T 2435/09 (Web site usage/ALCATEL LUCENT) of 10.11.2015

Summary of Facts and Submissions

I. The original applicant, Lucent Technologies Inc., appealed against the decision of the Examining Division to refuse the European patent application no. 99303604.5, and requested in the notice of appeal that the decision be set aside and a patent be granted.

II. In the contested decision, the Examining Division arrived at the conclusion that claim 1 filed with letter dated 23 April 2009 comprised subject-matter which extended beyond the content of the application as filed (Article 123(2) EPC). Furthermore, the application did not meet the requirements of Article 84 EPC, because claim 1 was not clear.

III. The Examining Division's decision did not address the questions of novelty and inventive step. However, in a communication dated 26 March 2003, the opinion was expressed that the subject-matter of claim 1 then on file did not involve an inventive step with respect to the following document:

D1: Bradford L. Barrett: "Readme" Internet article about the Web server log file analysis tool Webalizer, [Online] December 1997 (1997 - 12), XP002213789, retrieved from the Internet, <URL:ftp://ftp.mrunix.net/pub/webalizer> [retrieved on 2002-09-06].

In the same communication, it was furthermore noted that a similar reasoning could be set forth using any of the six other documents identified in the decision.

IV. With the statement of grounds of appeal, the then appellant maintained, as main request, the request considered by the Examining Division and submitted a new claim 1 by way of auxiliary request. Furthermore, the appellant pointed out that the issues for the appeal were whether the subject-matter of claim 1 filed with letter dated 23 April 2009 (main request) extended beyond the content of the application as filed, whether certain subject-matter of this claim was impermissibly defined in terms of a result to be achieved, and whether certain terms recited in claim 1 were unclear, as concluded by the Examining Division.

V. The registration of the transfer of the present application to Alcatel Lucent (the present appellant) took effect on 2 February 2012.

VI. With letter dated 19 December 2014, the appellant was summoned to oral proceedings to be held on 10 November 2015.

VII. In a communication dated 9 September 2015 pursuant to Article 15(1) RPBA, the Board expressed the preliminary opinion that claim 1 of the main request appeared not to be in compliance with Article 123(2) EPC and that some terms used in the claim appeared unclear or not in conformity with the description (Article 84 EPC).

Furthermore, the Board noted that, although claim 1 according to the auxiliary request appeared to overcome some of the objections under Articles 84 and 123(2) EPC raised by the first instance, it still contained features which were not clear or not in conformity with the application as originally filed.

VIII. In reply to the Board's communication, the appellant submitted with letter dated 23 September 2015 a new claim 1 as new main request, changed the previous main request into a first auxiliary request and withdrew the pending auxiliary request.

IX. At the oral proceedings, which were held as scheduled on 10 November 2015, the appellant submitted and subsequently amended a new first auxiliary request. The previous auxiliary request was maintained as the second auxiliary request.

X. The appellant's final requests were that the decision under appeal be set aside and that a patent be granted on the basis of the main request as filed with the letter of 23 September 2015 or, in the alternative, on the basis of the claim of the first auxiliary request as filed at the oral proceedings at 14:15 and amended at 14:25 or of the claim of the second auxiliary request filed as the then sole request in the examination proceedings with the letter of 23 April 2009.

XI. The only claim of the main request reads as follows:

"A method for building a database of information about visits to a Web site, the Web site comprising Web components organized in a file directory structure having directories and sub-directories, a Web component comprising a Webpage, or a basic data building block such as text, hypertext, images, embedded client software programs or other data displayable by a browser residing in a file which is accessible through its URL, the method comprising:

a) retrieving (95) raw hit records, a raw hit record comprising a logfile, each raw hit record pertaining to a particular Web component of the Web site;

b) extracting (85) selected information from said raw hit records, thereby to create, from each raw hit record, a filtered hit record, the selected information including information identifying the Web component to which the respective hit record pertains;

c) distinguishing (85) respective filtered hit records according to the visit to which each of them belongs, and associating (85) a particular visit index with each filtered hit record;

d) storing (100) the filtered hit records in a data storage device (15),

CHARACTERIZED IN THAT

said storing step (100) comprises the further step of indexing said records such that each record is retrievable by reference to the filename of the corresponding Web component;

e) proceeding hierarchically from lowest sub-directories to highest directories, compiling for each sub-directory and directory a respective record of collective usage information pertaining to all Web components associated with such sub-directory or directory;

f) storing said collective-hit-information records in the filtered-hit-record data storage device (15), said storing step comprising the further step of indexing each said record such that it is retrievable by reference to the name of the corresponding sub-directory or directory; and

h) [sic] organizing said filtered-hit records and said collective-hit records as a further Web site."

The only claim of the first auxiliary request reads as follows:

"A method for building a database of information about visits to a Web site, here denominated a primary Web site, comprising text, image, and other basic data building blocks residing in files accessible via their respective URL, said building blocks, which are here denominated Web components, organized in a file directory structure having directories (180) and sub-directories (180.1 - 180.3), the method comprising:

a) retrieving (95) data from a log file, said data here denominated raw hit records, each raw hit record pertaining to a browser request for a particular Web component of the Web site;

b) extracting (85) selected information from said raw hit records, thereby to create, from each raw hit record, a filtered hit record, the selected information including information identifying the Web component to which the respective hit record pertains;

c) distinguishing (85) respective filtered hit records according to the visit to which each of them belongs, and associating (85) a particular visit index with each filtered hit record;

d) storing (100) the filtered hit records in a data storage device (15),

CHARACTERIZED IN THAT

e) the storing step (100) comprises organizing the filtered hit records in a file structure of directories (180') and subdirectories (180-1'-180.3') which is corresponding to the file structure of the Web components to which the filtered hit records pertain;

f) the method further comprises compiling and storing collective usage information (190) for each directory and subdirectory of the hit-record file structure and organizing the collective usage information in said file structure;

g) the filtered hit records and collective usage information are organized so as to be retrievable via their respective file names, subdirectory names or directory names in said file structure, and

h) the method further comprises making the organized filtered hit records and collective usage information available as a secondary Web site such that if the primary and secondary Web sites are displayed concurrently in a first and a second window on a user terminal, each selection of a component of the primary Web site causes information to be displayed in the second window that describes visits in which the selected component was requested; and at least one selection of visit information from the secondary Web site causes a corresponding visited Web component to be displayed in the first window."

The text of the claim according to the second auxiliary request is not relevant to the outcome of the Board's decision.

XII. The appellant's arguments in support of the main request may be summarized as follows:

Throughout the proceedings of the present application a number of changes had been introduced which led to new objections under Articles 84 and 123(2) EPC. The claim of the new main request was essentially based on the original wording of claim 1, with minor amendments to overcome the clarity objections raised in the Examining Division's first communication dated 26 March 2003. In the present case, the appellant should be allowed to re-submit a previous claim version because it was hardly possible to overcome all the outstanding objections by further amending a claim that had already undergone many unsuccessful amendments. Furthermore, the claim of the new main request recited all the essential features of the invention, while avoiding objectionable language.

Reasons for the Decision

1. The appeal is admissible.

2. As indicated in the description (see page 1, first paragraph of the application as filed), the present "invention relates to tools, for use, e.g., by a content provider for a Web site, for summarizing and displaying information descriptive of usage patterns exhibited by visitors to the Web site".

2.1 The gist of the present invention consists essentially in organising a database of usage information concerning "Web components", i.e. essentially files, accessible at a Web site ("first Web site") so that the usage record relating to a Web component is retrievable from another Web site ("second Web site") under a file name corresponding to the Web component's name in the directory of the first Web site. Furthermore, by making the usage information accessible through the second Web site, a browser can easily display both a Web component and the corresponding statistics file.

Admission of the main request

3. In the letter dated 23 September 2015, the appellant noted that the new main request aimed at overcoming objections under Article 123(2) EPC which had resulted from amendments made to the original claim 1 in the first instance proceedings while seeking to overcome the Examining Division's clarity objections.

3.1 According to the appellant's submissions, the only claim of the main request was thus based on the original wording of claim 1 and included only minor amendments, such as a clarification of the term "Web component" and of "a raw hit record" which was said to comprise a "logfile". Both these amendments were clearly supported by the description as originally filed and had been occasioned by the clarity objections raised in the communication dated 26 March 2003 (see point 2.2).

4. With respect to claim 1 as originally filed, the only claim of the main request contains the following additions:

- "a Web component comprising a Webpage, or a basic data building block such as text, hypertext, images, embedded client software programs or other data displayable by a browser residing in a file which is accessible through its URL", inserted after "sub-directories,"­ in the first paragraph;

- "a raw hit record comprising a logfile", inserted after "retrieving (95) raw hit records," in feature a).

4.1 The history of the examination procedure reveals that, in reply to the objections under Articles 84 and 56 EPC raised against the original claims in the communication dated 26 March 2003, the applicant submitted a new claim 1 with letter dated 28 July 2003 and observed that claim 1 had been rewritten more clearly to highlight those features of the invention that the inventors believed to be novel.

In that letter the applicant identified the following features of the then claimed invention as being novel with respect to the prior art:

(i) according to the claimed method, data derived from the log file, "filtered hit records", are organised in a file structure of directories and sub-directories which is parallel to the file structure of the Web components to which the filtered hit records pertain;

(ii) the claimed method further includes the compilation and storage of collective usage information for each directory and sub-directory of the hit-record file structure and the organisation of the collective usage information in the hit-record file structure;

(iii) the filtered hit records and collective usage information are organised so as to be retrievable via their respective file names, sub-directory names, or directory names in the hit-record file structure; and

(iv) according to the claim method, the organised filtered hit records and collective usage information are made available as a secondary Web site having the property that when the primary and secondary Web sites are displayed concurrently on a user terminal, user selection of a Web component of either Web site can cause the automatic selection of a corresponding Web component of the other Web site.

4.2 Accordingly, the characterising part of claim 1 filed with letter dated 28 July 2003 comprised the following features:

(e) the storing step comprises organizing the filtered hit records in a file structure of directories and sub-directories which is parallel to the file structure of the Web components to which the filtered hit records pertain;

(f) the method further comprises compiling and storing collective usage information for each directory and sub-directory of the hit-record file structure and organizing the collective usage information in said file structure;

(g) the filtered hit records and collective usage information are organized so as to be retrievable via their respective file names, sub-directory names, or directory names in said file structure, and

(h) the method further comprises making the organized filtered hit records and collective usage information available as a secondary Web site having the property that when the primary and secondary Web sites are displayed concurrently on a user terminal, user selection of a Web component of either Web site can cause the automatic selection of a corresponding Web component of the other Web site.

4.3 In contrast to claim 1 of the request submitted with the letter dated 28 July 2003, the claim according to the present main request does not contain the features which, in the applicant's opinion expressed in that letter, distinguished the claimed invention from the prior art document D1. In particular it does no longer specify any features relating to the display of usage information. Nevertheless, in the letter dated 23 September 2015, the appellant essentially reiterated the novelty and inventive step arguments submitted in the first instance proceedings with the letter dated 28 July 2003, and thus relied on features which were no longer recited in the claim of the main request.

4.4 The Board acknowledges that in the present case the outstanding objections of added subject-matter (Article 123(2) EPC) were essentially directed against amendments which the applicant had made while attempting to overcome clarity objections raised in the first instance proceedings, and that, by discarding undisclosed features, it may be inevitable to arrive at a claim version similar to what had already been presented at an early stage of the first-instance proceedings. However, there is no justification in the present case for submitting, at a late stage of the appeal proceedings, a request which no longer contains those limiting features which the applicant had added in order to overcome the Examining Division's objection of lack of inventive step: an objection which the applicant appears not to have contested with respect to original claim 1 and which was no longer raised against the subsequent requests submitted to the Examining Division, which contained the above mentioned limiting features.

4.5 In view of the above, the Board decided to make use of its discretion under Rule 13(1) RPBA and not to admit the appellant's main request into the appeal proceedings.

First auxiliary request

5. Claim 1 according to the first auxiliary request submitted by the appellant at the oral proceedings at 14:15 essentially corresponded to claim 1 filed with letter 28 July 2003 in reply to the Examining Division's communication of 26 March 2003.

5.1 In a communication dated 14 September 2004, the Examining Division noted that the claim filed with letter dated 28 July 2003 seemed to meet the requirements of Article 52(1) EPC with respect to the available prior art. However, amendments were required to overcome objections under Article 123(2) and 84 EPC.

In particular, the Examining Division considered that the term "automatic selection" violated Article 123(2) EPC and that instead of "automatic selection", "retrieving and displaying a corresponding Web component of the Web site" appeared to be meant.

Furthermore, the Examining Division found that the terms "collective usage information" and "file directory structure" were unclear (Article 84 EPC).

5.2 At the oral proceedings, the Board considered that the claim proposed as first auxiliary request could be regarded as a promising basis for further examination since it corresponded to a claim which had overcome some of the objections raised by the Examining Division and in particular the objection under Article 56 EPC.

Although the wording of this claim was substantially different from the wording of the claim considered in the contested decision, the Board accepted the appellant's argument that it was difficult to overcome objections under Articles 123(2) and 84 EPC raised against a claim which, as a result of repeated amendments, had considerably departed from the original wording without effectively resubmitting some earlier claim version.

5.3 The Board also accepted some further minor amendments which aimed to improve the clarity of the claim wording and resulted in the claim according to the first auxiliary request now on file.

5.4 Finding that the claim filed as first auxiliary request at the oral proceedings prima facie addressed the Articles 123(2) and 56 EPC objections, raised by the Examining Division in the communications dated 26 March 2003 and 14 September 2004, and that the gist of the now claimed invention did not diverge in essence from the subject-matter submitted with the statement of grounds of appeal, the Board decided to admit it into the appeal proceedings pursuant to Article 13(1) RPBA.

Article 123(2) EPC

6. The first paragraph of the claim according to the first auxiliary request corresponds to the first paragraph of claim 1 as originally filed with the addition that the Web site is denominated "a primary Web site, comprising text, image, and other basic data building blocks residing in files accessible via their respective URL", and that the building blocks are "denominated Web components".

Corresponding definitions of Web site and Web component can be found in the application as originally filed (page 1, lines 10 to 13 and line 17 to page 2, line 2; page 5, first full paragraph; page 7 "Web component" and page 8 "Web page").

6.1 Step a) differs from step a) as defined in the claim as originally filed in that it specifies that data "denominated raw hit records" are "data from a log file".

The above feature finds support on page 2 of the application as filed, lines 13 to 17 and on page 12, lines 16 to 19. An example of the information stored in a logfile can also be found at page 2.

6.2 Steps b) and c) correspond to the same steps of claim 1 as originally filed.

6.3 Claim 1 as originally filed also discloses the step of storing the filtered hit records in a data storage device (step d)).

6.4 As to feature e) of claim 1, section 2.3 "Building the shadow directory" on page 16 of the original application explains that the "visit database" is processed into a new database, resident on a storage device, which is defined as "shadow-directory database".

Furthermore, it is specified in section 2.3, page 16, from line 17 onwards that "[i]t is a general characteristic of files containing Web components that they are organized in a hierarchical file structure. A correspondence to this hierarchical file structure is built into the shadow-directory database, such that its records are retrievable by the names of the corresponding files on the Web server.

By way of example, this correspondence can be achieved by using the same filenames, in the shadow-directory database, as the corresponding Web-page files. Alternatively, as will be appreciated by those skilled in the art, such a correspondence is readily achieved using well-known data-base technologies to relate files of the shadow-directory database to corresponding Web-page files.

As a consequence, each file in the Web-site directory will have a unique counterpart in the shadow-directory database. The correspondence described above is useful for facilitating on-line retrieval of information from the shadow-directory database by operating it as a parallel Web site" (underlining added).

An example of this file structure is shown in Figure 7.

In the Board's opinion, feature e) reflects the correspondence between the hierarchical file structure of the Web components and the usage information stored in the "shadow-directory database" disclosed in the description.

6.5 The step f) of compiling and storing "collective usage information" for each directory and subdirectory of the hit-record file structure recited in claim 1 corresponds to steps e) and f) (in part) of claim 1 as originally filed.

6.6 Feature g), which relates to the organisation of filtered hit records and collective usage information so as to be retrievable via their respective file, sub-directory and directory names in the file structure corresponds essentially to feature f) of claim 1 as originally filed.

6.7 Feature h) of claim 1 relates to the way the organised filtered hit records and collective usage information can be retrieved and displayed. In particular, feature h) covers the following aspects:

- the filtered hit records and collective usage information are made available as a secondary Web site;

- if the primary and secondary Web sites are displayed concurrently in a first and a second window on a user terminal, the selection of a Web component of the primary Web site in the first window causes the display in the second window of information relating to visits involving requests for that component, and similarly

- a selection of visit information from the secondary Web site causes a corresponding Web component to be displayed.

The wording of feature h) is essentially taken from dependent claim 3 as originally filed.

6.8 Strictly speaking, claim 3 depended on claim 2 and not on claim 1. Claim 1 related to a method for building a database whereas claim 2 was concerned with a method for navigating first and second Web sites. Indeed, feature h) establishes a link between the way the database is structured and the way the corresponding information can be displayed and specifies in particular how a Web component and the corresponding usage information are synchronized, as far as their display is concerned.

It is evident to the skilled person that these are two aspects of the same invention. Furthermore, this correspondence is clearly illustrated in Figure 7 as originally filed and constitutes an important feature of the invention.

6.9 In summary, the Board is satisfied that claim 1 according to the first auxiliary request does not contain subject-matter extending beyond the content of the application as originally filed (Article 123(2) EPC).

Article 84 EPC

7. In the Board's opinion, the claim according to the first auxiliary request comprises all the essential features of the invention.

7.1 In the communication dated 14 September 2004, sent in response to the filing on 28 July 2003 of a new single claim, the Examining Division objected that the term "collective usage information" was unclear and that the term "file structure" was vague.

7.2 According to the Board, however, claim 1 clearly specifies that the file structure of the Web component of a Web site and the file structure of the usage information are the same, and that the usage information utilises the same file name as the file of the corresponding Web component. For directories and sub-directories it may be necessary to define different names. This is, however, sufficiently explained in the description (see Figure 7 and page 19, lines 11 to 19).

8. In the course of the examination proceedings, the Examining Division also pointed out that the term "visit" was vague and left the reader in doubt as to the meaning of the technical feature to which it referred.

8.1 According to the appellant, however, there was no need to add a definition of "visit" to the claim. This term was well-known in the art and, therefore, the skilled reader would understand the meaning of "visit" without requiring any explicit definition.

9. As specified in the claim, the present invention relates to a method for building a database about visits to a Web site. The description specifies at page 2, lines 6 to 7 that a "visit to a Web site is defined as a series of downloads, from a specified Web server by a fixed client browser, that are contiguous in time".

As pointed out at page 3, first paragraph, "a visit to a given Web site begins with an initial request to the Web server (the entry point), consists of a number of consecutive downloads, and ends when the visitor either: (i) begins to request pages from a different Web site, or (ii) stops browsing altogether. The visitor's final request is referred to as the exit point".

9.1 As to the problem of identifying a "visit" and the "hit records" which belong to it, the appellant essentially argued that there were numerous well-known methods for inferring visits to a Web site, some of them listed in the description. In fact, how to identify individual visits was not an essential aspect of the invention. As the skilled person would appreciate that other methods not disclosed in the application might also be used, any attempt to define the precise information to be used for inferring visits would unduly limit the scope of the claim.

9.2 Section 2.2 of the description ("Identifying visits") specifies that several mechanisms can be used for grouping hits into visits and provides a number of examples which make use of different data.

9.3 The Board is satisfied that the meaning the skilled reader would attribute to "visit" in the context of claim 1 corresponds to the definition given to this term in the application. Furthermore, in the Board's opinion, it can be left to the skilled person, facing the task of implementing the claimed invention, to give the general concept of "visit to a Web site", as expressed in the claim, an appropriate "technical" definition which specifies how to identify a "visit" and to link "hit records" and "visits".

10. In a communication accompanying the summons to oral proceedings, the Examining Division noted that the terms "filtered hit records" and "collective usage information" were unclear and left the reader in doubt as to the technical meaning of the technical features to which they referred. In particular, the Examining Division expressed doubts as to the difference between such features and how data was obtained and where it was stored. Other unclear terms in the Examining Division's opinion were "building blocks" and "Web component" which had no well-recognized technical meaning.

10.1 The terms and expressions of the present claim which the Examining Division had found unclear occur in the original claim 1 and, in the Board's opinion, are sufficiently clear to the skilled reader, as they carry their customary meanings. For example, with respect to the expression "filtered hit records" it is specified in section 2.1 ("Filtering") of the description that every hit is typically logged by the Web server. "However, it is often advantageous to retain only selected lines of the log for further processing. (In particular, this reduces the volume of data that is subsequently manipulated and stored, and thus improves economy and speed.) Such selection can be carried out on any of the fields in the log. As is well known to those skilled in the art, such selection is readily specified in a configuration file. It is not critical where this filtering is carried out" (page 13, lines 5 to 10). Then follows a list of fields "filtered" from the logfile which may be of interest.

11. In summary, the Board considers that claim 1 according to the first auxiliary request is clear and supported by the description in accordance with Article 84 EPC.

Further issues

12. According to the view expressed by the primary examiner in the communication dated 14 September 2004, the claim filed with letter dated 28 July 2003 seemed to meet the requirements of Article 52(1) EPC "with respect to the available prior art". In fact, it appears from the file that in the remaining first-instance proceedings the question of novelty and inventive step was never again addressed.

However, in the communication dated 3 February 2009, accompanying the summons to oral proceedings, the Examining Division explicitly stated that if the objections under Articles 123(2) and 84 EPC were overcome by a new request, this request would have to be examined with respect to Articles 52(1), 54(1) and (2), and 56 EPC.

12.1 In these circumstances, the Board deems it appropriate to exercise its power under Article 111(1) EPC and remit the case to the department of first instance for further prosecution on the basis of the appellant's first auxiliary request. There is consequently no need to deal with the appellant's second auxiliary request.

Order

For these reasons it is decided that:

1. The decision under appeal is set aside.

2. The case is remitted to the department of first instance fur further prosecution.