ’02 patch. |
|
7. Porting White Paper |
WIP |
|
8. Further Posix smoothing |
No. |
|
9. HP Eloquence evaluation |
Done. Recommend for some customers. |
|
10. User level CI functions |
Investigated. Not likely |
|
11. Integrate internal system logging with syslogd |
No. |
|
12. Port Vplus to HP-UX or Linux |
No. |
|
13. Allow OP or NM users to kill Inetd |
No. Workaround available. |
The Expo seemed a
lot quieter than in years past – you could actually carry on a conversation
just about anywhere on the floor without being assaulted from all angles by
sound projected by amplifiers all set to “11”. Thanks Interex. That was the
good news about the Expo. The bad news, and remember this is just my
perception, is that even with reduced hours, the Expo was lightly attended.
When I ventured out onto the Expo floor I frequently felt that the vendors
outnumbered the regular attendees.
"The
functionaries of every government have propensities to command, at will, the
liberty and property of their constituents. There is no safe deposit for these,
but with the people themselves, nor can they be safe with them without
information. Where the press is free and every man able to read, all is
safe."
- Thomas Jefferson
Thanks to Gary Broder, Secretary of Interex and Board member for pointing me to this quotation. He was responding to an open letter I sent to various executives of HP and Interex and also distributed through several public forums protesting the exclusion of the press from the management roundtables. You can read it in its entirety at www.burke-consulting.com. By the way, Gary is the only representative of Interex who has replied to me as of the Wednesday following the conference.
Here is what happened. On Wednesday of HPWorld week, I found out that the press, including Interx’s own press, was going to be excluded from all the Management Roundtables. I was part of a conversation Wednesday afternoon with Ron Evans, Executive Director of Interex, who claimed not to have been the one to make the decision, but who said Interex had agreed to HP’s request in return for a favor it performed for Interex. For reasons that still baffle me, Ron Evans seemed unconcerned about it and not at all interested in trying to reverse it.
I figured it would not be a big deal; that any press wanting to attend a Roundtable would be able to bluster their way in, possibly with the help of an Interex member. I was wrong. Quite a scene developed Thursday morning at the HP 3000 Management Roundtable when Ron Seybold of the 3000 Newswire tried to enter. It got very ugly, but Ron eventually got in. [It should be noted that none of the HP management members of the HP 3000 Management Roundtable were aware of the exclusion policy and, in fact, specifically agreed they had no issue with any member of the press being present.] Before the incident, I noticed Ed Witkow, current Interex Board member and past Chairman in the room. I went up to him to question this new policy. He too professed ignorance as to who made the decision or why, but just like Ron Evans, seemed disinclined to do anything about it. After the Roundtable began, first Ken Sletten and then I rose to protest this policy. In the meantime, current Interex Board Chairman Bob Combs had entered the room and we had a bit of a verbal confrontation, with Bob basically wanting to defer discussion to sometime “later”.
By now I was on fire, and with the Strategic Management Roundtable next, I strode off determined to find out who at HP had made the request and why. I had to sit through the most boring 50 minutes of the week before the microphone was made available to members of the audience. At this point I rose to question whether this (the exclusion of the press from Roundtables at HP’s behest) was what we could expect from the “new” HP. The panelists started muttering about wanting to have an open dialogue with customers, ya-da-da-da. And about how the press is just looking for a story, not to convey information, etc… The usual excuses from people who feel they have something to hide. Fortunately the moderator, a gentleman from the Aberdeen Group, asked me why I thought it was important to have the press present. This gave me the opportunity to point to the room and the one hundred or so people present and note that the only way the tens of thousands of customers not present could learn about information that could be important to them was if the press was allowed to report on it.
I left the conference on Friday still without any commitment from Interex to never do it again or from HP to never ask again. I wrote my open letter to Interex and HP in LAX airport while waiting for my plane and sent it off Saturday at noon. As mentioned above, Gary Broder was the only person associated with Interex who has replied to date. Of the HP executives, the only person I’ve heard from so far is, and it appears to be a personal reply, Ann Livermore. She expressed surprise that HP asked for the press to be excluded and wrote, “I will check with the organizers of the HP participation in the event and have someone get back to you.” Makes me wonder if the wrong top female executive was chosen to lead HP.
Some at Interex have criticized me publicly, though not by name, for grandstanding and making much ado about nothing. I make no apologies though. Not speaking up would have amounted to tacit acceptance of a wrong policy. If someone’s feathers got ruffled, so be it. I would do the same again. And, will do the same next year if this colossally stupid policy is still in place.
Do not try this at home (at least not until you get
the patch)
Starting with
MPE/iX 7.5 you can use MPE “large files” to get greater than 4 gigabytes in one
dataset instead of using Jumbo “chunk files”. The major advantage of this is it
brings DDX functionality to datasets greater than 4 gigabytes. However,
Tien-You Chien noted that large file datasets are broken out-of-the-box in 7.5
and a patch, TIXMXA4 (?), has already been prepared to fix the problem.
Burke's
Manifesto for HP and the OpenMPE “gang of six”
In the September issue of the 3000 Newswire, in the net.digest column, I reported on OpenMPE Inc.'s so-called "gang of six", expanding on the originals in some cases and adding three items to the list (see www.burke-consulting.com for the original article - look under net.digest for the September issue). At HPWorld 2002, I distributed an expanded version of the article that I somewhat immodestly labeled "Burke's Manifesto for HP" (see also www.burke-consulting.com for the complete text). I did this for several reasons. First, and most important, was to stake out a public position that itemized the most that we in the community can wish for HP to do, hope for HP to do, or expect HP to do, the actual modifier depending upon your point of view. The second reason was to simply give these position statements an identifying name that can be used in the same manner as OpenMPE Inc.'s "gang of six" when having a dialogue. [Aside: A list of the "gang of six" does not appear to exist at http://www.openmpe.org/; however, both the 3000 Newswire article and Burke’s Manifesto contain the original text for the "gang of six".]
OpenMPE, Inc. is somewhat constrained in what they can propose because they have to temper it with their knowledge, based on privileged conversations, of what they may be able to get HP to agree to. Publicly, it cannot be confrontational. Fortunately, I am under no such restrictions and will continue to push, prod, cajole and shame HP into making decisions that maximize the benefits to its stranded customers.
HP's report card from HPWorld:
|
|
OpenMPE "gang of six" |
Burke's Manifesto |
|
Passwords |
Yes |
Partial |
|
Licensing |
Investigate |
No |
|
Allow emulator |
Yes |
Yes, with licensing restrictions that could torpedo the project |
|
Source code access |
Investigate |
Investigate |
|
Documentation in public |
Yes |
Partial |
|
Enable 3rd party software support after 2006 |
Investigate |
Investigate |
|
MPE/iX 7.x on 9x7 hardware |
N/A |
No |
|
Un-cripple A-/N- class |
N/A |
No |
|
Objectives for EOL strategy |
N/A |
See below |
The final, and ninth, item of Burke's Manifesto is different from the rest and was really designed to expose the underlying objectives driving HP's HP 3000 exit strategy. In that sense, it succeeded. Let me explain. In the ten months between 11/14/2001 and HPWorld 2002, HP seemed to be guided by one objective for its exit strategy "migrate all HP 3000 customers to another HP platform, preferably HP-UX". From responses obtained at HPWorld, a corollary objective became apparent, "retain tight control over MPE, releasing it only to the extent necessary so that HP will retain as customers in some fashion those customers unwilling to migrate". This is driving the whole licensing scheme and is something I will continue to address. As I said at the HP 3000 Management Roundtable, it is bad enough that HP has consigned MPE to the technology scrap heap. It is unconscionable that we are being asked to pay to haul the garbage away. Note to HP: just let it go.
Before anyone starts writing me that I am being too negative, let me agree with those people who say the very fact that HP is willing to do anything and, in particular, will not prevent the development of an emulator is big news. It is. HP could say No to everything. But it did not. Look at it this way; while some are satisfied with a partial loaf of bread, I want the whole loaf. And will continue campaigning for it as long as there is any hope of getting more.