Ex Parte Butler et al

Patent Trial and Appeal Board

Dec 13, 2012

11261752 (P.T.A.B. Dec. 13, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte MARK HENRY BUTLER, DAVID MURRAY BANKS,
SCOTT ALAN STANLEY, and RUEL CALALO
____________

Appeal 2010-007095
Application 11/261,752
Technology Center 2100
____________

Before JOHN A. JEFFERY, DENISE M. POTHIER, and
JENNIFER L. McKEOWN, Administrative Patent Judges.

JEFFERY, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner’s
rejection of claims 1-22 and 25. Claims 23 and 24 have been withdrawn
from consideration. App. Br. 5. We have jurisdiction under 35 U.S.C.
§ 6(b). We affirm.

STATEMENT OF THE CASE
Appellants’ invention indexes data in a directed graph that includes
nodes and associated properties. An algorithm indexes each indexable
Appeal 2010-007095
Application 11/261,752

2
property pointing to a resource as related to respective indexable nodes,
where the indexable property is attached through a chain of indexable links
from child to parent through intermediate nodes, all of which are
non-indexable. See generally Abstract; Spec. 5-9; Figs 1-2. Claim 1 is
illustrative:
1. A computer-implemented method of indexing for a content
repository, the method comprising:
providing a content repository for storing a plurality of nodes in the
form of a directed graph including at least one directed link, each directed
link linking a parent node to a child node;
providing items of information data, each item of information data
being attached to a respective node;
identifying each of the nodes as either an indexable node or a non-
indexable node;
identifying each of the items of information data and each of the
directed links as indexable or non-indexable, wherein at least one non-
indexable node has an indexable item of information attached to said at least
one non-indexable node;
indexing each item of indexable information data as relating to its
respective node if the respective node is indexable, and otherwise indexing
each item of indexable information data as relating to anyone or more
indexable nodes that are linked to its respective node starting from the
indexable node and linked to the respective node through a chain of one or
more indexable directed links in the direction from parent node to child node
without passing through any indexable nodes.

THE REJECTIONS
1. The Examiner rejected claims 1-9, 11-15, and 17-22 under 35
U.S.C. § 103(a) as unpatentable over Sedlar (US 6,427,123 B1; July 30,
2002) and Chinnappan (US 2002/0082932 A1; June 27, 2002). Ans. 4-39.1

1 Throughout this opinion, we refer to (1) the Appeal Brief filed September
15, 2009 (“App. Br.”); (2) the Examiner’s Answer mailed December 9, 2009
(“Ans.”); and (3) the Reply Brief filed January 28, 2010 (“Reply Br.”).
Appeal 2010-007095
Application 11/261,752

3
2. The Examiner rejected claims 10 and 16 under 35 U.S.C. § 103(a)
as unpatentable over Sedlar, Chinnappan, and Dutta (US 2002/0138471 A1;
Sept. 26, 2002). Ans. 39-41.
3. The Examiner rejected claim 25 under 35 U.S.C. § 103(a) as
unpatentable over Friedman (US 2004/0210552 A1; Oct. 21, 2004) and
Guay (US 5,907,844; May 25, 1999). Ans. 41-43.

THE OBVIOUSNESS REJECTION OVER SEDLAR AND CHINNAPPAN
The Examiner finds that Sedlar’s indexing method for a content
repository has every recited feature of representative claim 1 including a
repository for storing nodes in the form of a “directed graph” (file system
hierarchy in Figure 1) including at least one “directed link,” where each
directed link is said to be identified as indexable or non-indexable.
According to the Examiner, the Row ID in Sedlar’s hierarchical index in
Figure 5 supplies an “indexable directed link” between parent and child
items, and the index’s excluding items without children teaches
“non-indexable” directed links. Ans. 4-5, 44-48. Although the Examiner
acknowledges that Sedlar lacks at least one non-indexable node having an
attached indexable item, the Examiner cites Chinnappan as teaching this
feature in concluding that the claim would have been obvious. Ans. 6-7,
48-49.
Appellants argue that the Examiner not only “inappropriately mixes
and matches” elements from Sedlar’s relational database embodiment with
Sedlar’s prior art Figure 1 directed to a file system hierarchy, Sedlar’s
emulating this hierarchy via a database actually teaches away from a
directed graph as claimed. App. Br. 25-27; Reply Br. 2-4. Appellants add
Appeal 2010-007095
Application 11/261,752

4
that the cited prior art does not teach or suggest identifying each directed
link as indexable or non-indexable, let alone that at least one non-indexable
node has an attached indexable item as claimed. App. Br. 28-29; Reply Br.
5-6.2

ISSUE
Under § 103, has the Examiner erred in rejecting claim 1 by finding
that Sedlar and Chinnappan collectively would have taught or suggested an
indexing method for a content repository for storing nodes in the form of a
directed graph including at least one directed link, where (1) each link is
identified as indexable or non-indexable, and (2) at least one non-indexable
node has an attached indexable item?

ANALYSIS
Claims 1-9, 11-15, 17, 18, and 20-22
We begin by noting that it is undisputed that Sedlar’s Figure 1,
labelled as prior art, pertains to a conventional file system hierarchy, but
Sedlar’s Figure 5 shows a relational database that emulates this hierarchy.
See App. Br. 25; Ans. 44-45 (acknowledging this distinction). Nor is it
disputed that the Examiner relies on both figures in connection with the
disputed limitations of claim 1. See Ans. 4-5, 44-48. Nevertheless, we see
no error in this reliance since Sedlar’s Figures 1 and 5 are related, at least

2 Since Appellants withdrew the argument regarding providing items of
information data in view of the Examiner’s explanation in the Answer
(Reply Br. 5), we need not consider that argument here.
Appeal 2010-007095
Application 11/261,752

5
with respect to Figure 1’s particular items and hierarchical relationships that
are emulated in Figure 5 as the Examiner indicates. Ans. 44-45.
Sedlar describes two types of models for hierarchically-organized
systems: (1) a “tree” where each file has only one parent, and (2) a “directed
graph” where files can have multiple parents. Sedlar, col. 2, ll. 18-26. Even
assuming, without deciding, that Sedlar’s file system in Figure 1 shows a
“tree” system that is emulated via a relational database in Figure 5 as
Appellants contend (App. Br. 27; Reply Br. 2), we nonetheless find that
similarly emulating a “directed graph” would have been an obvious
variation, particularly since the only difference between “directed graph”
and “tree” systems is the number of parents.
Although emulating a directed graph via a database is not an actual
directed graph, nothing in the claim precludes this emulation, for the claim
merely recites, in pertinent part, that the repository stores a “directed graph”:
a limitation that is fully met by either an actual or emulated graph.
Appellants’ arguments regarding Sedlar’s alleged teaching away from
hierarchical systems (App. Br. 26-27; Reply Br. 3-4) are unavailing, for they
are not commensurate with the scope of the claim which does not preclude
emulated directed graphs using relational databases.
Nor are we persuaded of error in the Examiner’s finding that Sedlar’s
Row ID supplies an indexable directed link between parent and child, and
therefore at least suggests identifying indexable directed links. Ans. 47-48;
see also Sedlar, col. 7, ll. 6-45; Fig. 5. As the Examiner explains, the Row
ID “Y2” represents the “Windows” directory 114 which is the parent
directory to the “Word” and “Access” directories 116, 120 in the associated
“Dir Entry” list. Ans. 47-48 (citing Sedlar, col. 7, ll. 26-32); see also Sedlar
Appeal 2010-007095
Application 11/261,752

6
Fig. 1 (graphically showing this hierarchy). This relationship between
parent and children directories or nodes represented by the index data
effectively hierarchically links the nodes and therefore at least suggests a
directed link as claimed. Appellants’ contention that Sedlar’s hierarchical
index entries are nodes and not links (App. Br. 28; Reply Br. 5) is unavailing
since this argument ignores hierarchical links established between these
nodes represented by data in rows of the index. These “directed links” are
therefore “indexable” at least to the extent that they are represented by
Sedlar’s index. This “indexable” alternative therefore fully meets
identifying directed links as “indexable” or “non-indexable,” for the prior art
need only teach one of these alternatives to meet the limitation as the
Examiner indicates. See Ans. 48 (interpreting directed links identification
limitation “as requiring the directed links to be either indexable or non-
indexable” (emphases added)).
Although Appellants attempt to distinguish “indexable” from
“indexed” and “non-indexable” from “not indexed” by noting, among other
things, that an “indexable” item may not be indexed (Reply Br. 5-6), the fact
remains that all indexed items are “indexable”; otherwise, they could not be
indexed. Thus, identifying an indexed directed link in Sedlar likewise
identifies an “indexable” link. Although all items that are not indexed are
not necessarily “non-indexable” as Appellants indicate since the items may
not yet be indexed or deliberately omitted from the index (Reply Br. 6),
“non-indexable” items—like the indexable items noted by Appellants—
nonetheless are a subset of items that are not indexed. Since Sedlar does not
index directories without children as the Examiner indicates (Ans. 48 (citing
Sedlar, col. 7, ll. 15-23)), these nodes are “non-indexable” at least to the
Appeal 2010-007095
Application 11/261,752

7
extent that they do not comport with Sedlar’s scheme that indexes only
parent nodes.
We also see no error in the Examiner’s reliance on Chinnappan
merely to show that associating indexable information items with
non-indexed features is known in the art. Ans. 48-49 (citing Chinnappan
¶¶ 0028, 0065; Fig. 5). Even assuming, without deciding, that non-indexed
features in Chinnappan are indexable as Appellants assert (Reply Br. 7
(citing Chinnappan ¶¶ 0066-67)), we are still unpersuaded of error in the
Examiner’s reliance on Chinnappan merely to show that associating
indexable items with non-indexed items is known in the art, and that Sedlar
and Chinnappan collectively at least suggest the disputed limitations.
We are therefore not persuaded that the Examiner erred in rejecting
representative claim 1, and claims 2-9, 11-15, 17, 18, and 20-22 not
separately argued with particularity.

Claim 19
Although Appellants nominally argue claim 19 separately (App. Br.
29), Appellants reiterate similar arguments made in connection with claim 1
which we find unpersuasive for the reasons previously discussed.

THE OBVIOUSNESS REJECTION OVER SEDLAR, CHINNAPPAN, AND DUTTA
We also sustain the Examiner’s obviousness rejection of claims 10
and 16. Ans. 39-41. Despite nominally arguing these claims separately,
Appellants reiterate similar arguments made in connection with claim 19
(App. Br. 30) which we find unpersuasive for the reasons previously
discussed.
Appeal 2010-007095
Application 11/261,752

8

THE OBVIOUSNESS REJECTION OVER FRIEDMAN AND GUAY
The Examiner finds that Friedman discloses every recited feature of
claim 25 including associating a first set’s3 indexable properties with their
respective indexable owner nodes, namely by defining particular
relationships between subject and object nodes and their associated
predicates. Ans. 41-43, 50-53. Although the Examiner acknowledges that
Friedman does not associate a second set’s indexable properties with
respective indexable ancestors of the respective owner node and not the
owner node itself, the Examiner cites Guay for teaching this feature in
concluding that the claim would have been obvious. Id.
Appellants argue that not only is there no motivation to modify
Friedman with Guay’s teachings as the Examiner proposes, the proposed
combination still fails to teach an index with entries associating the second
set’s indexable properties with respective indexable ancestors of the
respective owner node and not the owner node itself as claimed. App. Br.
31-33; Reply Br. 8-10.

ISSUES
(1) Under § 103, has the Examiner erred in rejecting claim 25 by
finding that Friedman and Guay collectively would have taught or suggested
an index with entries associating a second set’s indexable properties with
respective indexable ancestors of the respective owner node and not the
owner node itself?

3 We note in passing that no antecedent basis exists for “said first set” and
“said second set” in claim 25.
Appeal 2010-007095
Application 11/261,752

9
(2) Is the Examiner’s combining the teachings of these references
supported by articulated reasoning with some rational underpinning to
justify the Examiner’s obviousness conclusion?

ANALYSIS
On this record, we find no error in the Examiner’s obviousness
rejection of claim 25. First, Friedman’s Resource Description Framework
(RDF) model represents metadata about Web resources (e.g., title, author,
etc.) by defining relationships between web identifiable objects, namely by
using (1) a subject; (2) predicates; and (3) objects mapped onto a directed
graph of nodes. Friedman ¶¶ 0002-04, 0015-17; Figs. 1-2. We emphasize
“directed graph” here, since, as noted previously, this is a particular form of
hierarchy where a child can have more than one parent, unlike a tree-type
hierarchy as Appellants acknowledge. Reply Br. 8. Leaving aside
Appellants’ recognition of Friedman’s directed graph hierarchy runs counter
to their earlier assertion that Friedman does not disclose hierarchies
(compare Reply Br. 8 with App. Br. 31-32), the fact remains that Friedman’s
scheme is hierarchical. Accord Friedman ¶ 0015 (noting that the directed
graph structure may only proceed from subject node 101 to other nodes).
Given this hierarchical scheme, we see no error in the Examiner’s
reliance on Guay merely to show that when a node in a hierarchy lacks
needed data, it is known to use data from ancestor nodes in that
circumstance essentially as a backup. See Ans. 51-53; see also Guay,
col. 3, ll. 31-54; col. 6, l. 29 – col. 7, l. 55; col. 16, l. 63 – col. 17, l. 32; Figs.
3, 8. Although this data is in the form of rules that are inherited from
ancestor nodes—rules that are not used in Friedman’s hierarchy as the
Appeal 2010-007095
Application 11/261,752

10
Examiner acknowledges (Ans. 51)—we nonetheless agree with the
Examiner that skilled artisans would understand that Guay fundamentally
teaches inheriting and using data from ancestor nodes when needed. As the
Examiner indicates (Ans. 52), this fundamental concept of inheriting data
from ancestor nodes applies regardless of whether Guay’s inherited data is in
the form of rules, or otherwise, since data is inherited when needed.
We therefore see no reason why this concept could not be applied to
Friedman’s hierarchical system as the Examiner proposes (Ans. 41-43; 50-
53), for this enhancement merely predictably uses prior art elements
according to their established functions—an obvious improvement. KSR,
550 U.S. at 417. Although inheriting data from ancestor nodes in a directed
graph hierarchy may be more complicated than a similar scheme in a tree-
type hierarchy since certain nodes may have multiple parents, Appellants
have nonetheless provided no persuasive evidence proving that modifying
Friedman as proposed would render it unsuitable for its intended purpose.
And where, as here, if a technique has been used to improve one device (i.e.,
inheriting data from ancestor nodes in a hierarchy when needed), and an
ordinarily skilled artisan would recognize that it would improve similar
devices in the same way (i.e., using inherited data in a directed graph
hierarchy when needed), using the technique is obvious unless its actual
application is beyond his or her skill. See id.
We are therefore not persuaded that the Examiner erred in rejecting
claim 25.

CONCLUSION
The Examiner did not err in rejecting claims 1-22 and 25 under § 103.
Appeal 2010-007095
Application 11/261,752

11

ORDER
The Examiner’s decision rejecting claims 1-22 and 25 is affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED

babc