KAYE CALDWELL

SOFTWARE ENTREPRENEURS FORUM

MS. CALDWELL:  Yes. I work with a lot of different  software  or-
ganizations and I'm speaking for the Software Entrepreneurs Forum
today.  I'm the President of that  organization.   I'm  also  the
Legislative  Awareness  Director.   SEF,  as we this Software En-
trepreneurs Forum, is a ten_year_old  nonprofit  organization  of
over one thousand present and future software developers.  Nearly
all of them are located in the Silicon Valley.  We  have  monthly
dinner  meetings  which attract over two hundred people and we've
had  as  many  as  six  hundred  people.   We  also  have  eleven
special_interest  groups that each hold monthly meetion a variety
of technical and business topics.  Our members are  mostly  small
companies, and I think our idea of small is probably not the same
as your idea of small.  By small I mean one to five people in the
company, very small.

The U.S. software industry is unique in that it includes a  large
number  of  very small companies. We have a thousand members just
in the San Francisco Bay Area.  Many of these small companies are
responsible  for the most creative new software developments.  In
the early days of microcomputer software development, nearly  all
software was created by individuals working on their own with one
or two associates; yet these types  of  development  environments
were  responsible for early word processors, spreadsheets and ac-
counting software.  In earlier generations of hardware, I'm  told
that  it was also the case that one_ or two_person companies were
a major factor in leading innovations.  Even today much  software
marketed  by  large  software companies is initially developed by
small, indepen software developers.

SEF's mission is to help these small companies succeed. While SEF
is  local  to  Silicon  Valley,  we  feel  that  SEF  members are
representative  of  the  thousands  of   software   entrepreneurs
throughout the United States. Our members have different opinions
on software patentability, indeed you're hearing  from  at  least
two  of  our members at other points during this hearing. Some of
our members are patent holders and are strongly in favor of broad
patent  software  protection.  Other members are against software
patents entirely.  However, there are issues on which SEF members
are  in general agreement.  Our members feel that the patent sys-
tem favors large companies over small companies, and we feel that
it's important that the patent system both in theory and in prac-
tice should not give big companies an advantage over small ones.

Most of our concerns have to do with Patent Office practices, and
we  understand  that  this  subject  is  scheduled for hearing in
February. We do appreciate the opportu to speak on  these  issues
today.

To the degree that software is patentable, SEF members  want  the
patent  system  to  produce good, clear patents, especially where
the patents are important. We want patents to be issued  and  in-
fringement issues resolved expeditiously.

Uncertainty as to the validity or scope of a patent hurts  paten-
tee  and  possible infringer alike. It makes it difficult to make
decisions, to raise capital, to develop  business  plans  and  to
make  business decisions.  Patents of uncertain scope or validity
are much more damaging to small companies than large ones.  Small
companies  can  rarely afford a good legal analysis on a patent's
scope and validity, and an uncertain situation can  often  put  a
major part of the small company's net worth at risk.  The current
patent system seems to encourage litigation.  We feel  that  it's
important  to  improve  the patent system so that it becomes more
self_enforcing.

While we realize that much progress has been made recently in the
ability  of the Patent Office  eal with software patents, initial
progress needs to be made. In proposing the following recommenda-
tions  for  consideration  we're less concerned with the specific
suggestions than we are in highlighting what we see  as  problems
and in stimulating the Patent Office to solve them.

Our first recommendation would be that the Patent  Office  should
continue  to improve its prior art database by adding to it text-
books, scholarly articles, user manuals  of  commercial  products
and   nonpatent   prior   art   cited  in  existing  and  pending
software_related patents. This would be particularly effective if
such  art  could be added to the PTO's computerized database, but
also be useful to review the trade publications for the last  ten
years  to  identify  significant  software  products  or  product
enhancements so that their manuals could be included in the data-
base.

Our second recommendation is that we  realize  that  patent  exa-
miners must have both technical ability and a knowledge of how to
apply legal principles to determine patentab$y. We encourage  the
Patent  Office's  recruiting  of  examiners with computer science
knowledge.  We encourage the Patent Office to continue to improve
the  quality  and expertise of its patent examiners and software.
We particularly suggest increasing the pay and professional  sta-
ture of examiners so that more examiners see it as a professional
career rather than just a stepping stone to private  practice;  I
think you've heard those suggestions earlier also.

We also suggest putting a high priority on identifying and exped-
itiously  examining  patents  which are likely to be asserted. We
feel that it's important to identify crucial patents and to focus
patent  office  resources on them as the place that will have the
most real_world effect.  The accelerated patent  examination  ap-
pears  to  be a mechanism for achieving this, but in practice the
accelerated examination does not seem to be  having  the  desired
effect.  We believe the criteria for making the accelerated exam-
ination also serve to select those patents which  are(ely  to  be
asserted,  those patents being reexamined, reissued, or where the
patent holder says there's a suspected infringer.

The performance criteria for the Patent Office should  give  more
weight to the examining of high_priority cases rather than simply
counting numbers of patents examined. We suggest that  expediters
be responsible for getting accelerated patents through the system
so they don't get stuck on individual desks. This and other prob-
lems  could be reduced by tracking patents based on the length of
time they've been in the Patent Office rather than  their  length
of time on a particular desk.  In brief we believe it's good pub-
lic policy to identify those patents that are likely  to  be  as-
serted  and  examine them promptly and thoroughly so as to reduce
the uncertainty of the scope of those patents.  This way the  pa-
tent  can take a place in the free_enterprise system as a negoti-
able commodity of reasonably_certain scope.

We understand that the Patent Office is trying out a  preexamina-
tion  interview. A, understand the way it works, prior to the ex-
amination the examiner, the patentee's lawyer  and  possibly  the
patentee  have an interview where they attempt to convey what the
invention is and to identify where relevant nonpatent  prior  art
might  be found.  We commend this idea, which we think has poten-
tial  to  both  speed  the  examination  process  and  create   a
better_quality  patent.   We  also  commend the Patent Office for
trying out new ideas on an experimental basis to try  to  improve
the patent process.

We understand that the level of  skill  and  the  art  needed  to
determine  obviousness  should  be  supported by printed publica-
tions. We also understand that the determination  of  obviousness
is  a legal question.  However, we encourage the patent office to
try to use software professionals and academics  to  help  locate
relevant  printed  publications which would document the level of
skill in the art.

We also encourage the Patent Office to provide further  education
for  patent  applicants,  which includes actual case exa0s illus-
trating how applicants can pursue the question of nonobviousness.
Particularly  important  would  be  actual examples outlining the
examiner's reasoning in determining nonobviousness.

We feel that there's a need for better education of the public on
patents  in general and software patents in particular. The reex-
amination process should be highlighted as a normal part  of  the
process.  The  role  of  prior  art  in the reexamination process
should be made known to the public and to the press in  order  to
reduce the concerns of possible infringers.

The Commissioner has ordered a  reexamination  of  the  Compton's
Multimedia  patent.  We applaud this action as it shows a respect
for the legitimate concerns of  possible  infringers,  especially
small ones.  We propose that the Commissioner, as a standard pol-
icy, order reexaminations of patents at no  cost  on  request  by
small  entities  which  both  present  evidence that they've been
given notice on the patent and produce prior art  or  other  evi-
dence of invalidity.

In the in4sts of reducing the time and expense it takes to deter-
mine  the  validity and scope of patents, we propose that the law
be changed in the following ways. A, have Federal  judges  remand
all  validity  issues  to  the Patent Office for reconsideration.
The courts should still be able to review such actions.   B,  re-
quire that anyone representing a possible infringer who has prior
art on a patent send that prior art to the patent office for sub-
mission into the patent's file wrapper. The penalty for not doing
so would be that the possible infringer could not use such  prior
art to challenge the validity of the patent.  C, limit the number
of reexaminations on any one patent to two except under extremely
unusual  cases,  in  order  to  bring forth prior art at an early
point; require rather than allow that the Patent Office  consider
all previously_unconsidered art in a file wrapper at any reexami-
nation, provided such prior art is filed at some  time  prior  to
three months after the reexamination notice is published.

Finall8e would encourage the Patent Office to take full advantage
of  public  participation  in  the patent process by making their
internal prior art database available on electronic form via  the
Internet  as  well as placing notices of reexamination on the In-
ternet. You spoke about this earlier today,  and  mentioned  that
this was a goal to be achieved several years down the road.  Here
in California there was a law passed last year that went into ef-
fect  January 1st to put all pending legislation on the Internet.
Last Friday that system went on_line.  It took them three  weeks.
You might want to speak a little bit to Jim Warren who's testify-
ing tomorrow morning.  He was very much involved in getting  this
legislation  passed and in getting this system implemented.  So I
think he could probably tell you something about that process.

We expect that the effects of these suggestions would be to force
out  prior art early on so as to more quickly determine the scope
and validity of the patents.

Commissioner LEHMAN, I'd like to thank you for holding  these  hear-
ings  here in Silicon Valley and for giving us the opportunity to
speak. Thank you.

COMMISSIONER LEHMAN:  Thank you very much. I want to commend  you
for a very interesting catalogue of suggestions, and I think it's
very gratifying how a group of individual inventors  like  yours,
not  a  big corporation, can really give so much thought to some-
thing like this and come up with so many very  intriguing  recom-
mendations  that  we're going to be looking at, and on that ques-
tion of the __ putting our system on the Internet, I  would  just
make  an  observation that the quantity of data in our files is a
little bit larger than the legislation currently  pending  before
the  California  Legislature,  and there are a few more technical
problems, but one of the things maybe if I can put in  an  adver-
tisement,  we  have  openings for two positions now in the Patent
and Trademark Office, basically the two top people  who  ran  our
information  systems  program  for  the last seven years have re-
tired.  So we're @uiting for new people to take this over.

Obviously I think one of the problems we have now,  there  was  a
suggestion  earlier that maybe we ought to move the Patent Office
out here, and maybe we should move at least Group 2300 out  here.
I  don't  know.   I see Gerry Goldberg is saying that wouldn't be
such a bad idea, after all we went  through  in  Washington  last
week.   But  this  certainly  is  where the talent is, so I think
we'll be publishing these openings pretty soon, and I think we'll
do  some aggressive recruiting out in this part of the world; and
maybe you can help us to get ourselves up to snuff technological-
ly.  Maybe it won't be three weeks, but maybe it won't have to be
the years that it's taken us thus far to make these improvements.

Anyway, I wanted to  thank  you  very  much  for  your  excellent
suggestions that are really appreciated. Thanks.

Next I'd like to ask Mr. James Chiddix to come  forward,  who  is
Senior   Vice_President   for   Engineering   and  Technology  of
Time_Warner Cable.