LEE PATCH

SUN MICROSYSTEMS

MR. PATCH:  Mr. Commissioner and colleagues, my name is Lee Patch.  I speak
today on behalf of Sun Microsystems, a 12_year_old $4 billion Silicon
Valley_based manufacturer of computer workstations and related software
products.

Sun invests approximately one_half of its substantial R&D budget in software
development, particularly UNIX_based operating systems, development tools
and application programs.

I serve as Deputy General Counsel and Chief Intellectual Property Counsel at
Sun.  I have the responsibility there for the patent activities.

You've heard and will no doubt continue to hear today widely diverging
opinions concerning the virtues, or on the other hand, great evils of
software patents, and you will note no lack of emotion and commitment to the
speakers on either side of the issue.  It's quite remarkable, I believe,
that the normally quiet, calm environment of the patent practice and of the
software development community has been so disrupted in recent years by loud
and impassioned philosophical debate on subject matter that historically and
traditionally had been only of interest to esoteric patent practitioners.

We've seen luminaries such as Mr. Warren take time out of his busy schedule
to speak to you very passionately about evils of software.  What I'd like
you to take with you from these hearings, if nothing else, is that from the
perspective of a company like Sun Microsystems, the system is indeed broken
and needs addressing.  The current operation of the system is creating an
unacceptable amount of uncertainty within the software and computer
industries, and as I'm sure you will appreciate, business executives who
routinely or daily make million dollar, multimillion dollar gambles on
issues of technology or on issues of the marketplace, hate the need to take
gambles and to make bets upon the outcome of a legal system.

In the face of this problem with the system, you have heard radically
different proposals for solution, ranging from abolition of software patents
outright to rather modest suggestions of improvement in the searching
capabilities that exist within the Patent & Trademark Office.

I would like to summarize if I could what I believe to be the three
fundamental problems that have been the subject of much testimony before you
this week.  The first problem that I believe has been identified and
addressed significantly relates to the quality problem within the Patent &
Trademark Office.  The second problem that I would like to address, which
has been discussed previously, relates to the surprise problem, that is of
considerable concern to the software industry.  And finally, I'd like to
address briefly the subject of the tied_hands problem, which is another
aspect of the frustration you may be hearing this week.

To the subject of the quality problem, I would say in my experience that it
is indeed the case that low quality software patents are routinely being
issued __ I should also mention that none of those are being issued to my
company __ but they are indeed being issued.  Largely it appears, due to the
fact that prior art is not available or being overlooked or being
misunderstood, and secondly, because it's a widely_held view that the
nonobviousness standard that is being applied is simply too low a threshold.

COMMISSIONER LEHMAN:  I hate to interrupt your train of thought, but what is
the __ I should probably ask some other people this, but __ what is the
relationship of the Court of Appeals or the Federal Circuit to that?  Do you
think they're not giving us the right kind of guidance that they should give
us on the obviousness standard here?  Or is this our problem more?

MR. PATCH:  I believe it's a merged problem.  I believe that the Court of
Appeals for the Federal Circuit has lowered that standard, and unduly so,
and I also believe that the standard is by no means uniformly being applied
within the Patent & Trademark Office.

In my role as corporate patent counsel, in_house counsel, I spend
considerable time defending against infringement charges, and I have had
personal opportunity to confront most of the famous bad software patents
that you will and have here heard about, including such as the exclusive or
patent, and the Soderblom patents and AT&T Pike patents, and other like the
Mark Williams byte order patent.  There's a list of them which have been
widely disseminated as problem patents.  I have personal experience in
having rebutted and defended against charges of infringement with relation
to many of those.  From that experience, I would like to offer a few
observations for your benefit:

First, there is indeed a quality problem that exists, and it is costing the
industry a great deal in terms of lost cycle time, a lot of expensive effort
being undertaken that would ideally not be necessary.

That having been said, I'd also like comment that no, it is not the case
that in the software industry the sky is falling.  We have not reached a
stage where these types of problems are bringing to a screeching halt
progress in the industry.  Our confrontations with this group of
questionable patents were fortunately all resolved prior to litigation, and
where there was some payment, they were relatively inconsequential amounts.

A third observation I'd like to mention is that to address these kinds of
infringement charges, which are perceived to have little merit, it's often
quite necessary in my position to do some mining in the memories of a series
of old and experienced practitioners of this art, who I happen to have at my
disposal inside the corporate environment.

There is not a great deal of public documentation, patent or otherwise, that
serves as a ready vehicle or mechanism for solving these problems when they
arise.  A lot of dusty basements have to be explored and old computers that
haven't been powered up for many years have to be discovered and reactivated
in order to deal with these problems as they exist today.

The last observation I wanted to make is that the quality problem that I
believe to exist is not unique to the software industry.  In my business
questionable patents are being enforced not just with respect to software
inventions but in the hardware arena and the semiconductor arena, there is a
considerable amount of that going on, driven not so much by the nature of
the technology but by the nature of the economic interests and an
opportunity being presented to people to realize some very significant money
to the bottom line.

The second issue I wanted to mention briefly was, as I said, the surprise
problem.  And that's a very real source of frustration as you've heard at
length this week.  The industry feels as though it's being kept in the dark
for long periods of time, and then surprised by some unanticipated patent
jeopardies. This is clearly the result of delayed publication on the one
hand, coupled with the ability in the Patent Office for an applicant to
prolong the prosecution for an inordinate period of time, with no penalty,
without loss of legal rights.  This creates an environment where surprises
become the rule rather than the exception to the industry.

Finally, the third issue of the tied_hands problem has also been mentioned,
and I'd like to simply summarize my experience as an in_house counsel as it
relates to that.  When a low_quality patent is perceived to have issued in
the software industry, most feel powerless to do anything about it.  There
exists today no quick, cost_effective mechanism to remedy this situation. 
The court system is typically unavailable unless you've already been accused
and would be able to initiate a declaratory judgment action.  And even if it
was available, it's viewed as slow, expensive and lacking expertise.

The current Patent Office re_examination procedure, which has been
discussed, is frankly considered in the industry as a trap to the unwary,
and it is consciously avoided in most cases of the type that I mentioned. 
It's viewed as biased in favor of the patent applicant, and it's also viewed
as dangerous, as a spoiler of otherwise powerful prior art.  Frankly, the
best prior art that you know about you would never offer up into the current
patent re_examination procedure.  You hold that back quite consciously.

COMMISSIONER LEHMAN:  Why is that?  Because you don't want to disclose it
because it's trade __

MR. PATCH:  No.  If you have a good reference, your very best reference, you
don't wish to throw it over the fence to the Patent & Trademark Office and__

COMMISSIONER LEHMAN:  You want to save that for litigation.

MR. PATCH:  You wish to have an opportunity to advocate aggressively the
significance of that reference, and the Patent & Trademark Office
re_examination proceeding simply does not provide that, and as a result of
that lacking, good references are spoiled and no longer of significant use
to you later in litigation.

COMMISSIONER LEHMAN:  And is that why a lot of people just choose to forego
the re_examination process and go into litigation then directly?

MR. PATCH:  Absolutely, absolutely.

A couple suggestions as to how you might address some of these problems:
The quality problem unfortunately is the most difficult of the three.  It's
not one that can be solved overnight, it's not one that can be solved with
the wave of the legislative wand.  It requires, like in operating a
business, daily, consistent execution.  And that is your challenge, Mr.
Commissioner, in implementing a solution to the quality problem that exists
today.

The notice problem is one which I think creates a great deal of the emotion
that you're seeing displayed today, and one which can be directly addressed
and resolved.  Since it's fundamentally due to the delay in the publication,
plus the process in the Patent & Trademark Office which allows one to
prolong extensively the prosecution history, the solution should be quite
clear.  Publish patent applications early and secondly, discourage prolonged
prosecution by measuring the life of the patent from its filing date.  If
somebody wishes to prolong their prosecution for eight or ten years after
that, it's at their own jeopardy.

This may sound in some regards like a call for harmonization, but I don't
wish it to be misunderstood as such, because I'm a strong advocate of
harmonization only when there's a specific identifiable problem being
resolved by such harmonization.

I'd also suggest that the software industry would probably be a supporter of
the idea of implementing a very short time period prior for publications
after filing, perhaps shorter than even the current European and Japanese
model of 18 months.

Concerning the tied_hands problem, the proposal or suggestion that I have is
really in two parts __ a minimal approach and then a more comprehensive
approach.  The minimal approach would involved converting the existing
re_examination procedure into an inter partes procedure, where opportunity
for equal advocacy and appeal would be available.  I would also strongly
recommend that you permit a much broader range of prior art to be introduced
into the proceeding.  Even when a challenged company would like to take
advantage of the re_examination procedure in our industry, it's very
commonly the case that the form of the prior art that we have available to
us is not acceptable in the re_examination procedure, and I would recommend
you open it up much more widely, allow oral testimony, allow physical
demonstration and make it a true inter partes procedure. Address all issues
relating to validity and reference to prior art.

Finally, I would remove any appearance of a bias in favor of the patent
applicant.  I would do so by changing the trier of fact in the
re_examination procedure, and I would strongly recommend you upgrade it and
so the credibility of the Patent & Trademark Office is enhanced and
prolonged.

The more aggressive approach that I would recommend would be a full_fledged
opposition approach.  Time for challenge to a patent should be set for a
reasonable period after its issuance.  I would recommend a
speak_now_or_forever_hold_your_peace approach to that opposition procedure. 
Companies can come forward during the time allotted, challenge with their
best shot, and thereafter the issue has been resolved before the Patent &
Trademark Office, no longer subject to repetitive review by every court that
might have the patent presented before it.  This will have the benefit of
unburdening the courts from dealing with frankly issues they are not
well_positioned to deal with, and it would streamline subsequent court
proceedings to deal with issues relating principally to infringement.

COMMISSIONER LEHMAN:  I think we're running a little over, so maybe we can
wrap up.

MR. PATCH:  I only have one more comment, and I will excuse myself.  Sorry.
The term of the patent in an opposition such as I've recommended should be
extended for the period in which the opposition goes forward to avoid abuses
of well_funded challengers keeping an opposition going for a long period of
time.

Thank you.

COMMISSIONER LEHMAN:  Thank you very much.  Those were interesting ideas. 
If you were here yesterday, you know, we announced actually that we have
already taken one of your suggestions, and that is that we are going to be
submitting legislation to Congress to deal with the so_called submarine
patent problem in part by moving to a system 20 years from filing will be
our new patent term, and that will at least in part address that difficulty. 
And then it turned out to be kind of a win_win situation because we were
able to get the Japanese to make some concessions to us in return for our
agreeing to do that.

And I'd like to also just use this as a forum to make a point about
harmonization process.  We announced earlier this week that we were going to
suspend the patent harmonization exercise that we had been engaged in with
the World Intellectual Property Organization which would have required us to
change to a first_to_file system in the United States.  That's not because
this administration or I am opposed to harmonization.  In fact, quite the
opposite.  Clearly the best patent system for Americans would be one in
which you could file a patent in our Patent Office and then with great
certainty and trustworthiness get very, very rapid protection everywhere
else in the world that would be sound and that you could trust and so on.

That was not the deal that we had cut in the last administration, that
wasn't the harmonization exercise that we were engaged in.  And so we're
basically going back and we're not going to abandon the principles of the
system that many Americans feel favor them until we really have a system in
which we receive very, very tangible results in other patent systems.  And
indeed this illustration of where we move to __ or we agree to do something
that is also in our benefit in this case, I think the 20_year term from
filing, we have received the tangible benefit from the Japanese.

I am very optimistic that we will, in fact, achieve true harmonization at
some point in the not too distant future, but it won't happen if we just
simply make all the changes unilaterally.  So since you raised that point, I
just wanted to explain where the Patent Office was on that.  Thanks.

Yeah, I guess Christopher Byrne, Senior Intellectual Property Counsel for
StorageTek, who is representing the American Committee for Interoperable
Systems, ACIS.



Back to The San Jose Index

Forward to Christopher Byrne