I have now received nearly forty private emails concerning my first posting
on this subject. Unfortunately, I've had to give up answering them all. But
as I mentioned earlier, one of the common themes has been, "What do we do
now?" The other primary theme has been on the subject of "once the trash has
been thrown out, it's there for anyone to use."

One person, a president of a well-known third-party and someone who I didn't
expect to say this, wrote the following:

======================================

> >Once you've abandoned it, it's free to be picked over by anyone who
>  >passes by. But even more than that, the users have paid many times over
for
>  >the development of MPE and have an obvious intrinsic right of ownership,
in
>  >and of themselves.
>
>  I have long believed this to be possible, the thinking being influenced
>  by vague memories of the GE 645 system being shuttered and the o/s being
>  treated as public domain. But this was around the late-60's / early-70's so
>  I could be totally off-base.

=======================================

On the other hand, a vice-president of another well-known third-party, wrote:

=======================================

> If I were you I would seriously avoid these suggestions like Sealand or
>  anything else that sounds like you're suggesting that anyone violate
>  licensing or copyright laws.  I think you need to take the high road on
>  these issues.  In today's legal climate, there is essentially no room
>  for "civil disobedience".  If you do something illegal, then they've won
>  and they can crush you.  It's one thing if they go after you for
>  honestly disagreeing with them (including suing them) but if you start
>  handing out copies of MPE, say, then they'll descend on you and the
>  public will probably not feel sorry for you either.
>
>  As far as the legal arguments go, I wouldn't think you could get
>  anywhere with the "once the garbage has been put out" theory, since it
>  proposes to wrest HP's control of their own property from them only
>  because of a choice they made.

=======================================

The legal theory behind "garbage on the street" is called "abandoned
property."

Last week, the Italian police arrested Roberto Cercelletta, a semi-homeless,
semi-mentally ill fellow, for every morning gathering up the money thrown
into the Trevi fountain in Rome. Although he'd been doing this in full view
of the police for the last fifteen years, his prior arrests had always
previously been thrown out because the Italian courts had declared the money
thrown into the fountain to be abandoned property, and thus the property of
no one, available to anyone who bent over to pick it up.

Three years ago, in 1999, Italy passed a law saying that the money that was
thrown into specific monuments of national importance, including the Trevi
fountain, was no longer to be considered abandoned property, but now the
property of the state. However, they set the fine for taking money from the
fountain fairly low, so Cercelletta simply continued to collect the money --
sometimes, on a good day, taking in as much as thousand dollars -- and paying
the fines when he was ticketed. Last week, he was arrested as a habitual
offender.

On this subject, WestLaw (4th Ed.) writes:

"Abandoned property is defined as 'Property that has been discarded by the
true owner, who has no intention of claiming title to it. Someone who finds
abandoned property, acquires title to it, and such title is good against the
whole world, including the original owner.'"

Since last Thursday, when I posted my first email on this subject, I was
surprised to find that there is already a substantial movement afoot in this
regard for software that is no longer sold or supported by the original
vendor. It's called "abandonware." See:

     http://www.wired.com/news/games/0,2101,49723,00.html

Up to now, the abandonware movement has primarily been associated with games,
but a few operating systems and emulators have fallen into this category as
well. Dozens of web sites have sprung up with hundreds of games that are no
longer sold or supported by their original vendors. The original vendors will
not sell, distribute or even discuss these old titles, thus other than these
web sites, there is no way to obtain a copy of the software.

Do the original vendors care? Yes. Surprisingly they don't care about the
copyright infringements, if there are any. They get no money from the
products now, and they know that they wouldn't get any under any sort of
vigorous prosecution. Rather, they sic the lawyers of the Interactive Digital
Software Association, their representative trade group, on these web site
operators simply to maintain their intellectual property rights. IP rights
that are undefended are lost after a bit of time, and they don't want to lose
those rights.

One website writes the following about the vendors' point of view:

=======================================

There have indeed been cases in history where the ownership of an invention
has been misplaced or abandoned, and another person has acquired it (much to
the chagrin of the original owner). But that's not the case with old
software. The people distributing old games over the web have obviously not
acquired the title to that software, nor do they publish proof that the
ownership was ever discarded by the software company. Just because the
company doesn't sell the software any more doesn't mean they've abandoned it.
They've abandoned the consumer, yes, but not the rights to the software.

Another issue floating around is a passage in the National Information
Infrastructure Copyright Act of 1995 that would allow copying the software
legally: Software in which the copyright is older than two years and is no
longer sold on the market would be deemed available free to the public, due
to the seemingly "apparent uselessness of outdated technology" as stated in
the NIICA. I have not been able to verify whether this is true or not; but
even if it was, it would still allow the entertainment software industry to
shut down websites in the name of intellectual property. The fact that the
software is "obsolete" by that definition is irrelevant, because there are
trademarked characters, music, and graphics in the software that are still
the intellectual property of the company that created them. Even the name of
the game is enforcable as IP.

     --http://www.mobygames.com/featured_article/section,26/feature,7/

=======================================

From this author's point of view, there is little hope. But the question
arises, to whom does MPE belong? The knee-jerk answer would seem to be HP,
but I'm not so sure. Like a hundred other people in the MPE community, I've
freely donated my time and expertise to the improvement of MPE. I donated
that effort not to HP, and not even to CSY, but quite explicitly to the MPE
community of users.

The parts of MPE that I will substantial intellectual property rights to are
CIU, the SQL shell on IMAGE, b-tree access in IMAGE, and advanced telnet. I'm
not claiming these rights for monetary purposes nor self-aggrandizement, but
rather because I believe they have now become important. Please allow me to
take a bit of your time and explain the history of these features:

I've been an HP3000 user for 22 years now. Beginning fourteen years ago, when
things were so bleak that SIGIMAGE has disbanded itself because CSY was
rejecting all of its suggestions, I began actively contributing to the MPE
community. Number one among those rejected suggestions had always been to
allow critical item update capabilities in IMAGE.

In 1989, I attended the first HP3000 vendor meeting in Maui. Orly Larson was
the only HP representative in attendance. I took with me an article to give
to Steve Cooper and Stan Sieler to review prior to the article's submission
to Interact, Interex's magazine of the time, outlining the simplicity of
actually putting CIU into IMAGE and the necessity of doing so. Orly Larson
also kindly read it and faxed it back to Cupertino.

The response that came back was, "Who is this person and why is he trying to
do harm to HP?"

I was fairly well flabbergasted by that response. It was completely
unexpected, although I was perhaps too naive. HP had an extraordinary
arrogance about itself then. As a compromise, Orly asked if I would hold off
publishing the article if he would do what he could to get CSY to implement
CIU. I said certainly. My intention was only to get CIU into IMAGE, not to
create a ruckus. Apparently, Orly's efforts were heroic, to the point that he
came very close to getting fired over the issue. Orly finally gave up in
June, 1990 and the article was published in the August issue of Interact,
entitled, "The World's Oldest Enhancement Request," just prior to the Boston
Interex meeting.

At the same time, Ron Seybold, then editor of the HP Chronicle, also
published a second article of mine, "An Open Letter to CSY", describing the
enormous blunder that unbundling IMAGE from the HP3000 was. I wrote this
article only as an open letter in the Chronicle after having the same text
rebuffed by one person after another within HP for nearly a year. These two
articles were purposefully timed to appear just prior to Interex, and they
played some significant part in creating the ambience of the "Boston Tea
Party," an event celebrated in song by Sasha Volokh, Eugene's younger
brother. See:

     http://kuznets.fas.harvard.edu/~volokh/unbundle.html

HP's reaction to the users' demands, as quotes overheard at Boston, was,
"We'll rebundle IMAGE over my dead body," and "This is just a bunch of noisy
vendors trying to save their own asses." Again, I was surprised by the
vitriol and vehemence of HP's reaction.

Following, Boston, I started almost immediately writing a series of articles,
about twelve in all, on the necessity of putting an SQL shell on top of
IMAGE. I considered the lack of that shell to be the most significant
competitive failing that IMAGE possessed at the time. Although I called the
new construct "Relational IMAGE", it eventually became to be called
"IMAGE/SQL." During that campaign, I also attended about a dozen local and
regional user group meetings, arguing for the necessity of an SQL shell on
IMAGE. At one of those meetings, a SCRUG meeting in 1991 in Los Angeles, I
met the new IMAGE lab director, Jim Sartain.

Jim was an enormous breath of fresh air. He wanted to find out what the users
really wanted and what would be necessary to dissipate the level of distrust
and anger that had built up since Boston -- and in comparison to many people
at HP that had preceeded him -- it was obvious that he meant it.  Beginning
in 1991, and lasting until about 1997, I and a number of others (Ken Sletten,
Alfredo Rego, Rene Woc, Fred White, Steve Cooper, Stan Sieler, etc.) flew
once or twice or three times a year to HP to consult on the design of
IMAGE/SQL, wholly at our expense, receiving no other compensation than an
occasional HP-supplied buffet lunch.

Working with Jim was a enormous pleasure, and it was a time of great
progress. Jim listened exceptionally carefully to SIGIMAGE's recommendations,
beginning with CIU, and made it work, although extremely cautiously at first.
This was the first modification made to IMAGE in 15 years, but as soon as CIU
was successfully implemented, work began in earnest on putting an SQL shell
on top of IMAGE as well.

In 1994, I began campaigning for adding b-trees into IMAGE, allowing for very
high-speed partial key searches, using a Boolean algebraic set technique we
had earlier invented for our own QueryCalc product. After just a few meetings
and public talks, Jim accepted that suggestion too and had Allegro put it
into IMAGE under contract.

And in 1997, I worked with Jeff Bandle of CSY's networking lab, again a great
pleasure, to put "advanced telnet" into MPE, making MPE the only O/S to have
a truly efficient and psychologically tolerable telnet architecture that is
usable over planet-wide distances.

I recount all of this, not for any form of braggadocio, but simply to stake
out those parts of the intellectual property rights of MPE that I feel I can
lay personal claim to. Most of these items would not exist in MPE if it had
not been for my efforts and I hereby claim them as my personal intellectual
property. Many other people can equally claim other parts of MPE as well.

Several people wrote me somewhat aghast that I would recommend skipping over
some of the legal niceties. One person wrote:

> Please let me know if I can do anything. I'm not sure Sealand could weather
> a legal assault, much less a military attack, if HP decided to pay for one.

Lawyers are very good at buffaloing and bluster, but the law -- any law --
only extends are far as its jurisdiction extends. FedEx now operates out of
Memphis for exactly that reason. FedEx was built around the hastily written
semester term paper that Fred Smith wrote while he was at Yale, and for which
he now-famously received a "C". After two tours of duty in Viet Nam, Fred
returned to his hometown of Little Rock, AR and used his inheritance of $4
million to purchase Arkansas Aviation Sales in 1971, never giving up on his
idea of airborne freight distribution system built around a hub-based
telephone switching network. Eventually he went through all of his money and
that of his two half-sisters, Fredette and Laura, along with the $8.7 million
dollars that the local Little Rock banks had loaned him.

At the worst of times, his sisters sued him for the eight million dollars of
theirs that he had "lost", he was indicted on forging some legal documents
(which he actually did), he killed a pedestrian at the height of his troubles
while driving inattentively with an expired license. And at basically the
same time, one employee took a regional manager hostage because Fred couldn't
meet payroll, telephoning back to Little Rock saying if he didn't get paid,
"I'm going to blow his head off and sh*t in the hole." But most seriously of
all, the Little Rock banks became extremely nervous and sent the sheriff to
repossess all of his aircraft. To avoid that repossession, Fred scrambled the
aircraft on the ground and radioed the inbound aircraft to divert to anywhere
else. Just stay out of Little Rock. Almost immediately thereafter, Fred moved
FedEx to Memphis, simply to get out of the legal jurisidictional reach of the
Arkansas banks to which he was indebted.

As a vendor, a creator and manufacturer of software, I am extremely concerned
about intellectual property rights. But I don't see MPE as the exclusive
province of HP. Because HP has always considered MPE a bastard stepchild, the
evolution of MPE has been at least as much a community-based effort as it has
been one designed and built from internal directives from within HP itself.
In the case of the "abandonware" games mentioned in the "Wired" article
above, no outside user contributed materially to the creation or manufacture
of any of the now-abandoned games, thus the legal rights of the creators
seems more clear. But things aren't nearly as clear in the case of MPE.

When did HP decide to kill MPE? The first real attempt was in 1988 and 1989,
when Wim Roelandts and Rich Sevick went around to the larger vendors and
customers and quietly told them that MPE was dead and that it was time to
move to "open systems," the area where HP was going to put all of its future
eggs. The Boston Tea Party was the result of the users fighting back.

The second orchestrated attempt was in 1994-95, and the "World's Largest
Poster" was the result that time. But that doesn't mean that some parts of HP
weren't just as grumpy as they were at Boston. While driving the U-Haul
rental truck that contained the poster to California, HP Legal called me on
my cell phone in mid-Arizona and demanded that I take the HP Channel Partner
logo off of the poster. They said it was an inappropriate and unauthorized
use of the logo. If I didn't, they said that we would have our channel
partner priviledges revoked.

I mentioned that because we weren't a reseller of HP equipment, but rather
purely a software developer, the only impact that revocation of our channel
partner status would have on us would be a cancellation of the HP Channel
Partner News "magazine", a small infrequently published folded piece of
paper. But I did readily agree to take the logo off if they wanted me to. I
explained however that the poster was put together in strips and that we
would have to unfurl and assemble the entire poster before we could cut the
logo out and flip it over. But I also explained that we had invited
Computerworld, InfoWorld, Interex, and the local newspapers (the Orange
County Register, the LA Times, etc.) to be there for the poster's assembly,
and they were bound to ask why we were flipping the logo over. I said I would
simply tell them the truth. I never heard from them again.

But HP has never been a monolithic organization. It's always been populated
by people who have cared about their customers as well. In the end, Harry
Sterling vigorously shook my hand, grinning ear-to-ear, and said "Thanks,
Wirt." The people in HP Legal are quite bright folks. They are not the
completely mindless yahoos they often appear to be. They understand
cost/benefit ratios as well as anyone. The cost of bad publicity almost
always outweighs juvenile victories. If, under the worst of all possible
conditions, MPE needs to be moved to Sealand, HP won't do anything,
especially given the fuzzy nature of the IP rights to MPE. MPE is an O/S that
exists only because the users have fought HP management for virtually all its
existence and have actively contributed to at least half of its development.

So when did HP decide to kill MPE this time? As best I can determine, the
first non-management people to know of the decision to kill MPE were told in
January, 2001, eleven months before the users were told, and just months
after Ann Livermore promised the users that HP would continue to actively
build and support MPE as long as the users wanted it. The subject was also
broached extremely obliquely, as a trial balloon, in early March, 2001 on a
vendor's list. While there was a great deal of discussion about the value of
an "Open MPE", among the most interesting exchanges was this:

A vendor wrote:

     "[Open MPE] would seem to be in conflict with HP's desire to retain MPE
as a tightly controlled proprietary operating system."

To which a CSY'er, obviously an old-timer, responded:

     "I am not sure that is an objective of CSY or HP."

And in the month before HP publicly announced the death of the HP3000, Jeff
Vance was still vainly holding onto the hope that MPE could be made open, in
one sense or another. Indeed, he filed this registration:

=======================================

Registrant:
   Vance, Jeff (OPENMPE-DOM)
   26865 Hester Ck Rd
   Los Gatos, CA 95033
   US

   Domain Name: OPENMPE.ORG

   Administrative Contact:
      Vance, Jeff  (JV9143)     [log in to unmask]

      26865 Hester Ck Rd
      Los Gatos, CA  95033
      US
      408-447-5666
   Technical Contact:
      VeriSign, Inc.  (HOST-ORG)        [log in to unmask]
      VeriSign, Inc.
      21355 Ridgetop Circle
      Dulles, VA 20166
      US
      1-888-642-9675 fax: - [log in to unmask]

   Record expires on 13-Sep-2003.
   Record created on 13-Sep-2001.
   Database last updated on 11-Aug-2002 13:34:37 EDT.

   Domain servers in listed order:

   NS.SITEPROTECT.COM           64.26.0.23
   NS2.SITEPROTECT.COM          64.26.38.2

======================================

But it's become exceedingly clear in the intervening nine months since
November 14th that HP management does not want to see MPE survive in any
form. If they did, the decision to allow a general license and an openly
distributed object code emulator could have been made in 36 hours, at any
time in the last eighteen months. Personally, I would not fill out any
further surveys from HP. These surveys are nothing but delaying tactics to
get past HP World and to allow the current user base to further dissipate. HP
perfectly well knows what the users want: they want to continue on with MPE,
eventually running the operating system on extremely inexpensive Intel
hardware, at multigigahertz speeds.

Can it be done? Absolutely, with or without HP's assistance.

Is there a market for an inexpensive version of MPE running on inexpensive
hardware? Again, absolutely, and no better proof exists than Paul McNally's
comments on Friday:

========================================

"I'm a newbie to the e3000. I'd like to learn more about this amazing little
machine. I got the Sys Admin book from Prentice Hall, and it's a great book
to get me started. I'm looking for some other references that I can go
further, preferably online, but getting one more book is not out of the
question.

"When I first started working with this computer, I sorely underestimated it.
In the last couple months, I've really gotten to like the little bugger. It's
stable and gets it's work done. What more can you ask for??"

=========================================

MPE could very likely give Linux a real run for its money in popularity, and
even more importantly, in actual value to a using organization. Indeed, it
were simplified a bit in order to better serve a much a larger audience,
doing things such as eliminating (or at least no longer mentioning) manual
masters, there would be enormous value in the hierarchical database
architecture of MPE, its transaction manager, its file system and its spooler
vis-a-vis the mess that Linux is. MPE is fast, reliable, simple, with
English-like commands and doesn't require a DBA to support it. While data
processing managers often focus on arcane attributes, such as shell scripts,
that isn't the value of MPE and IMAGE exhibit to a using organization. To a
business user, it's the extraordinarily efficient filing cabinet that is
IMAGE that is the core value of MPE.

I've gotten to the point that I no longer trust anything that HP says. If
they had wanted to do the right thing, they would have done it long before
this. They understand perfectly well the users' concerns and the users'
desires. Rather, I believe that it's time for the community of users to stand
up and stake their claim to MPE. It is a product that exists in its current
state only because of the long-term, deep abiding interest in its success
that the users have exhibited. The users have been the faithful shepards in
all of MPE's history, not HP. HP, on the other hand, has consistently
demonstrated itself to be a very unreliable, indeed almost treacherous,
partner for essentially all of its existence.

Wirt Atmar

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