# IPCR 399 Introduction of Electronic Data Illustrating the Contents by yah17499

VIEWS: 6 PAGES: 21

• pg 1
									                                                                                IPC/R 3/99
ORIGINAL: English/français
DATE: May 18, 1999/
18 mai 1999
.

WORLD INTELLECTUAL PROPERTY ORGANIZATION
ORGANISATION MONDIALE DE LA PROPRIÉTÉ INTELLECTUELLE
GENEVA/GENÈVE

COMMITTEE OF EXPERTS OF THE IPC UNION
COMITÉ D’EXPERTS DE L’UNION DE L’IPC

IPC REFORM PROJECT FILE/DOSSIER DE PROJET DE RÉFORME DE LA CIB

SUBJECT:                  INTRODUCTION OF ELECTRONIC DATA ILLUSTRATING THE CONTENTS OF IPC ENTRIES:
EXAMPLES OF PATENT DOCUMENTS, DETAILED NOTES AND CLASSIFICATION
DEFINITIONS, EXEMPLIFYING TECHNICAL TERMS
SUJET :                   INCORPORATION DE DONNÉES ÉLECTRONIQUES POUR ILLUSTRER LE CONTENU DES
ENTRÉES DE LA CIB : EXEMPLES DE DOCUMENTS DE BREVET, NOTES DÉTAILLÉES,
DÉFINITIONS ET EMPLOI DE TERMES TECHNIQUES CONCRETS

ANNEX/                                                                      SEE/VOIR    ORIGIN/
ANNEXE                                      CONTENT/CONTENU                   R 3/99    ORIGINE      DATE

6           Rapporteur report                 Rapport du rapporteur                  US       10.05.99

IPC/R 3/99

ANNEX 1/ANNEXE 1

Swedish Patent and                                               Introduction of electronic data
illustrating the contents of IPC entries
Registration Office                                                   Date: April 1st, 1999

on IPC/CE/28/5, Annex 5, Item 3

The Ad Hoc IPC Reform Working Group was given the task of considering the
introduction of electronic data illustrating the contents of IPC entries: examples of
patent documents, detailed notes and classification definitions, where appropriate, more
exemplifying technical terms.

We support the introduction of additional electronic data into the IPC. However, it should be
observed that this will necessarily mean that the paper and electronic versions will become
different. It is possible to include a lot of additional information in an electronic IPC, but there
is no way that the same information can be added to a paper version. In view of the technical
developments, we see it as unavoidable and natural, but it will be a major change in policy.

It is difficult to discuss "electronic data" without having a picture of what a future electronic
IPC will look like. We will presume that it will have one main layer, roughly corresponding to
the present paper version, which is linked to several additional layers giving further
information. These additional layers will only be visible as hyperlinks in the main layer.

The different types of additional information will have different status. Some will be of the
same character as notes, that is rules that have to be followed. Some of the additional
information will be references to paragraphs of the Guide, which is also information of high
dignity. Some will be definitions of technical terms. Other types will be mere user guidance,
such as informative references and example documents. It will be necessary to keep the
different types of additional information separated and clearly identifiable. The user must be
able to distinguish between important system information and mere user guidance. The IPC
already contains a huge volume of text, and it would probably be counterproductive to add
more text to the basic layer.

The linking of information to the basic text would give big advantages compared to the
present paper version. Notes and paragraphs of the Guide could be presented in their context,
and not somewhere else, far from the actual groups of the IPC, requiring the user to look in
another part of the book or in another book. We would also see it as an advantage if the
informative references could be removed from the text of the basic layer, leaving only the
essential limiting references.

Document IPC/CE/28/5, Annex 5, Item 3, mentioned three different types of electronic data.
We have the following comments regarding each of these types:
IPC/R 3/99
Annex 1/Annexe 1
page 2

Example patent documents

This is in our opinion the least desirable type of additional data. Anyone working at a
database. To give any added value, the documents would have to be carefully selected.
This would require a considerable amount of work. The example documents would
become outdated and would need to be replaced. We are also concerned that actual
documents would limit the interpretation of the scope of groups.

Detailed notes and classification definitions

We understand this as detailed, more explicit references and detailed definitions of the
coverage of technical terms and of the borderlines between different places.

The references are one of the big problems of the IPC. There are several reasons for this.
One reason is that it is difficult for non-IPC experts to see the difference between
informative and limiting references. Another reason is that the references do not take the
inventive concept into account. These are difficulties that could be avoided if more
elaborate texts were available.

The wording of the IPC itself is very condensed and does only seldom take the inventive
concept into account. In order to properly reflect the intentions of the Guide, most titles
in the IPC should really be worded along the lines of "inventions characterised by
features relating to xxx" instead of just simply "xxx". Again, more elaborate texts would
improve the situation.

In places where everyday language or technical terms of the trade are used, there is often
a need to be more clear. The normal language is not as exact as we want the
classification to be. As a long-term goal it would be better to replace the inexact
wordings used in the IPC itself by more exact functional wording, but additional
definitions will certainly help in many places.

Exemplifying technical terms

This relates to the opposite situation to the one mentioned in the last paragraph, where
to title in the IPC uses a functional language that might perhaps be difficult to
understand for the man in the street or the technical expert. In general, we see this as
the desirable situation – to have an exact title in the IPC and additional examples, rather
than having inexact titles that need additional definitions.
IPC/R 3/99
Annex 1/Annexe 1
page 3

Conclusion: We strongly support the addition of an additional information layer to the
electronic version of the IPC. However, providing this additional information will require a lot
of resources. Unless additional resources are committed from offices and from WIPO, it will
only be possible to introduce these additional data in very limited areas of the IPC.

Anders Bruun

[Annex 2 follows/
L´annexe 2 suit]
IPC/R 3/99

ANNEX 2/ANNEXE 2

April 14, 1999

Mr. Mikhail Makarov
Classification and Patent
Information Division,
World Intellectual Property Organization
34, chemin des Colombettes
1211 Geneva 20
SWITZERLAND

Re:   IPC Reform Working Group-Task 3 – Consider the introduction of electronic data
illustrating the contents of IPC entries
________________________________________________________________________

Dear Mr. Makarov:

Enclosed is the initial US response to Task 3 as set forth in Annex V of the Final Report
of the Twenty-Eighth Session Committee of Experts (IPC/CE/28/5) for your consideration.
Copies of this document have also been circulated via e-mail to the IPC Reform Working Group
Members.

Sincerely,

Robert W. Saifer, Director
International Liaison Staff

Enclosure
cc Reform WG Members
IPC/R 3/99
Annex 2/Annexe 2
page 2

REFORM WORKING GROUP

Consider the introduction of electronic data illustrating the contents of IPC entries

We are hesitant to begin the discussion of this topic by simply jumping into our opinion of
what we believe are useful additions to the information already found in the IPC. Prior to the
Working Group (WG) proceeding further on this topic, it is our conviction that general agreement
is needed on what our goals are and what this data should be required to accomplish.

We envision the ‘new IPC’classification schedules being simplified and enhanced with
electronic content information that:

1. Clarifies the ‘intended’scope (i.e., specifies the conceptual boundaries of the category of
invention covered) of each level of the classification to improve consistency of assignment.
This both reduces the time for searching and ensures a reasonably complete search of the most
appropriate art.

2. Improves the ability of trained search examiners to locate and search the most appropriate
groups for all the various search requirements of all the Intellectual Property Offices (IPOs).

s
3. Vastly increases the IPC’ usefulness to non-expert public searchers, which will expand the
future economic benefits of patents and encourage the use of new technologies.

4. Simplifies the presentation of the current and additional content information by utilizing
multiple layered electronic access techniques.

These goals are all admirable and noble in nature, but they can not be successfully
accomplished without one key factor. We firmly believe that the possibility of successfully
accomplishing many of the above goals is completely dependent upon the successful completion
of an IPC master classification file (i.e., IPC-MCF). Why is this essential? Because the IPC
cannot be converted from its current representative scope, to its future definitive scope, without
agreement by the WG on the population content of all of its classification symbols. Our views on
this subject are discussed in extensive detail in our task 6 paper in section 3 and can be referred to
for clarification if necessary.

It is our opinion that the WG must now definitely commit to the creation of an IPC-MCF.
The WG must not establish plans for the introduction of electronic data into the IPC for
illustrating the content of classifications without this commitment. This decision is fundamental
to determining what are the appropriate means for adding clarity to the IPC classifications.
IPC/R 3/99
Annex 2/Annexe 2
page 3

We also believe that one concept that is central to the creation of definitions in the IPC has
not been previously discussed by the IPOs. This concept is called ‘      coextensive titles and
.                                   s
definitions’ It requires that a classification’ title, its assigned obligatory patent documents, and
the definition of the classification all have the same scope. Although this concept is not currently
specifically required anywhere in the existing IPC, we believe that its incorporation into the new
reformed IPC is essential to its future success. Because of its importance, we have attached
Annex 1 with a more expansive explanation of the concept.

SUGGESTIONS FOR NEW IPC HAVING DEFINITIVE SCOPE

We recommend the creation of the following four types of electronic data that are useful
when the IPC-MCF is completed and the coextensive concept is added to the new IPC. These are:

A. the introduction of coextensive definitions for all subclasses and groups in the IPC,

B. the transfer of most ‘content information’found in the section and class level to the
definitions of the subclasses where it is most useful,

C. the transfer of all references and notes from their current location in the IPC schedules to the
definitions where they are appropriate, and

D. the introduction of multi-language ‘catch term lists’at the subclass or main group level for
selected technologies.

A. COEXTENSIVE DEFINITIONS FOR CLASSIFICATION TITLES

The use of separate definitions for all specific classifications in the new reformed IPC should be
s
the WG’ ultimate goal. In our opinion, this is the only way to keep the subclass schedules of
manageable length in a definitive system while still providing an essential level of guidance. In
most situations, this would reduce the need for separate ‘    term definitions’in the IPC and their
s
future use would be greatly reduced. Because of this topic’ importance, we have attached Annex
2 with a more expansive explanation of the concept and what we envision as a suitable standard
for IPC definitions.

However, we are also aware that definitions for all IPC classifications can not be created
overnight. In our opinion, the WG should make the creation of non-overlapping subclass
definitions a top priority, since this will help prevent the dispersion of similar inventions into
separate subclasses. We believe this task should be completed, to maximize its benefits, prior to
s
populating the reformed IPC. After the IPC-MCF’ initial population, our opinion is that
definitions should be written to at least the main group level during any significant revision
project in a subclass. Separate definitions for these main groups would also help classifiers
maintain the intended scope of all hierarchically indented groups and clarify many of the possible
ambiguities of their titles. Subgroup level definitions should also be encouraged and required
IPC/R 3/99
Annex 2/Annexe 2
page 4

when the WG believes they are useful.

i.e.’
Most term definitions, and many of the existing ‘ terms and expressions now included in
classification titles format (see paragraph 36 of Guide), should be deleted and incorporated into
the separate definitions for their classifications. Any term definitions that are still necessary
s
could be electronically stored separate from their subclass schedule’ text and displayed by mere
clicking on the defined terminology in the text of the schedule using hypertext links. This type
access would provide a high level of redundancy that would ensure correct usage of the term or
expression specifically defined throughout its subclass. In our opinion, no term definitions would
need to be specified at the section or class level. Those that are appropriate for more than one
subclass should now be repeated at the subclass level for all appropriate subclasses.

B. TRANSFER OF SECTION AND CLASS CONTENT INFORMATION

In our opinion, it would be beneficial to searchers and classifiers to remove all information (e.g.,
notes, definitions) currently found with the section, subsection, or class titles modifying the
effective scope of subclasses (paragraph 44 of Guide). This information should be included with
the definition for each appropriate subclass to avoid possible misinterpretations of the subclass’
scope by users. While the location of this information was understandable when paper schedules
were primarily used to reduce volume, it is not appropriate in a definitive system. In the new
reformed IPC, the section, subsection, and class titles will not need to be defined, since they are
only intended to be broadly indicative of related subject matter and not definitive (paragraphs 15
&16 of Guide).

C. LINE NOTES AND REFERENCES

Notes that are used to clarify the lines of distinction between their subclass or group and other
classifications in the IPC (i.e., line notes) are essential in maintaining consistency of assignment.
In addition references which show that the subject matter indicated by the reference is classified in
other referred to classifications (i.e., line references) are also essential in maintaining
consistency of assignment. Both types of information should be removed from the IPC schedules
and stored separately to provided redundant access to all classifications having their scope
modified thereby (see paragraphs 44 &45 of the Guide). Furthermore, line references and line
notes in the reverse direction should be incorporated in the specified classification(s) to ensure
that all users are aware of the interrelationship between the classifications. The viewing of this
type information during a search could be accomplished by clicking on the classification’      s
symbol in the schedule using hypertext links. In addition, whenever definitions exist for a
classification, we suggest that this type information be located adjacent to them in a systematic
and uniform manner.
IPC/R 3/99
Annex 2/Annexe 2
page 5

D. CREATION OF MULTIPLE LANGUAGE CATCH TERM LISTS

In our paper for the Advanced Seminar (i.e., IPC/SEM/98/5), US suggested in its Annex the
creation of a new IPC search tool. In that paper we referred to it as a ‘                     .
catch term list’ We
suggested that examiners create a separate list of terms for each subclass that is useful for on-line
full text searching in one of the official languages. These official catch terms could then be
interpreted by each IPO using collected references. Each IPO would create parallel catch term
lists and link them electronically to their official equivalent list to allow multi-language term
searching. The intended use of this tool is to avoid costly assignment of indexing codes to
individual patent documents whenever possible. While the primary purpose of this tool is for
online searching, we believe that it is also beneficial in illustrating the content of both its subclass
and dependent groups. In fact, linking each of the catch term lists with the range of titles and
definitions they cover would be an enormous help to searchers. Examiners will suggest the terms
in each list and they will frequently be the currently used art terms for their technology. These
lists can act as an expansion of the “Official Catchword Index” in an electronic term search of the
titles and definitions. This will be a particularly useful benefit to non-expert public searchers
who are using translations of the IPC that do not include a Catchword Index.

s
In conclusion, it is this Working Group’ mission to make effective use of the worldwide
electronic environment envisioned for the reformed IPC. This will provide an opportunity for all
users to take full advantage of the full text search and electronic association capability SCIT will
provide for them. It is essential that we maximize the benefits that an electronic environment can
provide to assist in properly classifying in and searching the IPC classifications. The words
found within classification schedules, definitions, notes, and references associated with the
classifications and patent documents can now easily be searched electronically. Therefore, we
should provide users with electronic associations between schedule titles, definitions, line notes,
line references and representative drawings and diagrams. Electronic links (e.g., hypertext links)
can be provided between different information within the IPC, and to external classification
schemes, for example economic or industrial classification schemes and the “second level”
schemes built by independent offices over the base or core level IPC.
IPC/R 3/99
Annex 2/Annexe 2
page 6

ANNEX 1

THE CONCEPT OF COEXTENSIVE CLASSIFICATION TITLES AND DEFINITIONS

The title of any IPC subclass or group in which patent documents are classified as
obligatory classifications must be written in such a manner as to be coextensive in scope with
both the obligatory patent documents classified therein and its definition along with note(s). This
task is more difficult than it at first seems, since it is often necessary during a classification
project to modify the wording of a title to reflect alterations in the scope of the obligatory patent
documents proper for the subclass or group. This guideline must be followed unwaveringly to
ensure that all obligatory patent documents meet every limitation found in both the
s
classification’ (a) definition along with note(s) and (b) all superior, hierarchically related (i.e.,
subclass and parent groups) definition with note(s) from which it depends.
However, the concept of coextension between a title and its definition, when its meaning
is taken in the most literal sense, is contradictory in principle to the concept of having a
definition. It is the purpose of a definition to point out the exact boundaries of its subclass or
group. This allows the title of a classification to maintain a reasonable length while the scope of
the obligatory patent documents classified therein is controlled over extended periods of time by
the limitations in the definition. Consequently, the concept of coextension has been interpreted
when applied to titles, to mean that the title of a subclass or group must be only indicative of all
its essential limitations. The title of a classification need not necessarily represent, or explicitly
indicate, the exact boundary of all the limitations required by its definition or explanatory notes
thereof. In addition, it is common for the title of a subclass not to specify that its scope includes
either unprovided for combinations, unprovided for components, or related subject matter not
specified as appropriate for another subclass. However, when this subject matter is found within
the subclass, it must be clearly stated in the subclass’definition and expressly provided for in a
dependent group when useful.
In situations where the scope of the claims are ambiguous, the original classification of such
patent documents should be assigned based on the full disclosure and the Classifier's opinion of
what the inventor intended as the invention. Patent documents having ambiguous claim scope
should not be used as the basis for creating subclasses or groups.
IPC/R 3/99
Annex 2/Annexe 2
page 7

ANNEX 2
THE CONCEPT OF IPC DEFINITIONS

The definition of an IPC subclass or group is a supplemental explanation of its title. It
identifies, in conjunction with its complementary notes and references, the scope of the subject
matter necessary for obligatory classification of patent documents into the subclass or group. The
definition must set forth in a clear, concise, and precise manner the required properties, attributes,
or characteristics possessed by patent documents included as obligatory type classifications in the
subclass or group. It is not unusual for some of the required properties, attributes, or
characteristics specified in the definition of one subclass or group to be found in patent
documents properly assigned as obligatory classifications in another subclass or group. However,
it is necessary in these situations for at least one additional distinct requirement of the invention
to be specified in the definition and notes of each classification to maintain their uniqueness and
prevent any overlap between the patent documents appropriately assigned to each.

As a supplemental explanation of the title, the definition clarifies what subject matter is
unique to its classification. To achieve its intended purpose, a definition must explicitly interpret
the title by elaborating on the essential details thereof. The definition and its complementary
notes must also correct all probable misunderstanding on the intended scope and reconcile
differences with other classifications in the IPC. For these reasons, a definition of a subclass or
group which, in effect, merely repeats its title, or the terms in its title which are relied on to
distinguish its scope, is unacceptable in almost all situations. However, the degree of elaboration
which is necessary to accomplish these goals, must be determined independently in each situation
by a consensus of the opinions of the classifiers and examiners concerned with creating the
subdivision and may, in certain situations, allow the repetition of terms in both.

Finally, as a general guide, a classifier should never define any term used in a title in greater
detail than is justified by the disclosure of all patent documents placed within the subclass or
group as an obligatory type classification. The reason for this is obvious; a patent document must
in all situations clearly disclose what is used as the basis for any obligatory classification. For
example, if most of the patent documents assigned to a group "Wood lathe" state only that "a
wood product is turned", then the definition should not go into details of the products that are not
disclosed. In this situation defining "wood" as "wooden pieces or blocks" would be acceptable.
Defining it as "a hard fibrous substance which is basically xylem located beneath the bark of a
tree" would be more restrictive than the disclosures and not acceptable.

[Annex 3 follows/
L´annexe 3 suit]
IPC/R 3/99

ANNEX 3/ANNEXE 3

AD HOC IPC REFORM WORKING GROUP

submitted by the International Bureau

Under Task 3 the introduction of electronic data illustrating the contents of IPC entries
should be considered. This could be achieved by introducing hyperlinks in the IPC:CLASS
CD-ROM and in the Internet version of the IPC in order to avoid overweighing of the text of
the IPC entries. A flag next to the entry or a button at the top would permit the appearance of
windows with the corresponding text. Below follows a non-exhaustive list of items that could
be available.

EXAMPLES OF PATENT DOCUMENTS ILLUSTRATING THE CONTENT OF A
GROUP

This would concern newly introduced groups since documents for older groups could be
easily available through patent databases. Attention should be made in order to indicate
unambiguous documents. A short comment could accompany the document, when necessary,
for more clarity. It should be considered whether the whole document should be present or
just its patent number with a link to some digital library. The provision of such examples and
texts would be the task of the proposing office or of the rapporteur.

EXEMPLIFYING TECHNICAL TERMS

In the current practice a limited number(usually one or two) of exemplifyin g technical
terms are added to the text of an entry, if necessary, usually non-obvious ones. On the
“hidden” layer a more exhaustive list of technical terms should be available including obvious
ones. These terms, although hidden, should be a part of the Classification for searching
purposes. If the inclusion of such terms is considered to be necessary even in the existing
entries this would necessitate a huge amount of work and offices, particularly the larger ones,
should be aware of this effort before embarking on such a project.

In the next edition of IPC:CLASS already hyperlinks to the relevant notes which
concern an entry will be provided, even if these notes are not present close to the said entry.
The Working Group should consider whether such a presentation is sufficient or if another
one should be more appropriate.

IPC/R 3/99
Annex 3/Annexe 3
page 2

Terms having a particular meaning explained in the Guide should have a particular
color and be hyperlinked to the relevant place in the Guide. The same should be provided with
the places of the IPC where special rules apply indicated in the Guide.

[Annex 4 follows/
L´annexe 4 suit]
IPC/R 3/99

ANNEX 4/ANNEXE 4

DEUTSCHES PATENT- UND MARKENAMT                                     IPC/REF/WG              Task 3
German Patent and Trademark Office                                           Date : 3.05.1999

Consider the introduction of electronic data illustrating the contents of IPC entries: examples of
patent documents, detailed notes and classification definitions, where appropriate, more
exemplifying technical terms

Re:    Comments on document IPC/CE/28/5, Annex V

Illustrating the content of an IPC group could be achieved by

Examples of patent documents
We think, this would at a first step only conc ern newly introduced groups or places where it is
very urgent to give examples because the wording of that place is ambiguous or difficult to
understand. Such examples should be submitted by the proposing office or by the rapporteur.

Examples of chemical formulae
In some areas of the chemical field of the IPC, e.g. C07D, it is rather difficult to work with the
text of the groups since it is not easy to understand the content of the group. In C07D there are a lot of
places where a chemical formula could be added so that the wording could be easier understood. The
formula is the speech of the chemists.

A proposal along that line was already initiated in project PCIPI/C152/91. At that time (1991) it
was decided not to introduce any of the proposed formulae since it was difficult to represent formulae
in textual databases. However, it was agreed upon that such formulae would be very useful.
After some years the problem was taken up again and considered at the session of Subgroup A in
Rijswijk, in April 1996, as well as at the seventeenth session of SI-Working Group in June 1996 (see
document PCIPI/S/XVII/9, Annex F). It was noticed that the reason of the representation of chemical
formulae mentioned above was no longer valid since now in many databases, e.g. CA, CLAIMS,
PATDPA, IPC:CLASS CD-ROM etc. chemical formulae are included. For a start one could take up the
part of the proposal of project C152 which was postponed.

[Annex 5 follows/
L´annexe 5 suit]

R3_an4.doc
IPC/R 3/99

ANNEX 5/ANNEXE 5

UK Patent Office

Electronic data illustrating contents of IPC entries

We think that this linked electronic data is essential to the understanding of the IPC. It is very
difficult for the actual entries to be clearly definitive but not to be too long and complex and
the explanation and exemplification of entries in other classifications appears an admirable
feature. We imagine this embodied as a hyperlink symbol in the text of the IIPC that can lead
directly to the electronic additional data when using an electronic version of IPC.

We can also envisage links from entries to lists of synonyms to enable more efficient database
searching.

On the other hand, IPC will need a set of rules giving the context of the electronic data and its
relationship with the entries themselves. It will follow that care will be needed to ensure that
the additional data will be consistent with the classification entries.

Jim Calvert
04 May 1999

[Annex 6 follows/
L´annexe 6 suit]
IPC/R 3/99

ANNEX 6/ANNEXE 6

May 10, 1999

Mr. Mikhail Makarov
Classification and Patent
Information Division,
World Intellectual Property Organization
34, chemin des Colombettes
1211 Geneva 20
SWITZERLAND

Re:   Rapporteur Report for IPC Reform Working Group- Task 3 – Consider the introduction
of electronic data illustrating the contents of IPC entries
___________________________________________________________________________

Dear Mr. Makarov:

Enclosed is the US rapporteur report for Task 3 as set forth in Annex V of the Final
Report of the Twenty-Eighth Session Committee of Experts (IPC/CE/28/5) for your
consideration. Copies of this document have also been circulated via e-mail to the IPC
Reform Working Group Members.

Sincerely,

Robert W. Saifer, Director
International Liaison Staff

Enclosure
cc Reform WG Members
IPC/R 3/99
Annex 6/Annexe 6
page 2

REFORM WORKING GROUP

US RAPPORTEUR REPORT ON TASK 3

Consider the introduction of electronic data illustrating the contents of IPC entries.

Our Purpose

Recurrent in discussions leading to the decision of the IPC Committee of Experts to
launch the IPC reform effort, were suggestions of the need for enhancing the illustrative
material associated with the IPC. There is little disagreement that such material would not
only increase the ease and effectiveness of IPC use, but also would make important
contribution towards achieving greater consistency in document placement within the IPC. In
the past, a major expansion of illustrative items such as category definitions, examples, search
notes, exemplary technical terms, etc., was seen as not feasible for two principal reasons.
Their creation and maintenance would require substantial additional work on the part of the
Offices maintaining the IPC; and they would result in an unwieldy and unmanageable bulk in
s
the IPC’ essentially paper based reference material. However, assumptions underlying the
IPC reform effort promise resolution of these obstacles. The reformed IPC is envisioned as
being optimized for use in an electronic environment wherein illustrative items would be in
the form of electronic data, easy to maintain and adding no meaningful bulk to reference
material.

The primary mission of our Working Group is to make effective use of the worldwide
electronic environment envisioned for the reformed IPC. This will provide an opportunity for
all current and potential users to take full advantage of the full text search and electronic
association capability which SCIT will provide for them. It is essential that we maximize the
benefits that an electronic environment can provide for classifying in and searching the IPC.
Failure to accomplish this goal in an extremely rapid manner will doom the IPC to oblivion.
Our rapidly changing, and ever expanding, technology has taught everyone the lesson that
history is not kind to the obsolete. A horse and carriage is now seldom used to travel across a
country, and it is kindly preserved only as a novelty in the industrial world and not as a
necessity. In the electronic world, IPC will be relegated to a similar status without our speedy
intervention.

distinct levels or layers of descriptive information for illustrating the contents of IPC entries.
This type of multiple layering of information is required for proper usage of the IPC by a
variety of divergent users and for specialized uses. The definitive information necessary for
limiting the scope of each particular classification symbol must be capable of customization
without sacrificing uniformity of presentation. Obviously, an expert EP searcher who has a
docket including only a very small portion of the IPC will not usually need to look up
definitions of subject matter. However, for a public searcher or an examiner from PT for
example, with a docket covering an extremely large portion of the IPC, frequent use of
IPC/R 3/99
Annex 6/Annexe 6
page 3

definitions will likely be quite common. Thus, the electronic illustrative information provided
and its mode of access must optimize searching for all users within reasonable cost.

Discussion of Threshold Issues

s
In the Rapporteur’ opinion there are two essential related issues that must be decided
prior to the WG determining what informative data to add or reorganize in the IPC. We
recommend that the WG begin the exploration of this topic by at least their discussion at this
meeting and hopefully their resolution prior to solidification of positions on this task.

The first issue is the creation of an independent IPC master classification file (i.e., IPC-
MCF). We firmly believe that the possibility of successfully accomplishing reformation of
the IPC is strongly dependent upon the successful creation, and continued maintenance of an
IPC-MCF. The IPC cannot be converted from its current representative scope to a definitive
scope unless the population content of all classification symbols within the ‘  official’IPC are
‘officially’established and maintained. Our suggestion to create this IPC-MCF is not
intended to preclude or discourage IPOs from maintaining their own separate classification
files with IPC information. However, it would allow IPOs to fully utilize the planned WIPO
IPDL for electronic patent document searches without maintaining their own separate files.

This concept is a radical departure in philosophy for the IPC. It requires us all to have a
pro-active worldwide approach to patent document, and related patent-like information,
searching. Currently, the documents searched by most IPOs are the ones that are in their
national Offices and under their sole control. The use and population of the IPC is
accomplished as each Office sees fit without external control or supervision. We all know the
existing IPC situation will not allow a useful worldwide search to be a possibility. Both the
EP and US long ago determined independently that their internal systems would need to
reclassify all foreign documents for useful worldwide document searching. The cost of such
an approach is rapidly becoming prohibitive even for the largest IPOs. We must act now or it
will be too late.

The second issue relates to the concept of ‘   coextensive titles and definitions’that, we feel,
must be incorporated into the new reformed IPC. It is essential to the creation of meaningful
definitions in the IPC which are necessary to ensure consistent assignment of documents
s
worldwide. It requires that a classification’ title, its assigned obligatory patent documents,
and the definition of the classification all have the same scope. If an IPC-MCF is created, this
concept becomes possible for the first time in the IPC. Without an IPC-MCF, classification
definitions are not useful and only very limited additional descriptive information for
clarifying scope can be added. Any real degree of useful definitiveness becomes unobtainable
for the IPC.

s
In the Rapporteur’ view, there are only three possible types of electronic data that are
useful without the coextensive concept being included in the reformed IPC. These are:

(a) the addition of example patents and chemical formulae,
(b) massive expansion of the use and location of term definitions, and
(c) increased use and location of line notes and references.
IPC/R 3/99
Annex 6/Annexe 6
page 4

These types of additional data are useful for adding clarity to the classification titles, but
they will not alone make the IPC a preferred worldwide search tool. Without the creation of an
IPC-MCF, we are merely getting out of the carriage to help the horse pull it along the
autobahn.

commenting Offices support the inclusion of at least some additional electronic data into the
IPC. It is clear from their comments, that these Offices also accept that this data will not be
limited to being viewable in the schedules and must be accessed by some type of hypertexts
links. Rapporteur will individually discuss each form of illustrative data covered in the

1. Inclusion of Example Patents

DE, IB, and SE, all had specific opinions on the use of example patents to illustrate the
content of groups. DE and SE support their use for newly introduced groups and believe that
the proposing Office or Rapporteur should provide them. DE states that they may also be
useful for existing classifications where the wording is ambiguous or difficult to understand.
IB suggests that only a portion of the document may be needed and that accompanying short
comments might be useful. SE is not enraptured by the concept, but gives limited support to
the concept. SE fears that the example documents will limit the interpretation of the scope of
the groups with which they are associated.
The US has experience in the use of example patent documents. In a system such as the
United States Patent Classification system (i.e., USPC), they are normally not needed except
s
where a classification’ wording is ambiguous or the mechanical interrelationship is difficult
to understand. However, the USPC populates its classifications prior to finalization of titles
and creates definitions for each classification to limit possible alternative interpretations of the
scope of classifications.
s
It is important in Rapporteur’ opinion for the IPC to narrow, as much as practical,
alternative interpretations of the scope of each IPC group and subclass. Under current IPC
practice, it is absolutely essential to accomplish this for newly created IPC groups and
subclasses, since they have no backfile in existence. In a representative classification system
such as the current IPC, without definitions or an approved official backfile collection of
patent documents, this can primarily be achieved by the use of example patent documents.
Whenever these documents are used to illustrate the content of a classification, they should be
in one of the official languages and obtainable by hypertext linked to their appropriate
classification. It would make the documents far more useful if specific text and drawings
pertinent to the classifications they are linked to were marked for the searcher as suggested by
IB. These example patent documents might also be invaluable to several IPOs during their
translation of new IPC classification from the official language IPC text into their preferred
languages.

2. Inclusion of Chemical Formulae

DE supports the inclusion of chemical formulae in some chemical fields of the IPC where
it is difficult to understand the content of the groups by their titles alone. In their opinion,
IPC/R 3/99
Annex 6/Annexe 6
page 5

formulae “are the speech of the chemists” in the same manner as drawings of patents are for
mechanical devices.
Rapporteur supports the concept of including chemical formulae in the IPC wherever it is
deemed useful by the WG. However, it seems that this information should not be part of the
subclass schedules as was necessary in the past, but merely linked by hypertexts to the
appropriate classification symbols. This will allow an expansive use of this concept to many
other groups without the fear of needlessly cluttering the subclass schedules.

3. Creation of Classification Definitions

GB and US, had specific opinions supporting the creation of separate definitions for
classifications.
GB believes that separate explanation and exemplification of classification titles is an
admirable feature. These definitions could be linked to their classification using hypertexts
US believes that the use of separate definitions for all specific classifications in the new
s
reformed IPC should be one of the WG’ primary goals. They think that it is the only way to
keep the subclass schedules of manageable length in a definitive system while still providing
an essential level of guidance. In their opinion, the WG should make the creation of non-
overlapping subclass definitions a top priority, since this will help prevent the dispersion of
similar inventions into separate subclasses. They also believe that most term definitions, and
i.e.’
many of the existing ‘ terms and expressions now included in classification titles format,
should be deleted and incorporated into the separate definitions for their classifications.
s
In Rapporteur’ opinion, it is important for the IPC to narrow as much as practical
alternative interpretations of the scope of each IPC group and subclass. Separate definitions
for classifications using different and more elaborate wording for the specified concept do
this. Rapporteur recommends the conceptual adoption and testing of this concept by the WG.

4. Expanding and Modifying Notes or References

IB, SE, and US all expressed specific opinions on various aspects of notes or references.
IB states that all notes that are concerned with a classification are already hypertexts
linked to the classification in the 7 th edition IPC CLASS.
SE wants the presentation of a reference to make it clear to searchers when a reference
is only informative and when it is intended to limit the scope of its classification.   SE also
supports additional definitions for technical terms for appropriate classifications.
US believes that both notes and references should be removed from the IPC schedules and
stored separately to provided redundant access to all classifications having their scope
modified thereby. The notes and references to a classification should also be applied in the
reverse direction to insure that the established line is known and followed by all. In US
opinion, it would be beneficial to transfer all notes and references currently found with the
section, subsection, or class titles to the subclasses they modify.
The expanded use of notes and references and the clarification of their text and meaning
seems to be supported by all of the commenting Offices. The WG needs to determine how
best to link this information to all appropriate classifications and when it should be viewed
(e.g., always with each appropriate title in the schedule, with a separate definition of the title
and viewed only when needed for clarification).
IPC/R 3/99
Annex 6/Annexe 6
page 6

5. Inclusion of Synonyms, Exemplifying Technical Terms, or Catch Term List

GB, IB, SE, and US all expressed specific opinions on the use of synonyms, example
technical terms, or catch-term lists for illustrating the content of IPC classifications.
Rapporteur has grouped these comments together since they overlap to some extent.
GB supports hypertexts linking of classifications to lists of synonyms to enable efficient
searching.
IB supports the inclusion of hypertexts linked technical terms for exemplifying
classifications. IB suggests that these lists be far more exhaustive than current term lists in the
schedules. These lists would include both readily apparent and non-readily apparent and they
should be part of the classification for searching purposes. IB warns that creation of these lists
is a huge amount of work.
SE would prefer additional examples in titles rather than inexact titles that need additional
definitions exemplifying the technical terms used in the titles.
US supports creation of a separate list of terms for each subclass that is useful for on-line
full text searching in one of the official languages. Each IPO could then create parallel catch
term lists and link them electronically to their equivalents in the official list to allow multi-
language term searching of all patent documents of WIPO members. US believes these lists
are also beneficial in illustrating the content of both their subclasses and dependent groups to
searchers.
The Rapporteur believes there is significant support for linking useful search terms and
example terms to classifications. The Rapporteur suggests that the WG investigate the
success of the EP with its existing synonym lists and discuss potential problems. EP is
currently expanding its full text patent document search capacity and it publishes in three
official languages. EP might consider testing the impact of adding equivalent French and
German terms to some of its existing synonym lists if other Offices agree to help with
translations. This would give the WG a useful indication of the degree of search improvement
over mere abstract searching of documents that might be expected. In many areas of the IPC,
the newly translated JP F-terms may also be useful in helping to form such lists. It is
s
Rapporteur’ opinion that of the improvements discussed in this paper, this one has the
highest potential for rapidly improving the search capability of IPOs using electronic search
tools provided by SCIT.