TIMOTHY CASEY

Senior Patent Counsel

SILICON GRAPHICS

MR. CASEY:  Hello, I am Timothy Casey, Senior Patent Counsel with Silicon
Graphics, and representing their views.

Silicon Graphics is the world's leading supplier of visual computing systems
targeted for technical, scientific and corporate marketplace.  The company
pioneered the development of color three_dimensional computing, transforming
it into practical and affordable mainstream solutions that improve the
productivity and increase operational efficiencies across a broad array of
industries, even though as of late, we seem to have gained our greatest
notoriety from our involvements in films like Jurassic Park and Terminator
2.

I would also like to point out that the company was originally founded on an
exclusive grant of a patent from Stanford University that's since run out,
but we did have our basis around the patent.

Although Silicon Graphics designs and manufactures personal computers,
workstations, servers and supercomputers based on our own designs for RISC
processing technology, a large part of our overall development effort is now
focussed on software, including, display, communication, development tools,
operating systems, applications and user interface technologies.

Naturally, Silicon Graphics files a large number of patent applications
related to both our hardware_ and software_based inventions, and has a
vested interest in maintaining such protections.  But rather than use my
time to further expand on the horrors or virtues of software patents, I
would rather state that we are spending too much of our time in these
hearings I believe, discussing the ill patient and not enough time
discussing the disease.  And shooting the patient I don't think is an
adequate solution.

Software as incorporated into the patent system is not the great villain
that many people would like us to see, but rather a misunderstood giant.  As
I mentioned earlier, most computer system manufacturers today invest a
majority of their research and development efforts on software technologies
in order to further distinguish their hardware products from their
competition.  I truly hate to think what would happen to this industry and
this nation's economy, if less than half of this development effort was
subject to patent protection.

It is also important to consider that in practical terms, software is not
really different from hardware.  It's just that the Patent Office
understands hardware and is better prepared to adequately examine
hardware_related cases in most situations.  Patents have been granted on
transistors, resistors, capacitors, clock circuits, filters, and the like,
all necessary building blocks of many electronic designs.  But has that
seriously impeded the electronics industry?  Why is software so different?

Are not most software products composed of basic elemental blocks of code
arranged in new ways to perform new tasks, much like the hardware elements
of any modern electronic product?  The reason software patents have
developed into such a controversial topic is primarily because a number of
overly_broad software patents have been allowed to issue.  And why is that?

Well, we can argue that the prior art is inadequate, and that is true.  And
we can argue that the statutory subject matter tests are inadequate, and
that is also true.  But I would argue that one of the biggest causes is the
fact that the Patent Office has not had the most important tool it needs to
adequately examine software patents; and that is examiners who have the same
fundamental understanding of basic software elements as they have of basic
hardware elements.

I find it absolutely amazing that the Patent Office has issued so few
overly_broad software patents given the level of training of many of the
examiners and the complexity of the application subject matter.  While I
understand that the Patent Office has already undertaken steps to hire
computer science majors in the future to help solve this problem __ which I
loudly applaud __ there are many additional measures that can be taken to
improve the services of the Patent Office, both with respect to already
issued software patents, and any applications that might be examined in the
future.

Some of these measures include revamping the re_examination process, so that
re_examinations can be used to achieve the goals for which they originally
intended.  One step would be to make re_examinations an inter partes
proceeding. Another would be to make them less expensive, both in terms of
fees and the formal requirements of a re_examination request that presently
force potential applicants to seek exceedingly expensive professional
assistance in order to comply with the regulations.

It may also be in the country's best interest if an amnesty period is
implemented over the next year or so during which applicants could institute
a re_examination of any software patent, based on new art, meeting the
requirements __ which may indeed need some revamping __ by simply filling
out a one_page application form and filing a minimal fee, say of $500
instead of the present fees which is well over 2,000, I believe.

Second, instituting some form of prepublication of applications for the
purpose of eliciting industry comment, such as publish issue patents on a
tentative basis pending the discovery of new art during the comment period.

Third, establish routine industry_supported education programs to provide
continuing education for examiners, even up to and including advance degree
study for examiners who commit to an extended tenure with the Patent Office.

Silicon Graphics has participated in both bringing people to the Patent
Office to do presentations on graphics technologies for the examiners.  And
we have also hosted a number of examiner groups at our office in Mountain
View to attempt to give them additional education on our industry and our
technologies.  And I applaud the Patent Office's efforts in that area.

Four, employing technical specialists with broad industry knowledge and
allowing them to roam between examining groups so they can provide expert
assistance when needed to less highly trained examiners.

Five, allowing examiners more exposure to a variety of technologies instead
of pigeon_holing some of them in narrowly constructed examination areas.

Six, providing the examiners with better technical tools, such as network
computer systems that allow examiners to do key element searching of both
text and graphics on a single screen at the same desktop system that they
use for word processing, video teleconferencing, and Internet
communications.

Seven, introducing legislation to turn the Patent Office into a government
corporation, so that the Patent Office can attract and maintain examiners at
competitive pay scales to the industry without being constrained by the
Civil Service pay guidelines, and allowing the Patent Office to actually
keep and utilize all the money it raises from user fees.

Eight, instituting new limitations on the maximum number of claims permitted
in an issued patent, such as three independent claims and no more that 30
total claims, in order to simplify the examination process and reduce the
burden of accused infringers who are often forced to prepare opinions on
patents with hundreds of primarily duplicative claims.

Nine, instituting per_page surcharges for patent applications with more than
35 pages of text and 10 drawing figures to force applicants to be more
succinctly_descriptive of their inventions.

As long as I am on the subject of steps that the Patent Office can undertake
to improve its services to its clients, I will introduce two additional
measures that should be considered: namely, instituting a new type of
expedited patent application, and regionalizing the Patent Office.

Silicon Graphics recently obtained a patent on some technology that was
critical to the protection of one part of our systems, and that we knew has
already been duplicated by after_market suppliers of such parts.  Despite
filing a Petition To Make Special with our application and being extremely
diligent in our efforts to shepherd the application through the Patent
Office as soon as possible, it still took eight months from the filing of
the application for it to finally issue as an enforceable patent.

Now, you may think, "Eight months, that's pretty fast". But it would have
been done in two or three months had it not been for the fact that it took
three months to get from the mail room to the examiner and another four
months to get from the examiner to the final print.  The examiner completed
the entire examination of the patent in less than one month.

In a similar case, we have a patent application on a fairly simplistic
mechanical assembly sitting in the Patent Office since July of '91, despite
the filing of a Petition To Make Special.  During the pendency of this
application, that mechanical assembly was copied by a number of other
companies. Although this only resulted in a small amount of competition for
us in a relatively narrow market for this part, it was a greater concern for
us because the knock_off products are not always of adequate quality, it
could cause damage to our customers' systems when used, for which we would
ultimately be responsible for correcting in order to maintain our customer
loyalty.  This would not have been the case had this application been
allowed to issue.

The reason the one application took eight months and this other application
has taken over two and a half years is because the Petition To Make Special
Process only applies to the examination and not the remainder of the Patent
Office process. What is needed therefore, is a new type of application that
not only gets expedited when in front of the examiner, but throughout the
entire process.  I would be more than happy to pay a higher fee, such as a
1,000 to $1,500 in filing fees for such an application in these types of
situations, because I would surely make that up in outside counsel fees when
the attorneys have to relearn the technology two or three years later after
we get a final office action.

My last suggestion is that the Patent Office seriously consider working to
regionalize the Patent Office.  No one in Washington can truly appreciate
the difficulties in communicating with the Patent Office from the West
Coast.  Because of the time differences between the two parts of the
country, and the new flexible hour programs instituted by the Patent Office,
which I do think is a good idea, there's only a one_hour period each day,
typically between 1:30 to 2:30 Pacific Standard Time, during which a
practitioner on the West Coast or an applicant can expect to get ahold of an
examiner on the telephone.  Because of this, it has sometimes taken over a
month to arrange a telephone interview with an examiner.

My colleagues on the East Coast however, can call at much more convenient
hours, or even walk over to the Patent Office for an in_person visit,
something which would cost me well over a thousand dollars to attempt. 
Given these restrictions, I hate to imagine what the independent inventor or
startup organization on the West Coast thinks of our patent system.

Some other benefits of a regionalized Patent Office would include:  An
ability to draw from a larger pool of potential examiners; new economic
growth in the parts of the country selected for the new Patent Office sites;
and greater public accessibility to the Patent Office records and examiners,
as well as greater exposure for the examiners to the relevant industries. 
It would certainly cut down on your costs of sending examiners across the
country.

I hope these suggestions prove to be useful guides to the Patent Office and
as an invitation to the Patent Office to consult with West Coast companies
and practitioners and applicants for addition solutions that could be
implemented by the Patent Office to resolve our present difficulties, but
also to raise the Patent Office to new heights of service.

Thank you for this opportunity.

COMMISSIONER LEHMAN:  Thank you.

Do you think it would be a good idea, for example, to have Group 2300
located here in this area?

MR. CASEY:  I would imagine that a majority of their clients are in this
area so it would probably be a good idea, or at least some portion of them.

COMMISSIONER LEHMAN:  There were some reactions I had to certain things that
you said, but I don't want to take away from some of the other __ we are
already I think making changes along the lines of some of the things you've
recommended, and certainly part of it has to do with our automation system
and other reforms that we are making.  But thanks very much for your help.

MR. CASEY:  Thank you.

COMMISSIONER LEHMAN:  Next I'd like to call Robert Sterne of Sterne,
Kessler, Goldstein and Fox.