Public Comments Resulting From:
Notice of Public Hearing and Request for Comments on the Proposed Plan for an Electronic Public Search Facility
67 FR 17055, April 9, 2002
NOTE: All comments were scanned from paper documents whether e-mail, mail, fax or hand-delivered. Personal home addresses, USPTO e-mail header information, and fax/copier information have been removed. Comments were extracted from the original and are viewable from the links below. Click on the "Original Document" button to open a new window and view the original. Comments are listed in the order they were processed. Also included are the comments presented at the Public Hearing. (6/4/2002)
|NAME ||AFFILIATION |
| || || |
| ||Jim Davie ||Private Individual |
| ||Nils Victor Montan ||President, International Trademark Association |
|3 ||Daphne Hammond ||TM Office Public User Soc. |
|4 ||Allan M. Lowe, Esq. ||Lowe, Hauptman, Gilman & Berner |
|5 ||James P. Chandler |
>>Post-Public Hearing Statement
|President, National Intellectual Property Law Institute |
| ||Vincent E. Garlock |
Michael K. Kirk
>>Public Hearing Statement
|American Intellectual Property Law Association |
| ||Jennison & Shultz, P.C. |
>>Public Hearing Statement
|Jennison & Shultz, P.C. |
| ||Christopher E. Kondracki ||Specialized Patent Services |
| ||Joseph E. Clawson, Jr. |
>>Public Hearing Statement
|Private Individual |
| ||Calvin Van Sant |
>>Public Hearing Statement
|VanSant Patent Services |
| ||Robert B. Weir |
>>Public Hearing Statement
|National Intellectual Property Researchers, Assoc. |
| || |
>>Public Hearing Statement
|Patent Arts, LLC |
| ||Harold L. Novick ||Nath & Associates, PLLC |
| ||Glenn E. Wise ||Registered Patent Agent |
| ||Terrence L. B. Brown ||Shlesinger, Arkwright & Garvey |
| ||R. Lee Grantham |
>>Public Hearing Statement
|Nixon and Vanderhye, P.C. |
| ||David Testardi ||Patent Attorney & searcher |
| ||Henry W. Leeds ||Drinker, Biddle & Reath, LLP |
| ||Hayden Gregory ||ABA-IPL Section |
| ||Charlotte M. Kraebel ||Attorney |
| || |
Michael H. Minns
Mark A. Watkins
|Hahn Loeseer and Parks, LLP |
| || |
Donal B. Tobin
|The Gillette Company |
Start of Comments
I would like to comment on the Federal Register Notice of April 9,2002. On page 17055
the following information is set forth:
111. Proposed Plan for an Electronic Public Search Facility
A. Status of Replacement Electronic Search Systems
The U.S. patent image database contains the complete printed patent of all patents
granted from 1790 to date, including bibliographic information, specifications,
drawings and claims. Extensive efforts over a number of years have been undertaken to
ensure that this database is as complete as possible. As patents issue each week on
Tuesday, this database is updated the same day with that week's issue of patents.
The line "The US. patent image database contains the complete printed patent of all
patents granted from 1 790 to date, including bibliographic information, specifications,
drawings and claims." Is substantially correct but cannot stand with out explanation. The
PTO is doing a fantastic job. We should give our customers all the facts and not overstate
our capabilities. The optical file is a fantastic search tool but it cannot be better than what
we put into it.
A significant number of the patents granted from 1790 to July 2, 1836 are not in the
database and will never be in the database. Drawings for a few of the utility patents have
been lost and thus "the complete printed patent of all patents" cannot be made available.
Also, I believe that there are situations where some of the data in currently issued patents
is not compatible with current OCR capabilities and thus I don't believe that these patents
are currently available in the optical file.
The line "Extensive efforts over a number of years have been undertaken to ensure that
this database is as complete as possible" would seem to imply that this task has
maintained the optical file to include all patents available in the paper file. The paper
patent file contains design drawings, which have never been reproduced by the PTO and
are not currently in the optical file. These were not disposed of as scheduled and there is a
current effort to scan these into the optical file; this task is far from finished. The people
responsible for ''to ensure that this database is as complete as possible" would have
unknowingly allowed the only source of thousands of design patent drawings to be
disposed of. It was only through the volunteer efforts of myself, Elizabeth Dougherty,
Diane Davie and the fantastic Holly Baynahm that these irreplaceable design drawings
have been preserved.
In response to the request for comments published in the April 9,2002 Federal Register,
the International Trademark Association (INTA) is pleased to offer its support for the proposed
plan to eliminate the paper trademark collections from the public search facilities and to
transition to electronic trademark information collections. INTA takes no position with respect
to the paper patent collections.
INTA is a 123-year-old not-for-profit organization comprised of over 4,000 member
companies and firms. It is the largest organization in the world dedicated solely to the interests of
trademark owners, The membership of INTA, which crosses all industry lines and includes both
manufacturers and retailers, values the essential role that trademarks play in promoting effective
commerce, protecting the interests of consumers, and encouraging free and fair competition. The
members of INTA routinely apply for and maintain trademark registrations. They are the
customers of the U.S. Patent and Trademark Office (PTO). INTA therefore has a vested interest
in working with the PTO to ensure that a fully automated search system has a thorough archival
capability, is entirely accurate, and is fully accessible to the public. And in supporting the
elimination of the paper files, we trust that the PTO has taken the necessary steps to ensure that
the present automated system continues to meet these criteria. In that light, we note with approval
that the PTO intends to develop enhanced software vehicles that maintain and improve the
automated search system.
We strongly suggest that as part of the planned enhancement, the PTO upgrade the
manner in which the electronic search system categorizes and accesses design marks. Because
the availability of design records is dependent upon those designs being categorized and coded
by a human, having a consistent, well-coded database is not always a certainty. In certain
instances, this can be an obstacle to a thorough search. INTA encourages the PTO to resolve this
problem, so as to ensure that the database of registrations and applications consisting of design
marks is consistently coded, such that a search for a design mark will yield the same consistent,
reliable result as a search for a word mark. Commercial searching companies, who have for some
time been manually coding all design records to enable thorough searching and recall, and a
review by the PTO of existing paper tiles prior to their elimination, may prove useful to the PTO
in enhancing the electronic system. JNTA stands ready to assist the PTO with this ongoing effort.
Thank you for the opportunity to submit comments on the proposed plan to eliminate the
paper trademark collections. Should the PTO have any questions or comments concerning the
INTA response, please contact INTA Washington Representative Jon Kent at (202) 223-6222.
Nils Victor Montan
This is a request to speak at the public hearing scheduled for May 16, 2002 regarding the "Proposed Plan for an Electronic
Public Search Facility". Daphne Hammond will speak on behalf of the Trademark Office Public User Society. Following are
the Society's written comments for inclusion in the record. Please contact the sender for any further information or to
confirm this request.
Trademark Office Public User Society
2518 Fort Scott Drive
Arlington, VA 22202
The United States Patent and Trademark Office (USPTO) is proposing a plan to convert its search library to an all
electronic facility. Although comments or questions were solicited in a Federal Register published August 27,2001, it
is disconcerting that we have received no response to any of the questions or issues we raised at that time.
There is no doubt that an excellent automated search system should exist and could replace the paper search system,
despite a concern that the "browse factor", a unique and irreplaceable element of the paper system, would be lost in an
electronic environment. Searching is frequently a subjective endeavor, where in the process of conducting a search, a
stray reference will alert the searcher to a new and rewarding strategy not previously considered. That being said, years
of neglect and insufficient quality control have taken a toll on the paper records. The automated search system has been
a welcome adjunct to fill in the gaps and errors that have crept into the paper system. The automated search system has
its own series of problems which are different from the problems in the paper record. First, there appears to be a
significant lack of quality control during the input period resulting in uncounted errors in bibliographic data, improper
or missing design codes, illegible or missing images, images associated with the wrong mark and other significant
missing data elements. Keeping in mind that this is the same system utilized by the Examiners, the economic and
business ramifications to the agencys internal and external customers are significant.
The USPTO is mandated to maintain a searchable record of trademarks. It would be intellectually dishonest to interpret
that mandate to refer to registrations only, since previously filed applications can be just as significant to an examiner
and devastating to an applicant. It would seem to be in the United States best interest for the USPTO to create and
maintain the most accurate and complete record possible. Despite significant expenditure of funds, many of the
problems identified by previous Government Accounting Office reports still remain and have not been addressed by the
agency. Thus, it seems premature to eliminate the paper search file until such time as significant improvement in the
electronic systems data integrity can be verified.
By way of example, the USPTO proposes elimination of the following specific items currently existing in the paper
files: color marks, Paris Convention marks, government agency logos, Defense Department weapons names and
sensory marks. The USPTO never created an adequate coding system to make sensory or color marks searchable in the
initial design of the electronic search system. During subsequent redesigns and upgrades, some new codes were created
to accommodate those marks but the thousands of existing registrations and applications were not retrofitted with those
codes. Similarly, there has been no cohesive effort to ensure complete capture of Paris Convention, Government
Agency or weapons marks in their entirety. Many are missing the image associated with the mark and some may be
missing entirely. It makes little sense to eliminate one system in favor of another if such significant discrepancies in the
data have not been identified and corrected. The automated system as it stands now is not a reliable substitute for the
paper, just as the Internet search system offered by the USPTO (TESS, TAM) is a poor substitute for X-Search and
While we applaud the significant efforts the USPTO has made to move into the electronic age, much has to be done to
ensure a smooth transition with reliable data for accurate information dissemination. We continue to offer our
assistance in any manner that might aid the agency in this endeavor.
Very truly yours,
I believe the PTO will be making a grave error by eliminating paper files. The most expeditious way to perform many searches is
by visual inspection of the patent drawing. In addition, word matching searches are frequently ineffective because the
searcher and the patent application drafter do not use the same terms. If the paper files were eliminated searches relying on the
patent drawing could not be performed.
Allan M.Lowe, Esq.
Lowe, Hauptman, Gilrnan and Berner
1700 Diagonal Rd.
Alexandria, VA 223 14
This notice is to confirm that the National Intellectual Property Law
Institute, a non-profit organization, is interested in acquiring the
paper files from the USPTO and is further interested in participation
in the public hearing on May 16, 2002.
If you have any questions or require additional information, please do
not hesitate to contact me by return e-mail, telephone or facsimile at
the coordinates listed below.
Professor James P. Chandler
NATIONAL INTELLECTUAL PROPERTY LAW INSITUTE
1815 Pennsylvania Ave., NW
Washington, D.C. 20006
Post-Public Hearing Statement
Pursuant to 17055 Federal Register Vol. 67, No. 68 dated April 9,2002, please be advised
that the National Intellectual Property Law Inutitute, a non-profit organization. is interested in
acquiring the paper patent and trademark registration collections to be removed from the USPTO's
public search facilities.
If you have any questions or require additional information, please do not hesitale to contact
me at 202.789.0234.
Professor James P. Chandler President NATIONAL INTELLECTUAL
PROPERTY LAW lNSTITUTE
On behalf of the members of the American Intellectual Property Law Association (AIPLA), allow this to serve as a
request to speak at the public hearing concerning the Proposed Plan for an Electric Public Search Facility on May
16, 2002. A written copy of any and all remarks for inclusion in the record will be available at a later date.
I would appreciate it if you could confirm receipt of this request. If you have any questions, please feel free to
Vincent E. Garlock
Deputy Executive Director
The American Intellectual Property Law Association (AIPLA)
2001 Jefferson Davis Highway, Suite 203
Arlington, VA 22202
(703) 4 12-4355 Direct
(703) 415-0786 Fax
Public Hearing Statement
The American Intellectual Property Law Association (AIPLA) appreciates the
opportunity to present its views on the Notice of Public Hearing and Request for Comments on
the Proposed Plan for an Electronic Public Search Facility, published in the Federal Register on
The AIPLA is a national bar association of more than 13,000 members engaged in private
and corporate practice, in government service, and in the academic community. The AIPLA
represents a wide and diverse spectrum of individuals, companies, and institutions involved
directly or indirectly in the practice of patent, trademark, copyright, and unfair competition law,
as well as other fields of law affecting intellectual property.
This Association, in letters of October 9, 2001 and October 29, 2001 to the USPTO
regarding the removal of the patent and trademark document collections, acknowledged the
burden placed on the USPTO of having to maintain both paper and electronic collections and
supported the removal of the paper collections. We also stressed the need for the Office to
continue to improve the electronic search systems and, upon removal of the paper collections, to
redirect the resources used to maintain those collections to the further improvement of the
We believe that the electronic collections are now capable of offering all the hctionality
of paper collections. In addition, the electronic collections provide numerous advantages that the
paper collections do not provide, such as collection integrity, convenient remote access, easy
copy making, and space saving, among others. With the vast improvements made in the
effectiveness of electronic searching, the electronic collections are being used by more and more
users and the paper collections are being used by fewer and fewer. Looking at the use of both
types of collections in the USPTO public search rooms, most searchers use the electronic
systems and very few use the paper collections.
AIPLA continues to believe that every effort should be made to improve the electronic
collections of patent and trademark documents. These improvements could take the form of
more public electronic search terminals, internet access to the electronic collections which are
now available in the public search rooms, better public access to the same databases and
collections that are available to examiners including improved foreign patent document search
capabilities, reduced search systems down time and adequate backup search systems.
Given the present functionality of the USPTO electronic collections and the expectation
that the Office will continue to improve and update them, AIPLA supports the elimination of the
paper patent and trademark registration collections from its public search facilities and the
transition to electronic collections. While we would welcome another organization acquiring
these paper collections, the absence of any willing recipient should not prevent their elimination.
We appreciate the opportunity to offer our views and would be pleased to answer any
questions that you might have.
Michael K. Kirk
A representative of the Jennison & Shultz, P.C. Law Firm will speak at the public hearing on May 16, 2002.
Public Hearing Statement
Normally at public hearings a speaker gives thanks for the opportunity to speak. I
am not here to thank the bureaucrats who, by caveat, have erroneously determined that
the paper search records of the Patent and Trademark Office Crystal City Patent Search
Room and Trademark Search Library are no longer needed for public reference. I fear
that a decision has already been made. I think that would be disrespectful to the purposes
of holding public hearings, such as this one.
We are members of a law firm that specializes in trademark law. Members of our
firm have been using the Trademark Search Library of the United States Patent and
Trademark Office since 1939. In addition, members of our family have been associated
with the USPTO and its predecessor agencies since 1909. Therefore, we have had a close
relationship with the workings of the trademark search facilities for close to a century.
Based on our knowledge and experience, we are very aware of the value of the
data contained in the public search records. We are familiar with the many reasons that
the public needs and uses the information contained in the records through our
continuous relationships with and representations of individuals as well as small and
large companies and corporations. In addition, we are and have been the Washington
associates for many U.S. and international firms. We understand the need for
maintaining the integrity of the valuable resources located in the public records of the
USPTO. As a result of our constant daily working with the records, both automated and
paper, maintained by the agency, we have been able to study the benefits and the
problems that appear in each of the formats.
We do not object to the development of a plan to remove the trademark classified
paper files from the public search facilities, provided that prior to the plans
implementation and removal of any paper files, the USPTO must completely demonstrate
to the satisfaction of the user-public and the Congress that the automated records that
replace the paper files are complete, up-to-date and reliable with respect to all of the data
currently maintained in the non-automated records.
To date, the USPTO has been unable and unwilling to publicly make that
Discrepancies in Search Results
A simple comparison of two exact mark searches illustrates the lack of reliance
and accuracy of the X-search system and the lack of federal trademark notice of complete
reliance on the automated records. Attachment A shows Registration No. 1,377,536 for
the service mark of the letters R F with a design of a shield, house and stars. The
registration is searchable and locatable under the letters and, for example, the house
buildings and scenery design, in the paper search room records. By contrast, a review of
the automated record displays the service mark as consisting of only the letters R F. No
design elements are searchable, therefore in the automated records, the design has no
trademark notice to potential users of confusingly similar marks.
Attachment B displays Registration No. 1,585,102 for the letter 0 with
geometric horizontal lines, shadows and a leaf design. The trademark is searchable and
locatable in the paper records as the letter O, the design element bars and the
vegetation element. The automated records, by comparison identifies a pseudo mark as a
letter 0 and does not identify the actual mark as an 0. Further, only the leaf design
element is entered in the design field of the mark. In other words, an electronic search
record of the Ohio State University 0 with lines registration is incomplete.
Of course, many more examples may be given, but I am not up here to belabor the
point that when something is not right, it is wrong. The examples simply illustrate that
the electronic system is not complete, not accurate and not reliable enough, at this time,
to justify the elimination of the paper search records.
This is a lesson previously taught to but not learned by the PTO. Attachment C is
a copy of the Government Accounting Report (GAOAMTEC-91- 1) dated October 1990
that identifies historical data quality problems at page 5 under the heading "Data Base
Inaccuracies May Compromise Quality of Registration Process"
The GAO report brought to mind Assistant Commissioner of Trademarks
Margaret Laurence's quote that "We (Trademarks) wanted to automate in the worse way,
and we did."
Equivalence and Completeness of Systems
Another point of controversy over the plan to eliminate the Trademark Search
Library paper records is that it fails to meet any obvious test of comprehensibility or
coherence. It is a faqade of rationality. The Federal Notice glosses over the pending
application and abandoned application searchable records. By citing Section 4 1 (i)( 1) of
Title 35 of the United States Code, the PTO states that it is only responsible for
maintaining trademark registrations arranged to permit search for and retrieval of
information. That responsibility is not met with only the automated records.
a. Historical Completeness
There is no equivalence of the completeness of the automated records versus the
paper records. The automated records only carry the registration and application records
from 1983 to date, and many of these records are missing and incomplete. We have also
discovered and reported to PTO that information has been inexplicably purged from the
automated search system. It is the paper collection records with its microfilm of canceled
and expired trademarks that are arranged to permit search for and retrieval of information
on all trademarks from the first registration to those issued two days ago. The automated
records hold 19 years of trademark registrations; the paper search record system holds
over 100 years of search and retrieval on all trademark registrations and applications. The
classified paper records maintain registration certificates, application drawings, and
registration and application status data that are not available in the automated system.
They also maintain amendment, assignment, consent, correction and status information
that the Office has failed to capture and maintain in the automated search and status
systems. Attachment D is an example of an amendment to a design trademark entered in
the paper search records, but missing from the automated system.
b. Statutory Notices
The Federal Notices misleadingly states that the database also includes the marks
protected under Article 6ter of the Paris Convention. Attachment E displays the
Convention mark by the WIPO for European Atomic Energy Community (EURATOM)
as missing from the automated records, but with full copy of notice and image from the
paper records. The PTO acknowledges this shortcoming in the TMEPS 1205 Copies are
filed in paper records of the Trademark Search Library, and pertinent information is
entered in the automated search records. However, since many of the images associated
with these entries are not currently available by computer, they must be found in the
Further statutory notices, including United States Government agency notices under
Executive Order 1 1628 have been filed and maintained throughout the paper records.
Attachment F shows a typical government agency notice for the Federal Bureau of
Investigation. No such notice exists in the automated records, and if the paper records are
eliminated, none will exist as statutorily required. The missing electronic notices bring
into question the agencys commitment to providing the govemment mandated
information to the public.
From the trademark examining attorneys perspective, the only relevant trademark
information is the live trademark registrations and applications. But from the publics
need for research, the entire collection is needed and only the paper records at this time
permit the search and retrieval of all trademark information. In order to provide our
clients with the most comprehensive and accurate information from the public records,
we conduct searches of the paper and the automated records currently maintained in the
Trademark Search Library. Yes, we continue to find discrepancies in both formats. We
have documented and reported thousands of references inexplicably missing from the
automated records or that are incorrect in the paper records. The problem is caused by
several factors, including input errors, data maintenance and the limited capability to
retrieve the information from the automated search systems. Neither system is equivalent
of the other.
Negative Impact to the Public
In our opinion, officials looking into the subject issue do not fully understand how
the Trademark Search Library is used. The Office is only concerned with 2(d) citations.
They do not have any use for information relating to abandoned applications or canceled
or expired registrations; therefore, they do not maintain this information indefinitely in
the automated records. It is maintained, however, in the paper records and the microfilm
records in the Search Library. These records provide valuable information in the areas of
possible common law use, marks that have run into problems in the past and ownership
The public needs a comprehensive search system that provides completeness. The
electronic system does not provide the completeness, at this time. Only a search system
incorporated from components of the electronic, paper and microfilm records provides
the most complete records and meets the need of the public. At the present time,
however, our clients will be damaged by the elimination of the classified paper drawings
and registrations because the automated records alone fail to give notice of trademark
As experts in the field of trademarks, we certify that the implementation of the
plan to eliminate the paper search record system will negatively impact the public.
Good Faith Issue
We are deeply disappointed and concerned by the attitude recently shown by the
agency. For centuries the intent of a trademark from common law to statutory protection
is to give notice of claimed rights. For over one hundred years the agency has maintained
the paper records with full support for ensuring the best public trademark notice possible.
Any decision to eliminate the classified paper search records is premature at best.
Anyone with first hand knowledge on the how poorly the automation search systems
have been developed and implemented is horrified that the best possible back up system
will be eliminated. The hasty decision to eliminate the paper records will make waste.
This administration departs from the best information available, regardless of the
medium. This administration now wants all things electronic and proposes to shirk its
agency responsibility of maintaining public notice to obtain the electronic environment.
Those of us who use the records in the Trademark Search Library know that great harm
will be done throughout the trademark world by the callous approach being considered. It
is our opinion that anyone who conducts a search without using both the paper and
electronic records may be negligent. Further, while the Trademark Office is not
accountable for missing citations during examination, such omissions cost the public
dearly in opposition and infringement costs.
There is a PTO story that Thomas Jefferson started the shoe paper filing system
for patents that was integrated into the trademark side as well. I do not think the lore is
true. I suspect that some worthy public servant started the practice, but that credit was
given to Jefferson. So I want to now thank the generations of exemplary public servants
who have strived for and sweated over the establishment and maintenance of the paper
records of the Patent and Trademark Office. Whether it was Carl Jennison, my
grandfather, or Harold Pitta, Cathy Terry or Virginia Johnson, it is their efforts that were
true to the notice requirements of the American public. It is their efforts that the present
administration should consider and compare themselves to for justification.
These comments were prepared by Kathryn Jennison Shultz, John N. Jennison,
Carl E. Jennison.
Respectfully submitted, Law Offices of
Jennison & Shultz, P.C.
Attachments: A B C D E F
I request the opportunity to speak at the Public Hearing on May 16,2002.
Christopher E. Kondracki
SPECIALIZED PATENT SERVICES
2001 Jefferson Davis Highway, Suite 505
Arlington, VA 22202
Phone: (703) 41 5-1 555
Fax: (793) 41 5-1 557
I would like to speak at the hearings. I would represent only myself.
Public Hearing Statement
My name is Joseph Clawson and I live in McLean, Virginia. I do not claim
to represent anyone other than myself in this testimony. Next month will mark 30
years of my life devoted to the Intellectual Property aspects of the patentability of
patent claims, patentability determination, and the searching of the Prior Art. In
that time I have accessed perhaps 10 to 12 million documents relating to the
patentability of claimed subject matter.
I was a patent examiner in the semiconductor and computer static memory
technologies from June, 1972, until December of 1997. In February, 1983, I was
personally asked by then Asst. Commissioner Frank Burnett to aid the General
Counsels Office of the Copyright Office in formulating a position on integrated
circuit mask design. I was co-developer of the semiconductor Class 257 structure
classification beginning about 1990; I reclassified the non-volatile floating gate
static memory subclass 185 into 33 new subclasses in 1995. Since then I have
acted as searcher in numerous litigation cases, both here in the U.S. and overseas;
the estimated total of litigation I was involved in is over $100 million in the last
three years alone.
Thus I can speak with some authority as an expert as to the relative merits of
the various manners in which the Prior art can be best and most completely
recovered, and the judiciousness of the proposed elimination of the paper search
files at the U.S.P.T.O. as is stated in the April 9,2002, Federal Register, Vol. 67,
NO. 68, pp. 17055-17060.
It is repeatedly stated in the Federal Register that:
...the paper patent and trademark registration collections are no longer
needed for public reference because of the availability of mature and reliable
electronic search systems in its public search facilities.
There is NO EVIDENCE for such an assertion or conclusion.
Further, these mature and reliable electronic search systems are never
identified. They certainly cannot be the current error prone, unreliable, and
inherently logically defective systems which use the present BRS search engine.
This is the system which lost the patents of months of October and November in
early December, 2000, and is the search engine which never could even adequately
do its original designed purpose of looking up authors in a library catalog. It is
unclear if these missing months of October and November, 2000, or other months,
have ever been fully returned. It is known that the text files of over 100,000
patents from 197 1 to the present are also lost.
Such assertions in the Register thus appear to fly in the face of the daily
experience of myself and others who use - or more properly put irttempt to use -
the electronic search systems of the USPTO, and who routinely find such
electronic search systems clearly inferior to the existing classified paper patent
library in most applications.
Further, the important existing classified Foreign patent documents and
technical Literature paper collections are also NOT available to the public with
these mature and reliable electronic search systems in its public search facilities
now, and there have been no proposals to provide such in the future. Only the
current classified paper files provide this essential search resource.
Thus the prima facie case of the Register notice has not been made.
Amazingly, using the classified paper search files at the USPTO in a foreign
suit, I was able to find better Japanese Prior Art than the Japanese patent examiner
in litigation involving a Japanese patent [kokoku]. This shows the enormous
power and strength of the U.S. classified paper file system, something which
cannot be duplicated by using the commercially available electronic data or
Abstracts which accompany these Foreign references. It is not clear exactly how
applications are presently searched in Japan, but whatever particular system they
use, the U.S. Classified paper file system is clearly superior. Asan examiner when
classifying foreign patents, we would routinely ignore the Abstract, which was
almost always written by someone with little knowledge of either English or the
technology involved, and instead rely upon the drawings and brief translation of
section(s) of the text for classification purposes. Clearly, no one uses the so-called
International Patent Classification system - neither Europe nor Japan. It is largely
useless for searching. From my professional experience, only the robust classified
U.S. search system provides an adequate basis for determining the differences
between the Prior Art and the claims at issue, and only the paper files can do this in
a time effective manner.
Thus we need to preserve and expand not only the U.S. paper fles library
but also the U. S. Classification system as well.
In the early 1980s many patent examiners had an opportunity to start using
electronic database searching, using "key word" search strategies. Their initial
enthusiasm was often soon blunted as they found that they could not locate,
electronically, the references that they personally knew were there from their
manual searching. Others who were expert in the various technologies also came to
much the same conclusion. This finding was hrther buttressed by scientific
evaluations such as the March 1985 paper by David C. Blair et al., in the
Communications of the ACM, "An Evaluation of Retrieval Effectiveness for a Full-
Text Document-Retrieval System." This result was repeatedly verified in later
studies by others.
While this published paper reports the results done using a generalized
database of only 350,000 pages, when applied to a Patent database, another unique
problem arises: in many, if not most, cases the invention is never fully described
"in the words." The Patent Law requires only that the Specification, including the
Drawings, together be understandable and enabling to one of ordinary skill in the
art to make and use the invention. "The words," in many if not most cases,
merely "flesh out" what is shown in the Drawings and do not replicate "in words"
what is in the Drawings, but are ancillary thereto. Thus, in a Patent database
electronic search one is often presented the additional problem of "searching" for
"words" which were never there to begin with.
Thus "automated electronic searching," or "electronic search systems" while
at times useful in a limited manner cannot hlly replace an actual properly
classified manual paper search file, in many searchers' experience. You cannot,
quite literally, "find the references'' searching electronically. Only a properly
classified paper file can do this.
A more correct way to view the searching of an electronic database using
"word search" is to consider that "ALL the information is lost," until you can
discover some manner - usually a random process - of getting a portion of it back.
In the beginning of the Gulf War, the first night of the attack on Baghdad saw
CNN broadcasting a sky filled with antiaircraft shelling. And all missing their
target. They were shooting randomly, blindly into the air - hoping, praying to hit
something. But they hit nothing. This is the exact same blind nature of "keyword"
searching. As Mr. Randy Rabin points out [Intellectual Property Today, Vol. 9,
No. 5, pp.60-63, May 2002, and NOT available electronically], in such
keyword searches ...The searcher is his or her own lexicographer, and search
success depends on whether the searchers verbal imagination is a match for that of
the writer of the patent. As above noted, in the Blair et al. paper, there was a
staggering electronic loss of about 80% in this small database of 350,000 pages. As
the database gets larger, the retrieval rate goes down even further and the error rate
goes up. Given the size of the present USPTO database, in the terabytes, it is not
seen possible how one can extract the proper information using electronic word
Many from their own experience have shown that only an unacceptably
small percentage of relevant Prior Art could be rodinely retrieved electronically,
and what was recovered was most often NOT the best, and NOT the most pertinent
Prior Art. If one considers a closed stack paper search file where nothing
leaves, then one has absolute file integrity, and if properly Classified it then
becomes easy for any person who can readily read English to find ALL the
pertinent Prior Art in a particular subject area. A complete search and
consideration of ALL the relevant documents - absolutely necessary to be in
compliance with Graham vs. John Deere - then becomes available, and doable, and
is available to the ordinary person without any special training.
It has been said that the paper files are not necessary because: ...It [all data
and information] is on the Internet.
This is simply not true. While some data is available on the Internet, or other
electronic data bases, much if not most of it is certainly not. Many journals have
only recently kept their data electronically. The journal Nature, for instance, only
goes back to January, 1996; the journal Science, October 1995. Electronic and print
versions routinely differ, sometimes significantly. And from Blair et al. and the
followup studies confirming their results, even if it was true that everything is on
the Net, no one would be able to locate the requisite information.
The assertion is also made, page 17056, second column, of the FR, that these
electronic search systems are:
...user friendly ...
yet admits that considerable training is necessary for their use, at page 17056,
third column, and at page 17059, first column of the FR. The Notice fbrther states,
as at page 17057, first column, that:
There are a steady new stream of customers who use the facilities for a very
limited time and for purposes of a fairly narrow scope. There are approximately
300 new users every month.
Yet due to the extremely arcane, nonintuitive, and difficult interface used
for these electronic search systems, this system is de facto INACCESSIBLE
TO THE ORDINARY PERSON. Even after training, the new user (and even
the experienced user) will not know if: he has formulated his question
properly; if the computer is down or dead; if all of the data is available to the
computer; or if all of the data hits will be available to him. He cannot look at the
complete set of data, even assuming arguendo that the computer search system
has all the relevant data at that moment at hand; he must pare down his search to a
few documents to be displayed electronically.
Thus, if the classified paper search file is destroyed, then the individual
Inventor will be forever precluded from searching his own invention. Destruction
of the classified paper search file would form an undue, permanent, and
irreversible hardship for the small and medium sized inventive entities.
The Register hrther states that electronic searching and the classified paper
search files are equivalent. However, they are not.
The electronic display of data is still far inferior to that of paper and no
display system exists which can duplicate the resolution of paper. The USPTO uses
a minimal display raster of 300 dots/inch [D.P.I.] to electronically store and
display data; in many cases, involving subscripts and superscripts in mathematical
and chemical representations, this is insufficient. The printed paper patent will be
perfectly clear; the electronic facsimile of the patent will not be and will often be
ambiguous. Two 81/2 x 11 inch sheets of paper present a display requirement of
5100 x 3300 pixels, an electronic display which does not currently exist. To be
able to view a screen for extended periods of time to be comparable to paper
requires at least a 100Hz refresh rate. So one has a 16,830,000 pixel requirement
with a 1.68Ghz data rate generator requirement, just to be able to reproduce
electronically the substandard 300D.P.I. of the current electronic storage. NO
SUCH ELECTRONIC DISPLAY AND NO SUCH 1.68GHZ DATA
GENERATION CURRENTLY EXITS. And this is for the substandard 300 DPI
standard currently in use.
The Office has admitted that the labor costs for their Millennium Agreement
with the examining corps which calls for the destruction of the examiner paper
files will cost them from between $20 million and $30 million per year in
increased costs. Yet the examiner paper file system, from an independent study,
only costs the Office about $1.3 million a year - about 0.1% of current budget -,
and the Public Search Room files are almost self supporting.
So why would someone spend $30 million per year to save $1.3 million?
As a Performance Based Organization, the Office is supposed to be run like a
business. This makes no discernable economic or business sense.
Recently a senior Patent Office employee told me: As soon as they get rid
of the paper [files], then theyre shipping examination overseas. The April, 2002,
edition of POPA News [Vol. 02, No. 2] also has the headline: USPTO Moves
Forward on Contracting Out Patent Examination.
Is this the reason why the paper files - which have served the public and the
Patent Office so well for two hundred years - have to be destroyed? So that
examination can be contracted out overseas? What will happen when the Russian
basic-bottle patent is re-introduced and repatented in this country? What will
happen when the Australian basic-wheel patent is re-patented in this country?
What recourse will the public and every American corporation have, but to submit
to any sort of foreign green-mail when the only resource capable of truly
protecting the public - the existing classified paper search files - are destroyed?
The destruction of the existing classified paper search files thus have no
economic benefit, and would severely impact the public and every American
corporation by essentially gutting this most important economic line of defense
against bogus litigation, either foreign or domestic.
Thus it is sincerely asked that the Offices intention of destroying the
classified paper search files as proposed in the Register be reconsidered and
rescinded. Should the Office ever actually contract out examination overseas - and
such is specifically provided for in the A.I.P.A. - then the classified paper files
would be the only real defense of the American public against such foreign
predation. The Constitution would seem to require such a prudent safeguard as
keeping and maintaining the classified paper search files, an irreplaceable treasure,
especially when they routinely outperform the electronic search systems and cost
virtually nothing to the Office.
Please consider this email a request to speak at the public hearing on May
16, 2002 relating to the disposal of the public paper patent collection.
Calvin Van Sant
VanSant Patent Services
21 Fairmount Rd
Kirkwood, PA 17536
I will be speaking for myself as a public user of the USPTO facilities. The
focus of my remarks will be in support of an electronic search facility, but
stressing the importance of the USPTO being diligent in how the transition
Public Hearing Statement
The following are the comments of Calvin E. VanSant, a public user of the USPTO public search
VanSant Patent Services is a sole-proprietor business providing patent searching services to
individuals, corporations and law firms, principally in the mechanical arts.
I have been using the public search facilities of the USPTO for 10 years. Over those years, I have
spent over 3900 hours using the various iterations of the computer-based image systems.
It seems inevitable that the electronic searching tools will supplant the paper files. Accepting that
fact, my comments relate to making the electronic tools as robust and usable as possible before
removing the paper from public use.
The following items should be addressed prior to removing the paper:
Outside of a fire, or similar catastrophic event, the paper collection is always available. The
same cannot be said of the databases accessed from EAST and WEST. Every effort should
be made to insure system availability to the examiners during core working hours and to the
public searchers while the pubic search room is open. System upgrades and maintenance
should be conducted in a manner to minimize or eliminate disruptions to the examiners and
the public. System "up-time" is a commonly tracked statistic in most corporate information
technology departments. The PTO should benchmark database providers such as Denient,
Micropatent, Delphion, etc. with a goal of exceeding their system availability percentage.
To help keep the PTO accountable, a graph showing the "up-time" of the search systems
should be posted in the public search room and updated on regular basis (i.e. monthly).
The best way to win over the advocates of keeping the paper is to make the electronic
system better. Better, in this case, meaning more accurate. When a data entry or similar
error occurs, the correction only needs to be made in one place rather than attaching a
certificate of correction to every paper copy filed in the shoes. The "proposed plan for an
electronic public search facility" addressed this point, but is a bit confusing. The Federal
Register announcement states:
"Like paper files, errors can occur in electronic search systems. However,
mechanisms are in place for tracking, reporting and fixing errors that are made as a
result of internal processes"
However, a recent discussion thread on the Patent Information Users Group ( " More stinky
biblio" http://www.derwent.com/archive.piug/0018.html) would contradict the previous
statement. On January 9,2002, Stu Kaback of ExxonMobil made note of erroneous fields in
the bibliographic data of US 6,160,029. Larry Larson of the USPTO offered his expertise in
response to Mr. Kaback's observation.
"The simple fact is that the 6,160,029 bibliographic data in the full-text
database (and in PTO-distributed full-text data) *can't* be fixed, because
it is correct, in that it agrees with the issued patent as printed (see the
full-page image for confirmation.) Under present PTO processes and systems ,
it will stay that way forever. PTO electronic data, both in databases and in
bulk data on magnetic media, is not intended to be a collection of
absolutely correct information; rather, it is intended to be an accurate
rendering of PTO's legal publications.
PTO does not have any process or system which actually corrects errors in
published documents by either re-printing those documents or by correcting
electronic data (either in-house or as distributed to the public), no matter
how egregious the errors might be. PTO has only Certificates of Correction
and Reissue Patents as tools. Certificates of Correction become appended to
the full-page image database, and Reissues are added to the database as new
documents, but published documents containing errors correctly reflected in
PTO full-text can never actually be corrected. This obviously complicates
the automation of patent searches and makes problematic full reliance on
electronic, rather than paper, patent collections.
It would theoretically be possible to correct errors by reprinting corrected
patents with a new publication date (as distinguished fiom the issue date),
and then updating electronic data to the most recently published content
(akin to software version control), but PTO does not have such a process.
Larry Larson, USPTO "
Please excuse my naiveté, but not being able to correct known errors in the database seems
rather unbelievable. Mr. Larson indicated that a committee has been formed to address a
process for correcting errors. I would encourage the PTO to implement procedures to
correct database errors as soon as possible.
To this, I must add Mr. Kaback's response to Mr. Larson explanation.. .
I believe that my notoriety in the field of patent information is due in
substantial part to the fact that I am not accepting of unacceptable
conditions. I think it is sad that the most imporiant patent office in the
world doesn't seem to comprehend how seriously wrong it is for it to
disseminate faulty information without providing a method to correct
errors. Everyone makes errors, every system makes errors. Other
documentation systems that I deal with have appropriate systems which
permit them, when such errors are pointed out, to correct them--they take
seriously a responsibility to provide correct information to their users.
Sadly, the USPTO still seems to consider itself just an organization for
issuing patents; any documentation that results apparently isn't worth
worrying about too much.
I said sadly--but sad is far too mild a term. I challenge the USPTO to
take seriously the responsibility of producing an archive of information
that is as accurate as possible--and that can be corrected when the
inevitable errors creep in.
A brief final note on an issue related to the electronic versus paper debate - Reclassification
One of the principle services I provide to the corporations who support the PTO through application
and maintenance fees is infringement or clearance searches. Before a manufacturer produces and
sells a product, they need a reasonable assurance that they will not be infringing upon another
company or individuals patent rights.
The ability to review the pertinent patent art in an efficient manner is vital to my clients. The
USPTO's decision to reduce resources in the area of reclassification has caused and inordinate
increase in the time it takes to complete a clearance search. KEYWORD SEARCHING CANNOT
REPLACE A CLASSIFICATION SEARCH. The information imparted by a human being placing
a patent in an appropriate subclass adds tremendous return-on-investment to the many users who
need a reasonable search set to prevent infringement of another's patent rights. Using VanSant
Patent Services as an example, between 1998 and 2001 the hours required to complete an
infringement search have increased by 32% (search set in excess of 1500 patents are common).
With the addition of pre-grant publications, and the increased number of patents issued, the problem
of subclasses with too many patents is accelerating. Please direct appropriate resources to
reclassification projects in areas where emerging technology has caused outdated classification
areas to explode in size.
In summary, make the electronic systems dependable, make them accurate, and use human
intelligence, applied by way of reclassification, to make the systems more efficient.
Thank you for your time.
Please consider this a request for the National
Intellectual Property Researchers Association (NIPRA)
to speak at the public hearing scheduled for May 16,
2002. The identity of the individual to speak on
NIPRA's behalf and our written comments will follow.
Robert B. Weir
Public Hearing Statement
The United States Patent and Trademark Office (USPTO) is proposing to eliminate the paper
patent and trademark registration collections fiom its public search facilities and replace them
with electronic information collections. The National Intellectual Property Researchers
Association (NIPRA) is opposed to such a transition at this time.
NIPRA is a not-for-profit organization comprised of intellectual property attorneys, agents,
researchers and inventors dedicated to the maintenance and improvement of the
United States Patent and Trademark Office with particular emphasis on improving the ability of
its members to access patent and trademark information so that they might effectively support the
Intellectual Property Community.
Upon review of the current state of the USPTO electronic systems, NIPRA is convinced that the
USPTO is not yet ready to transition to an exclusively electronic search environment. Although
the automated search systems have been a welcome adjunct to fill in the gaps and errors that have
crept into the paper systems and NIPRA supports the development of a superior automated
search system that could replace the paper collections, at present such a system does not exist.
USPTO and recent independent studies have confirmed that the USPTO electronic search
systems are not mature and reliable and although they may provide equivalent functionality to
the paper collections, they do not provide the more important criteria, equivalent results.
The USPTO argues that a fully electronic search system will result in cost savings and that the
paper classified files are incomplete by nature of the format. There may be missing or misfiled
documents, potentially impacted search results which rely only on the paper classified files. By
the same token, the USPTOs proposal will force reliance on electronic search systems such as
EAST, WEST, X-Search and TESS, systems that have been demonstrated to have numerous
software issues and similar data quality problems. A recent NlPRA survey demonstrated a 52%
error rate in the X-Search and TESS design code fields alone. (See Exhibit A) Those results are
buttressed by an August 2001 assessment of that same data by PriceWaterhouseCoopers (PWC),
the USPTOs internal consultant. The PWC assessment indicated an 46% error rate in the initial
data entered by USPTO contractors and a 36% error rate in the data uploaded to the electronic
search systems subsequent to quality review by USPTO employees. (See Exhibit B) Similarly,
test queries conducted on EAST indicate system problems such as a.) in excess of a one hundred
thousand patents issued since 1971 that are not text searchable b.) numerous reclassification
efforts that have not been entered into the database despite reclassification of the paper collection
c.) identical search queries returning different results d.) discrepancies between the number of
patents filed in a particular subclass in the paper collection and in the electronic database.
The inadequacy of the electronic search systems and the USPTOs development practices has
been well documented for more than twenty years. Despite consistent negative reports by the
Government Accounting Office (GAO) and the Department of Commerces Office of the
Inspector General, the USPTO has not learned from their previous mistakes. Since as early as
1979 the GAO has raised concerns about the effectiveness of electronic searching and noted that
the USPTO has a difficult time defining quality measures, electronic systems development costs
and proper systems specifications. These findings are repeated annually through the mid 1990s
in a series of GAO reports lambasting the agency for poor systems development, extravagant
costs and questionable contractor agreements. Similarly, as late as March 2001 the Office of the
Inspector General (OIG) had found weakness in the development of electronic search systems.
Specifically, the OIG identified the following problems with the development of the EAST
system in Inspection Report No. PSE-12679March 2001 (See Exhibit C.) 1.) ineffective
management and monitoring by the Commissioner for Patents and the Chief Information Officer,
2.) incomplete systems specifications, 3.) poor communication with the end-users 4.) poor
acceptance testing, and 4.) insufficient training in the use of the system.
Those official reports bear striking similarities to the comments received by the USPTO in
response to its initial Notice of Request for Comments on Development of a Plan
To Remove the Patent and Trademark Classified Paper Files From the Public Search Facilities
published in the Federal Register August 27,2001. The fifty comments received were
overwhelming negative and reiterated many of the same concerns expressed by the GAO and
OIG. (See Exhibit D)
There is no defense for these findings. The USPTO has however consistently engaged in a ritual
of obfuscation and blame shifting. The PWC assessment of the electronic trademark data was
never released to the public or, on information and belief, to the Chief Information Officer.
Although the findings of the assessment have great import to the public user, the USPTO, in
violation of OMB Circular A-130, suppressed the findings and did not inform the public about
these limitations in the information dissemination product so that they might be fully aware of
the quality and integrity of the information. Further, the USPTO has consistently responded to
GAO and OIG criticism with a line by line dissection of their reports, pointing out minute
semantic errors and ignoring the larger issues, ultimately placing blame for systems failures on
budget cuts, uncertain funding, contractor delays and Government Services Administration
interference. Similarly, the USPTO response to the overwhelming negative comments to its
August 27,2001 notice was to dismiss the substantive comments and questions regarding the
plan to eliminate the paper files as not germane to the issue. Given that the USPTO cunningly
asked for comments on the development of a plan to remove the files rather than comments on
whether the existent systems were adequate to proceed with the development of a such a plan,
such contempt for the public input is not surprising. Nevertheless, it is curious that of the fifty
Reponses received to the notice, not one had a single comment or question that the USPTO
considered germane to the issue.
Furthermore, the USPTO has published its Draft Data Quality Guidelines and appears to have
determined that the data in the electronic search systems is exempt from those quality guidelines.
We have contacted the USPTO Information Products Division for clarification on this issue and
are convinced by their response that they neither understand what the data in those systems is nor
how the data is entered. Our query with regard to what data was or was not exempt from the
Data Quality Guidelines resulted in the following response ten days in the making: The
databases or systems you mention contain a variety of data or information types, and according to
the guidelines it is possible that some of the information contained within a database or system
may be exempt, while other information is not and essentially review the OMB guidelines and
make the determination for yourself.
Lastly, although the Federal Register notice denotes what the USPTO shall not do under Section
4804(d)(2) of the AIPA, it does not underscore that the Office is mandated to maintain, for use
by the public, paper, microform, or electronic collections of United States patents, foreign patent
documents, and United States trademark registrations and that those records must be arranged
to permit search for and retrieval of information. It should be noted that the trademark
electronic search systems, principally X-Search and TESS do not fulfill that requirement as they
are composed of bibliographic and image data that has been keypunched or tagged by USPTO
employees and contractors and are not collections of the trademark registrations as issued. Given
the General Counsels fine interpretation of Section 4804(d)(2) in its determination that the
pending trademark applications were not subject to the Act, that same fine interpretation should
be used to analyze the electronic search systems compliance with Section 4804(d)(2).
In short, the USPTO has embarked on twenty year campaign of misinformation and
contemptuous disregard for GAO and OIG findings as well as public input with regard to its
electronic systems development. To date, the potential damage that could result from absolute
reliance on the USPTO electronic search systems has been mitigated by the existence of the
paper patent and trademark collections that serve as validation for the results generated by those
systems. Should the USPTO be allowed to proceed with the elimination of those records this
valuable function will also be eliminated, jeopardizing the validity of all research conducted at
the USPTO and resulting in serious damage to the United States intellectual property system and
economic harm for those who rely on the faulty USPTO electronic systems.
Given the USPTOs mandate to disseminate information it would seem reasonable that the
USPTO should strive to create and maintain the most accurate and complete record possible.
Despite the expenditure of untold millions of dollars, many of the problems identified by
previous GAO, OIG and independent reports still remain and have not been addressed by the
agency. Thus, it seems premature to eliminate the paper search file until such time as significant
improvement in the electronic systems data integrity can be verified.
NIPRA recommends that the USPTO immediately commission an independent study of the
automated search systems by an independent organization to ensure correction of the existing
data and creation of guidelines to correct the data flow and ensure future data quality. (See
Exhibit E) This study should consist of a side-by-side comparison of the electronic and paper
search systems until such time as the results of an exclusively electronic search are consistently
the equivalent of a combined electronic and paper collection search. Pending the results of that
study the agency must suspend all efforts to eliminate the paper patent and trademark collections.
The office is also urged to advise users of the automated search systems of their deficiencies in
accordance with OMB Circular A- 130
Very truly yours,
Robert B. Weir
Exhibits A B C D E
With respect to the May 16 hearing, please register the following individual
Randy Rabin, President, PatentArts, LLC
2101 Crystal Plaza Arc, PMB 400
Arlington, VA 22202
I will speak as a professional searcher on my own behalf,
and I wish to express an objection to the removal of the paper
Patent and Trademark Libraries, and for the improvement of the
Respectfully submitted ,
Public Hearing Statement
Good Morning. My name is Randy Rabin. I am president of PatentArts, LLC, and have
conducted patent research for companies and law firms for over 20 years. I am speaking
today on my own behalf. During this time, I have performed several thousand patent
searches in nearly every technology, with an emphasis on computer technology,
including software, electronics and communications. From 1995 through 1997 I served
on PTO public advisory committees regarding computer implementation, and most
recently, at PTOs request, participated in sessions to explore how to improve patent
search result quality.
I join my colleagues today in expressing as strongly as possible the necessity to retain the
paper based collection of patents until the computer system (which we refer to as EAST)
has proven itself as a dependable tool for accessing critical patent information. For
simplicity, I will restrict my comments to the patent side, but most of my remarks are also
appropriate for trademarks.
One might assume that those of us who support preservation of the paper collection are
perhaps not ready to move into the future. Quite the opposite is true. Many of us who
are most vocal on this issue also happen to be among the most computer knowledgeable -
not only are we skilled computer users, but much of our patent work involves inventions
based on computer technology itself. Many of us have been accessing online databases
since even before the first PC appeared. Part of my own history goes back to 1981, when
after 4 years of searching patents strictly in paper form, I proposed to then Commissioner
Mossinghof that we have terminals in the PSR that would enable us to access the
databases of Dialog and other services. The power of the computer for quickly accessing
data was obvious to anyone who sometimes spent many hours searching for a single
detail in hundreds of patents. But an expression was born in the 1980s: a computer
search is a good aid, but not a replacement for, a paper search. I had hoped that by now,
nearly 20 years later, a computer search would be good enough to replace a classified
As complex as the patent system is, it has a fundamental basis, and that is the act of
comparing. This act of comparing can happen many times in the life of a patent: the
inventor may use a collection of patents as a unique knowledge source in developing his
invention, the patent attorney will use related patents to focus the claims of a patent
application, and the examiner will conduct his own search and comparison in acting on
an application. Later, the patent may be the subject of a validity or infkingement search,
with even deeper comparisons. Every aspect of a patent, fkom prosecution through
litigation, is based on these comparisons with the written record. Obviously, whether the
record exists in paper or electronic form, the written record must be accurate, complete
and usable. Otherwise, every other aspect of the process is compromised. In the Federal
Register notice announcing this hearing, the electronic database was described many
times as mature and reliable. Every day, every one of us who uses the electronic
system is faced with its flaws in the form of missing or corrupt data. And faulty search
results sometimes lead to the issuance of faulty patents, at great cost to the parties
FLAWS IN THE TEXT FILE
When conducting a search using EAST, the number of patents that can be searched using
text input, whether it be a technical term, the name of an inventor or a company, or cited
references, is limited to those patents having a text file. Of the 6.8 million issued patents,
only 3.1 million, less than half, fall into that text searchable group, which spans the
period 1971 to date. Of that latter group, however, more than 103,000 are missing text
files and therefore are not retrievable using any terms other than patent number or
classification.2 Therefore, an examiner or searcher looking for patents using any words,
for example light amplification or halogenated biphenyls, would be limited to those
patents issued in the 3 1 year period of 197 1 to date, even though many patents for those
technologies exist prior to 1971. More tragic though, is that during that period of 1971 to
date, there is a 3% chance that the patent you need is missing a text file and therefore will
not befound. In addition, if you tried to find all of the patents assigned to, for example,
Merck Pharmaceuticals, or Dow Chemical, there is a significantly higher chance that
patents will be missing due to the inclusion of chemical symbols in the text (note Exhibit
A). PTO has been aware of this problem of missing data since at least 1992, but either
through neglect or choice, has not posted a warning notice to users, nor have the missing
files been restored in all that time. To the contrary, notices are routinely posted
announcing that the date range of patents available for text searching is 1971 to date,
giving the user a false sense of confidence that all data is present.
Another major problem is the complete inability to text search prior to 1971. Even small
private companies have managed to OCR older patents back to at least the early 1900s.
EAST still does not provide that ability. Since the appearance of an article in the New
York Times, I have received a number of calls fiom across the country fiom people
looking for patents to Thomas Edison, Nikola Tesla, Philo Farnsworth, and Chester
Carlson. One sought wartime patents to his father. One sought patents to her uncles
company that operated during the 1950s. Not one of these could be found on the
computer system due to its date range limitation (See Exhibit B).
FLAWS IN THE IMAGE FILE
A large number of patents have flaws within the image file, which of course contains text
as well as drawings. In Exhibit C are several patents found in the past few days during
the course of normal searching by one searcher. You will note drawings and text that are
little more than black blobs. In the paper file, this of course does not occur, but
sometimes a patent is missing. Since almost all patents are cross-referenced, the same
patent can be located in another subclass.
In the computer database, however, a patent is recorded only once, without a clean
copy to fall back on. Though a heroic effort is being made by one examiner on his own
time to locate and replace defective Design Patent scans, too many exist in the Utility
Patent database to correct in the near future. Despite the fact that examiners and searchers
routinely encounter these flaws in the course of our regular work, there is no reporting
system in place to allow this army of more than 3,500 to assist in the correction process.
A very significant problem in the study of patents, especially when many hundreds must
be reviewed, is the poor image quality of text and drawings displayed on a monitor. In the
PTO, 21 color CRTs are used to present black and white information. In Exhibit D,
photographs were taken of a screen display of a patent, with a corresponding paper image
of the patent taped to the screen for a side-by-side comparison. Reading a patent for
content and meaning on a computer monitor is a very significant problem: many searches
are aborted early due to visual fatigue, or patents are printed out in large numbers for
THE PAPER PATENT COLLECTION
The classified patent collection has been demonized as wasteful of space, money and
resources. We have already witnessed the merging of the two separate collections that
nearly cuts the space requirement and other costs in half. The cost of maintaining one
complete classified library has been conservatively estimated at between $5-7 million per
year, including space and utilities, copies, and staff (compare this with the computer
system, which has cost well over $1 billion so far, with annual expenditures exceeding
$100 million, and $239 million in year 2001 alone).
The space requirements to house US and Foreign Patents, plus Literature, has been
estimated at 74,000 square feet. For visualization purposes, this is equivalent to the floor
space of an average K-Mart department store. Of course, a multi-story building of
equivalent floor space is preferable, and would serve the needs of examiners and the
public alike. In short, the classified paper system is the very cheapest, most dependable,
user friendly, hacker-proof, already-existing backup system that could be devised or
procured, for use by examiners and public alike. And it is already in service. But the
paper collection is far more than a backup system: to experienced users, the paper
collection has no equal as a research tool, allowing careful, in-depth study of a subject,
and permitting the reader to far surpass the 1-2 hour limit of viewing a computer monitor.
To be useful, the collection must be located in the immediate vicinity of the USPTO.
It is impossible to discuss any aspect of the search process without mentioning
classification. If there is one very critical tool valued by examiner and searcher alike, it is
the Classification System. It deserves its own patent, if not the Pulitzer. Without it,
searching paper patents would be impossible. However, it is just as essential for
conducting computer searches, especially in light of the failings of text searching. The
classification system has been one of the dependable tools we have had as examiners and
searchers. Yet, almost fkom the day the computer system was installed, classification has
been slowly neglected until its reliability is in question. The importance of classification
must be reevaluated and its decline turned around before it is too late.
In summation, it is hard to believe, given the flaws pointed out in this and other hearings,
that the computer system can be considered robust and mature as contended in the
Federal Register notice. Such confidence is little more than wishful thinking. It is
apparent that any decision affecting the fate of the paper libraries is being rushed by a
plan to move into new buildings that are too small for either the paper collection or the
examiner corps. 4
I want to conclude by making some observations that will not be popular, but must be
acknowledged. There are four primary groups affected by this pending decision to
remove all paper records: 1)Attomeys and Agents; 2) Examiners; 3) Professional
Searchers; and 4) Inventors and their Companies.
In my preparation for this hearing, I assumed that attorneys and agents would be eager to
participate, seeking to guarantee that the system would be intact, and the tools would be
in place to ensure the integrity of the patent system. Even the most adventurous attorneys
I know balked at the idea of opposing PTOs actions, and being so identified. Two
attorneys, one of them a former examiner, told me the following:
1) My clients want patents, not necessarily valid patents, and the more the better. If a
searcher or examiner does a good job at finding prior art, my clients arent happy,
and neither am I.
2) Attorneys make money whether we represent the inventor or an infringer. The
more inefficient the system is, the better it is for us.
3) If a patent application is rejected, or a patent is found invalid, the attorney is not
considered liable, nor is the PTO, at any level.
4) Companies can afford lengthy litigation to prevail over smaller entities.
The second group, Examiners, unfortunately are forbidden by contract to speak on the
subject, but privately have expressed strong opposition to the removal of their libraries,
and the need to totally depend on EAST, despite having received a pay raise in exchange
for their libraries. In official PTO surveys, examiners have often expressed the
importance of a job well done in evaluating their own sense of job satisfaction, and the
frustration they feel as their efforts are blocked or undermined by office policy.
The third group, professional searchers, has been the most vocal group on the issue we
are discussing today because we are the closest to it. We feel an obligation to speak for
those who cannot. The time we have spent on this issue, including preparation for this
hearing, is considerable, and at our own expense. We are simply trylng to do the best job
for our clients that we can.
So that leaves the individual inventors, the corporate inventors and the companies who
have often invested millions in their inventions and in their patents. I doubt more than a
handful of those individuals know what is happening here today. These are the people
who should be speaking today, the people who have invested their time and money in the
creation of their inventions, who fund this system, and who rely on this system to work
One final note - it was difficult to prepare for this hearing today, because of the widely
held, and often experienced feeling that PTO management - to paraphrase Professor
Higgins in My Fair Lady - will listen very nicely, and then do precisely what they were
going to do anyway. But I hope it isnt true.
Thank you for the opportunity to speak.
1. See the Website: bustpatents.com for a list of patent lawsuits and their costs.
2. The following search statement is used: text available.ti. AND @py>=1971.
3. The merits of the classification system were unanimously extolled in a USPTO
public hearing in 1996.
4. Many of the points I have made today are further expanded in an article I wrote in
the May, 2002 issue of Intellectual Property Today, pp.60-63, Exhibit E.
Exhibits A B C D E
I hereby request the opportunity t speak at the May 16th hearing.
Name: Harold L. Novick
Nath & Associates PLLC
1030 15th Street, NW, 6F
Tel. (202) 775-8383
Fax. (202) 775-8396
email: firstname.lastname@example.org <mailto:email@example.com>
Organization represented: None
Contact Information: see above
Interest: The affect that the removal of the paper files will have on
the integrity of the patent data base; the current status of the electronic
data base; the effect on patent classification; the effect on those arts
where a word search is difficult.
Harold L. Novick
Kindly add my name to the list of those persons wishing to
speak at the public hearings (set forth in the above-identified notice)
Glenn E. Wise
Please add my name to the list of speakers.
Very truly yours,
Terrence L. B. Brown
Reg. No. 32,685
Shlesinger, Arkwright& Garvey, LLP
3000 S. Eads ST
Arlington, VA 22202
I was recently informed that few reservations to speak at the public hearing had been made
by the 4/30/02 deadline. I would like the opportunity to speak at that hearing. If a
guaranteed spot can not be made, please let me know if there is any other opportunities
available to address the subject.
Reservation in the name of R. Lee Grantham.
Public Hearing Statement
Im Lee Grantham. I work for a local 35 attorney firm. I manage the firms search
department. Ive been searching for 13 years.
Weve been asked to comment today on a plan to create an electronic search
facility which means the elimination the PTO paper files. PTO argues that functional
equivalency has been reached regarding electronic patent storage and retrieval. I have to
agree. The new search tools are congruent with the necessary methods of the information
age. The paper based search system has been made better. I expect additional features to
be added regularly thereby further enhancing our work effort.
The question at this point, today, regards expectations that the work product
derived from using electronic techniques in comparison to the paper files will produce
comparable results. With electronic capabiity we expect efficiency, we expect
thoroughness and, especially, we expect to identify germane prior art. We expect the new
methods and techniques to enable the user to produce a work product of equal quality to
that provided by the system now being replaced.
The new search tool is powerful. The precision of keyword searching directs the
user to relevant prior art. One can check cited references with the snap of a finger. You
can even insert a search term in a particular document after youve identified the
document as potentially pertinent. So, the introduction of electronic techniques must
result in an improved system to find prior art, right? I dont think so. (The key word is
must.) The expectations are not guaranteed.
The fruits of technological progress do not result without a vision and a reasonable
understanding of current systemic dynamics. The decision makers must understand that
the cross impact effects of various players, even those simply engaging in their daily tasks,
can have totally unanticipated effects. Current trends must be identified and carried to
their logical conclusion so the decision maker can better evaluate the steps necessary to
reach expected outcomes. The assumption, it seems, is that we have a fine search system
that works so contributing more capacity or more searchg power or more efficiency will
results in an improved outcome. Its as if PT0 feels there is black box where you put in
something good, let it mix around a little and out pops progress - new and improved.
That is not how success works. The goal must be defined and attentive steps taken to
reach it. The technology analysts long ago recognized that every time a new technology is
introduced into a social system there are unintended consequences.
Thats what I am going to address today -- unintended consequences.
There are two different types of searchers. There is the public sector searcher and
the private sector searcher. Even though each access the same source material each has
different objectives and each faces different constraints.
The public sector searcher is considered an expert due to their narrow focus. They
are expected to learn the art within their areas. They are expected to search quickly and
over time, as they learn the art, to search even quicker. Historically, this has been a fact, it
has worked. Presumably, the public sector searcher is expected to be thorough despite the
fact that they work under strict time constraints that preclude extensive searching.
Searching is only one responsibilty that this group has. They, the public searchers, are the
The private sector searcher is a generalist. We work in all the art units. We
always have more time to search. We regularly spend 8-12 hours doing patentability
studies. We sometimes rely on primary examiners to provide a search field thereby the
saving the generalist some time.
The reality of the matter is that the examining corp heavily influence the overall
makeup of the search system. This is true even though the examiners search results are
the product of minimal searching time. As the shear number of patents has exploded
searching hap become more entailed.
In older to achieve a proficient system it is imperative to evaluate current systemic
trends that can impede attainment of the goal. The goal in this case is a well organized,
accessible patent searching facility. Because the examining corp has great influence on the
system it is necessary to look at trends emanating from that goup. I want to be clear that
what I am stating here are the observations of a private sector searcher but I believe that
they are representative of intrinsic trends that do not bode well for a proficient search
system. My point is not to attack but only to illuminate.
The following encounters have all happen to me in the past year. The experiences
are replicated fairy often by other private searchers. For instance:
1. Showing a disclosure to an examiner and being told (Im not sure where to
look but I would just keyword it;
2. Asking an examiner where the paper files are for their class and are told "I
dont know, we dont search paper anymore. When you mention that youre looking for
the foreign shoes the response -- the foreign is on the database;
3. Take a disclosure that acknowledges the use of known old technology
(although the application in the environment was new), see two examiners, get two
different search fields, and get the same puzzlement/surprise from each along with the
same comment that the old technology was different; only to find out latter that the
technology was not obscure and that there were about 2 dozen examples patented
between 1900 and 1920. The art was not in any of the subclasses provided by either
examiner. The correct subclass was eventually identified using electronic search techniques
but remember I have ample time to look.
4. We are starting to see simple patents issuing that have pertinent issued prior art
not identified by the examiner. We are finding invalidating prior art that predates 1971 or
uses different lexicography.
5. I am subject to comments from attorneys that office actions are being issued
that cite patents that have little to do with the invention but do contain appropriate
The five examples show a number of things. I believe that importantly they show a
disconnect regarding the existence of old art. Old art is buried in the subclasses. More to
the point there is a disconnect regarding prior art that is not key word searchable. The
general trend, therefore, indicates a reluctance to manually search class/subclass even
though this is doable electronically. The result is diminished reliance on the classification
The elimination of the paper files removes the need to physically put yourself in the
art. The organization on which the paper system was founded is now rendered out of
hand, out of sight, and out of mind. This development is understandable and in some
cases it was even predictable. Subclasses in many mechanical fields are simply growing
too large, which is another trend, The problem is acute in the mechanical arts because the
mechanical arts do not lend themselves to keyword searching and one must manually
Brigid Quinn recently commented that the paper had no intrinsic value. On this
matter she was wrong. The value of the paper files resides in the fact that they provided a
well organized method of finding prior art. The foundation of the paper system must be
continued. If the classification system is allowed to atrophy the goal of improved search
capability will be missed; it might not even be up to the demands of the information age.
A system dependent on text searching will not be efficient and it will not be
comprehensive. Technology historians realized long ago that with the introduction of new
technology something old is lost. In this case it appear that we are losing the classitlcation
system. The private Sector requires a system that works especially in view of burgeoning
I am an independent patent attorney and an independent patent searcher. My direct clients
include IBM Corp. and Whirlpool Corp., and my indirect clients include Toyota Motor Co. and
Honda Giken KK (whom I work for through the patent firm NGB Corp. in Tokyo, Japan).
I would like to comment on the U.S.P.T.O.'s misleading statement regarding "mature and
reliable electronic search systems" that are available in the public search facilities which, in the
U.S.P.T.O.'s view, renders the paper collections superfluous.
The electronic search systems in the public search facilities are neither mature nor reliable.
The systems are constantly being changed in an effort to make them work properly, and are
constantly plagued with bugs and instabilities.
For the sake of brevity, I will limit myself to issues which have arisen in the last two
weeks. I will also limit myself to a discussion of the EAST system, though in terms of fiascos,
I could just as easily address WEST (described by the Washington Post as "the computer system
from hell"), CASSlS2 (a classic example of replacing an excellent CD-ROM system with a newer
and virtually useless DVD-ROM system) or the new PALM system. In each instance, the current
PTO system is less stable and less functional in significant ways than the system which it
IN THE LAST TWO WEEKS OF EAST:
- As verified by PTO staff, as of Tuesday, May 14th, the bibliographic "classification
data" for pre-1971 U.S. Patents was no longer available on EAST. Classification data is vital
to the searching of patents. It was postulated that the loss of the pre-1971 cIassification data in
EAST may have been related to the attempted incorporation of the long overdue [pre-1971]
"OCR" file into WEST.
- The entire EAST system in the public search facility crashed on Friday, May 10th at
9:10 am. The crash was called in by PTO staff, and the PTO staff was assured that it would only
take "15 to 20 minutes to get the service back up". Handwritten signs were posted by the PTO
staff that EAST was experiencing "global" networking problems. These signs were still posted
when I left the search facility around noon.
- During the week of May 6th to May 10th, the following seach engine error in EAST
was reported and verified by PTO staff: when a search for patents issued to U.S. inventors is
conducted in the USPAT database with thequeries "us.inco." or "us.incc.", EAST indicates that
a total of three (3) patents satisfy the criteria! ( Actually, upon follow-up by PTO staff, it appears
that one of the three patents returned by EAST was invented by persons from the old U.S.S.R.
in the 1970s!)
- During the week from May 6th to May 10th, the following search engine error in EAST
was reported and verified by PTO staff: when a search by document number (e.g.
"3900004.pn.") is conducted while in the USPAT and EPO files simultaneously, and when the
number corresponds to both a U.S. Patent and a German Application, once you view the U.S.
Patent, if you try to retrieve the image for the German Application, EAST instead returns the
image of the U.S. Patent while retrieving the bibliographic information for the German
Application. (This seems to be an temporary networking or caching glitch, and happens most
of the times on most UPWS terminals, including staff terminals.)
- A large proportion (greater than 50%) of Japanese patent images were unavailable for
display on EAST on Friday, May 3rd and Monday, May 6th.
- The basic graphical user interface for EAST was changed in the last two weeks. The
change incorporated no perceptible improvements, but it did take several steps backwards.
Previously, the patent list screen in the upper left-hand corner would display both the patent
number and the patent title; seeing the title in this screen facilitates browsing. Now the default
patent list screen displays the patent number, kind code, database source, and issue date. The
"kind code" and "database source" are much Iess informative than the title, and users must now
change the default patent list screen manually to show patent titles.
- When flipping pages, EAST hangs now much more than it did when it was
implemented. When it was implemented in 2000, EAST could flip pages at under a second a
page semi-consistently. Now, when "autoflip" is set, minimum page flip times that can be
achieved by the system are often 1.8 seconds per page, or more. This is not an acceptable flip
rate for browsing. When paper patents are flipped, human dexterity can achieve flip rates of
under 0.5 seconds per page. During "picture" searches, it is often necessary to flip though many
thousands of drawing pages looking for a particular design, and a 1.8 second flip rate achieved
by EAST literally adds hours to the required searching time.
I am not sure what would lead U.S.P.T.O. oficials to allege that EAST is a "mature and
reliable" system. I do hope it is out of ignorance (and not out of a desire to deceive) that this
statement has been made. I would, however, caution U.S.P.T.O officials not to make such a
statement to the Congress.
David Testardi, Reg. 33,639, VSB No 38608
2101 Crystal Plaza Arcade, PMB #189
Arlington, VA 22202
I note that the PTO proposes to eliminate the paper patent and trademark registration
collections from its search facilities, and that a hearing thereon has been scheduled for May
I have no intention to speak at this hearing. However, I am writing to endorse the
For nearly 28 years, I have opposed the elimination of the paper trademark search
records until the reliability of the electronic files had been demonstrated. I believe that
time has now come.
For the last ten (1 0) years, I have relied primarily on the search reports of
Thomson and Thomson and Corsearch, which, in turn, are based on the electronic data
received from the Trademark Office. With minor exceptions, the electronic records have
proven to be reliable. As matters now stand, they are more reliable than the paper
Having worked with the paper trademark search records in the Trademark Office
for over 44 years, it is with some sadness that I will see their demise. On the other hand, I
applaud the Trademark Office for waiting until its electronic records at least were as
accurate as the paper records before eliminating the paper records.
Very truly yours,
Henry W. Leeds
In the Federal Register Notice dated April 9,2002, the PTO requested public comments on its
development of a plan to eliminate the paper patent and trademark registration collections from
its public search facilities and transition to electronic patent and trademark information
collections. The comments of the Section of Intellectual Property Law of the American Bar
Association on this proposal follow.
I. PATENT FILES
A. MAINTAIN THE PUBLIC SEARCH ROOM
The American Bar Association Section of Intellectual Property Law (IPL Section) continues to
oppose, at least until an equivalent or better electronic system is demonstrated, the removal of
the paper patent files from the PTO facilities, on the grounds that the paper files continue to be
an important tool for searching patents.
B. MINIMIZE DELETERIOUS EFFECTS OF PAPER FILE REMOVAL
The IPL section recommends that the following areas be addressed to eliminate deleterious
effects in any proposal by the PTO to remove the paper files from the public search room:
Printing - Since copying will not be possible, printing is an issue that should be discussed.
Printouts should be easily retrieved by searchers and the cost of printing should be reasonable
and reflect the actual cost of producing the printed output.
Adequate Number of Work Stations -Searchers should not have to wait for an available
workstation. Therefore, there should be an adequate number of functional work stations such
that searchers rarely need to wait for one to become available.
Desk Space - The library currently provides plenty of desk space to public searchers who need a
place to spread out their work. Digital work stations should be provided with ample desk space
Aesthetics - The library currently has a unique aesthetic quality that echoes the aesthetics of
great historical libraries. Aesthetics was not mentioned in plan proposal, and we urge the PTO to
consider aesthetics an important aspect of the new electronic public search facility.
Processor Time / Search speed - With paper files, you can flip through paper patents very
rapidly. The PTO should ensure that the electronic search systems permit searching at a rate
comparable to (or better than) searching manually.
Classification Index - Paper files were maintained in order of classification. There should be
no move by the PTO to abandon the classification system, as it remains an effective search tool.
Foreign Art Collections - With the loss of Examiners art collections in shoes, free public
access to foreign art collections at the PTO is eliminated. The IPL encourages the PTO to
explore relationships with WIPO, EPO, P O and other national patent offices to establish
comprehensive searching of foreign art at the PTO.
Assistance - With paper searches, the only assistance necessary was in pointing the person to the
patents they wanted to view. The electronic systems are much more complex and adequate
assistance should be made available to members of the public learning to use the electronic
systems for the first time. The software should be user friendly and the interface should be as
intuitive as possible. However, the search tools should also be capable of powerful searches as
would be expected of professional searchers.
Hours of operation - We ask that the hours of operation not be reduced.
Above all, the IPL section encourages the PTO to ensure the highest integrity and quality for the
electronic records. It is crucial that the electronic records be maintained up to date, and in
searchable form. Furthermore, it is important that there be easy access to the electronic records,
with easy and quick access to non-text images, including drawings and chemical structures.
Without the ability to review these parts of the patent, any electronic access is insuffkient and
cannot replace the paper files. The search engines must allow for use of wildcards, truncators,
expressions, proximity operators and the (and) and (or) functions, as well as classified searches.
II. TRADEMARK FILES
The IPL Section of the ABA does not object to the PTO proposal to remove paper files relating
to trademarks from its public search facilities. Our members report that most practitioners are
already using on-line services, either the PTO site or commercial services, primarily if not exclusively. We
also note that the PTO stopped filing information on new trademark applications in these files last
November. As a result, even now it is not possible to conduct a complete trademark search in the paper
files. We therefore do not object to the removal of these paper files, and we believe that PTO efforts
should now be directed toward monitoring and maintaining the quality of its automated trademark files.
III. OVERALL PLAN OF ACCESS
The IPL section suggests that the plan for any new electronic search facility be a part of an
overall plan to provide free or low-cost public access to patent and trademark documents. Such a
plan should include available free or low-cost parking, improved web-based searching, and
improved facilities at the regional Patent and Trademark Depository Libraries. With regard to
the latter, the IPL section suggests and encourages the PTO to provide EASTMST access at
the regional Patent and Trademark Depository Libraries. While the section recognizes that
search speed at the depository libraries may be limited by the network connection, such access
would be very valuable to businesses and individuals alike.
Thank you for the opportunity to comment on issues relating to converting the Patent
and Trademark Public Search Rooms to all-electronic facilities, I am a frequent, almost daily,
user of the Public Search facilities. The comments are my opinions, not those of any
The Patent and Trademark Office has made impressive strides in making patent and
trademark information available to public users in electronic form. More particularly, the full-
text search capabilities, as initially available using the "MESSENGER" software, significantly
enhanced access to U.S. patent information.
Nevertheless, I believe the proposal to convert to all-electronic access is premature.
Necessary attributes of an all-electronic system include high system reliability, equivalence to
classified searches in the paper files, full-text capability and workstations suitable for use for
extended periods or time. For reasons stated below, the electronic prototypes do not yet meet
The electronic systems are subject to down time, in which the terminals are dark;
crashes when data or searches are lost and system failures, particularly with respect to the
group printing function.
Experienced searchers have reported significant mismatches between the contents of
classified paper subclasses and the same subclasses, accessed by EAST or WEST. Generally,
the electronic subclasses contain fewer references than the corresponding paper subclasses.
The electronically-available subclasses thus provide a less comprehensive search than the
corresponding paper files.
The full-text search capability, now utilizing the BRS software, complements the
classified search capability, but does not replace the classified sesearch capability because:
1. an OCR electronic file of patent documents between 1970 and about 1974
is not available to public users of the search facilities;
2. no pre-1970 patent document is searchable in full-text form to users of
EAST or WEST;
3. chemical formulae are not searchable in EAST or WEST; and
4. full-text data base is not indexed, so that the output of a fill-text search
is inherently limited by thc language of the query input.
The full-text electronic search capabilily is necessarily limited by the fact that the
applicant is his/her own lexicographer. The full-text and classified search capabilities are thus
complementary, rather than coextensive, in their scope.
Regardless of the form in which the search files are available, it is imperative to
maintain the US. Classification system (UCLA) and constantly update the classification system
to accommodate the need for manageable searches of increasingly large and active subclasses
(over 1000 documents). Search strategies proposing to obviate the need for reclassification of
excessively large subclasses by using the Boolean "and" to combine a subclass(es) with full-
text concept(s) can, and do, fail because of unavailability of unscanned or unavailable pre-1975
art in the full-text side of the data base and the creativity with which applicants are permitted
to define their inventions.
Concept- based classification, as exemplified by UCL A, is independent of terminology.
Concept-based classification allows the applicant/patentee to describe the invention in
terminology of his/her choosing and provides a framework in which the searcher can find
pertinent references, regardless of how the invention is described. The importance of
maintaining the UCLA and reclassifying overly large subclasses into manageable subclasses
should not be underestimated.
There is legitimate concern about the quality of the images displayed, particularly the
quality of text images. People who normally search by viewing drawings are generally more
favorably disposed toward the available image quality than those who need to read mainly
textual material. Workstation users require higher image quality than presently available in
order to be able to search for several hours each day without eyestrain. Improvemats in the
monitors with respect to glare/angle of viewing etc. are almost certainly needed for the
monitors to become acceptable replacements for viewing of paper document copies for hours
at a time.
Other attributes, required for acceptable workstations, were discussed at a focus session
held on July 17, 2000. Most of the '"requirements" are merely elements of good design,
minimal crileria if the PTO is to provide world-class electronic search rooms.
Despite the progress made in developing the electronic workstations, the workstations
are not yet an adequate replacement for searching of the paper files. Therefore, at least highly
used paper search files should continue to be available for use at the PTO for the foreseeable
The existing paper files are an irreplaceable, tried and true, source of technological
information, a source which should not be put into dead storage or destroyed unless and until
electronic information products have been demonstrated to be equivalent to or better than the
paper search files. 'I'herefore, the PTO should act conservatively in reducing its reIiance on
the paper files.
Thank you for your interest in this matter.
Charlotte M. Kraebel
Reg. No. 25,784
In response to the request for comments published in the April 9, 2002 Federal Register, Hahn
Loeser & Parks, LLP opposes the proposed plan to eliminate the paper patent and trademark
collections fiom the public search facilities and to transition to electronic only patent and
trademark information collections.
Hahn Loeser & Parks is a business and business litigation firm with more than 95 attorneys in
three offices. Hahn Loeser & Parks has been active in patent and trademark matters since 1947
through the firm of Oldham & Oldham Co. LPA, which joined Hahn Loeser & Parks in 2001.
Our clients include individual inventors, Fortune 500 corporations and international clients.
Client industries include organic, inorganic and polymer chemistry; medical and life sciences;
electromechanical systems; mechanical systems; and electronics.
A major limitation to the electronic searching is that the inventor can be his own lexicographer.
As a result, the use of keyword searching with words that may be commonly used by one skilled
in the art will often not result in a complete or comprehensive search. For example, valve may
be called a gate, or stopcock, or be described in many different ways, particularly if the
application originated in a non-English speaking country. If a critical word appears once in a
document and it is spelled incorrectly, the document may be missed completely. Another
limitation with electronic searching is that invention can precede the technology. For example,
lasers were invented several years before the acronym laser was coined to identify the
invention. Thus locating a critical document can be totally dependent on the skill, or luck, of the
searcher to pick a comprehensive list of keywords. Often, a document located using a computer
text search will, after studying the patent; turn out to be irrelevant because the keyword was
listed as background material in the patent. While keyword searching can find many relevant
patents, there is no substitute for a manual search. For many inventions, a picture is necessary to
convey the invention. Therefore, flipping or browsing through the figures can quickly determine
if a reference is relevant or not.
Currently, text searching is only available for patents issued since 1971. Therefore, electronic
searching for patents prior to 1971 must be done by classification and subsequent downloading
and viewing of each patent. The electronic systems do not provide the same ease and speed as
the paper collection for the ability to quickly flip through the front pages and to readily switch to
reading the relevant sections of a reference of interest. Even with newer faster computer
equipment, viewing screen after screen of patent images is simply not as fast or efficient as
scanning through paper pages.
Trademark records are such that, by their very unique and sometimes perverse nature, as a whole
they often defy standard rules of boolean logic or even constructed design codes and thus
cost effective electronic searching. As to word mark searching, during a search the paper files
educate the searcher as to various aspects of the word(s) in question more effectively than do
electronic systems by providing the researcher a overview of the word(s) spanning product and
classification categories. That is not to say that at some time in the future cost effective and
improved electronic search systems may not be devised such that the classified paper search
records will no longer be unique, but for now such is not the case.
Many electronic searches using the USPTO current system are missing the image associated with
the mark and some appear to be missing entirely. It makes little sense to eliminate one system in
favor of another if such significant discrepancies in the data have not been identified and
corrected. The automated system as it stands now is not a reliable substitute for the paper, just as
the internet search system offered by the USPTO (TESS, TAM) is a poor substitute for X Search
In closing, while computer searching is and will no doubt continue to improve as a useful adjunct
to paper searching, paper is still the resource of choice, particularly where long and often tedious
searching efforts are concerned. Such searchers have also expressed their feeling that higher
quality search results flow when paper is one of the searching resources used.
We applaud the significant efforts the USPTO has made to move into the electronic age, yet
much has to be done to ensure a smooth transition with reliable data for accurate information
dissemination. We continue to offer our assistance in any manner that might aid the agency in
Michael H. Minns
Mark A. Watkins HAHN LOESER & PARKS LLP
I have been reading, with alarm, the proposal of the U.S. PTO to replace certain paper patent and trademark files
now in the public search rooms with an electronic search facility. Since I first began as a patent searcher in 1967,
I have come to regard the paper collections of the Patent and Trademark Office as an essential research and legal
tool. I have followed the development of electronic searching media throughout my career. Although electronic
searching is possible, it is slow and incomplete.
To conduct a serious novelty, validity, or infringement search, there is no substitute for hand searching in
Washington. The community of search firms available in the Washington, D.C. area offers important and
valuable services to the corporate and private bar, and to the independent inventors and small companies alike.
Particularly useful electronic improvements that have been provided in the Patent Ofice search room are
application locator files, and class and sub-class integrity checks. But I assure you there is no substitute for hand
searching in the shoes.
I speak for myself and for the Gillette Company, and I implore you to recommend keeping the paper files
available in both the patent and trademark search rooms.
Donal B. Tobin
PS The undersigned attorneys practicing in my office join in this letter.
|Chester Cekala ||Reg. No. 32,971 |
|Paul I. Douglas ||Reg. No. 31,244 |
|Thomas G. Krivulka ||Reg. No. 38,525 |
|Stephan P. Williams ||Reg. No. 282546 |
|Edward S. Podszus ||Reg. No. 35,983 |
Some contents linked to on this page require a plug-in for PDF file.