Ex Parte Waddington

Patent Trial and Appeal Board

Sep 23, 2013

11093517 (P.T.A.B. Sep. 23, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte DANIEL G. WADDINGTON
____________

Appeal 2011-001984
Application 11/093,517
Technology Center 2100
____________

Before JOSEPH L. DIXON, ST. JOHN COURTENAY III, and
CARLA M. KRIVAK, Administrative Patent Judges.

DIXON, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2011-001984
Application 11/093,517

2
STATEMENT OF THE CASE
Appellant appeals under 35 U.S.C. § 134(a) from a rejection of claims
1-18. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.

INVENTION
The invention is directed to “N-way merging of source code into a
single, unified and semantically correct version” (Spec. 1:16-17). Claim 1,
reproduced below, is representative of the claimed subject matter:
1. A method for performing conditionalized N-way
merging of source code, comprising:

collecting differential information for each of N versions
of an original source code program, the original source code
program having lines of text;

constructing a merge data structure from the differential
information; and

merging the N versions into a single, semantically
correct, unified program using the merge data structure, the
single, semantically correct, unified program having
preprocessing conditionals to capture differences between the N
versions.

REFERENCES
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Livshits US 2004/0230886 A1 Nov. 18, 2004

Andreas Zeller, “Configuration Management with Version Sets: A Unified
Software Versioning Model and its Applications,” April 1, 1997.
Appeal 2011-001984
Application 11/093,517

3
REJECTION
Claims 1-18 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Livshits and Zeller.

ANALYSIS
Claims 1-17
Appellant contends that neitherLivshits, Zeller, nor the combination of
the two references, discloses or would have suggested “merging the N
versions into a single, semantically correct, unified program using the merge
data structure, the single, semantically correct, unified program having
preprocessing conditionals to capture differences between the N version,” as
recited in claim 1 (App. Br. 9-15). We disagree.
The Examiner cites Zeller’s Abstract (Ans. 4-5, 11), which discloses:
Unified versioning. Version sets provide one single
formalism to express all versioning dimensions as well as
constraints on them, integrating SCM [software configuration
management] concepts like revisions, variants, workspaces, and
configurations in one single model. . . .

. . . .

We have implemented the version set model in an
experimental SCM system name ICE for Incremental
Configuration Environment. In ICE, the version set model
shows up numerous user-visible benefits. Through the FFS, a
virtual file system, users can access version sets consisting of
arbitrary combinations of revisions, changes, variants, and
workspaces. Individual versions are accessed as files; version
sets as a whole can be handled via version directories or
through the well-known C preprocessor representation.

Appeal 2011-001984
Application 11/093,517

4
(Zeller, Abstract). Considering Zeller as a whole to determine how Zeller
handles version sets with the C preprocessor, as quoted above, we find
Zeller discloses the following:
A completely different “repository” concept used for
variant management is realized in the C programming language.
All variants are stored in a single component visible to the
programmer; variant-specific parts are enclosed in C
preprocessor (CPP) #if . . . #endif directives. As part of the
compilation, CPP selects a single variant from the source code
determined by a conjunction of attribute/value pairs. CPP
evaluates each #if-expression, with any attribute being replaced
by its respective value. The code piece enclosed by the #if . . .
#endif is included only if the #if-expression evaluates to a non-
zero value.

. . . .

It is common to see conditional compilation, as
exemplified by CPP, as a programming language feature. In
our context, conditional compilation should rather be regarded
as a compiler- and language-independent version-control
technique.

(Zeller, p. 15, sec. 2.6.1). Thus, Zeller discloses handling version sets by
using C preprocessor directives, i.e., “conditionals” as recited in claim 1, to
determine which lines of code from which versions are merged into the
program to be compiled. We agree with the Examiner (Ans. 5, 12-13) and
find this meets the claim 1 limitation of a “unified program having
preprocessing conditionals to capture differences between the N versions.”
The Examiner also cites Zeller’s discussion of semantics-based
merging (Ans. 5, 12), which discloses the following:
While syntax-based merging guarantees the syntactic
correctness of the merge result M, one still has no guarantee
Appeal 2011-001984
Application 11/093,517

5
about how the execution behavior of M relates to the execution
behavior of the merged variants V1 and V2. . . . [T]he first
approach that performed true semantics-based merging was
presented by Horwitz, Prins, and Reps in [HPR89]. Their
algorithm relies on the assumption that behavior differences,
rather than textual or structural differences, are significant and
must be preserved in M.
. . . To determine interference of changes, the algorithm
determines the program slices in V1 and V2 that are changed
from the base V0 and the slices that are unchanged from V0.
The changed and unchanged slices are then merged, and if there
is no interference, a merged program M is produced from the
merged slices. The algorithm ensures that M captures the
changed behavior of both V1 and V2 as well as the behavior that
was unchanged from the base V0.

(Zeller, p. 51, sec. 5.6.3). Thus, Zeller discloses semantics-based merging
approaches exist in order to maintain the semantic integrity of a merged
program. This suggests merging various versions of a program such that the
resulting unified program is semantically correct, as recited in claim 1.
Appellant additionally argues the Examiner’s motivation for
combining the references is deficient (App. Br. 15-16). In stating why it
would have been obvious to combine the references, the Examiner cites
Zeller, page 195, which discloses:
ICE [Zeller’s inventive incremental configuration
environment] uses CPP [C preprocessor] expressions to
represent feature terms and files enriched with CPP directives
to represent version sets. The intent is to give users a familiar,
well-understood representation of multiple items in one
representation. . . .
The primary advantage of the CPP format is that only the
differences between versions are specified.

Appeal 2011-001984
Application 11/093,517

6
(Zeller, p. 195, sec. 16.6). In other words, Zeller discloses using C
preprocessor directives to represent differences between version sets makes
it easy and clear for a user to assess the conflicts between different versions
of a program. Appellant has not explained why one of ordinary skill in the
art would not have been motivated to modify Livshits in view of Zeller’s
specific disclosure of the advantages of using C preprocessor directives.
Rather, Appellant states the Examiner merely “articulates the result of
combining Livshits with Zeller” and the Examiner’s “conclusory statement
is unaccompanied by evidence or reasoning” (App. Br. 15-16). This
argument, however, overlooks the explicit motivation in Zeller to use C
preprocessor directives for software version control as discussed above.
We are therefore not persuaded the Examiner erred in rejecting
representative claim 1, and claims 2-17 not specifically argued separately.

Claim 18
Appellant contends that “claim 18 utilizes ‘means for’ recitations, and
as such requires the Examiner to consider the specific structure described in
the specification (e.g., FIG. 2) to interpret these limitations” (App. Br. 16).
“The structure disclosed by the Appellant[] cannot be disregarded” (App. Br.
17). We agree with Appellant.
Means-plus-function claims “shall be construed to cover the
corresponding structure, material, or acts described in the specification and
equivalents thereof.” 35 U.S.C. § 112, sixth paragraph (emphasis added).
The Examiner has not made factual findings based on Livshits and Zeller
regarding the specific structure required by the claimed means, for example,
the “means for merging the N versions into a single, semantically correct,
Appeal 2011-001984
Application 11/093,517

7
unified program using the merge data structure, the single, semantically
correct, unified program having preprocessing conditionals to represent
differences between the N versions” recited in claim 18. The Examiner’s
rejection relies on the same findings as for claim 1 (Ans. 3-6, 14). However,
claim 1 does not recite any specific structure, but rather recites steps to be
performed in a method. Absent specific factual findings based on an
interpretation of the corresponding structure for the means-plus-function
limitations as described in the Specification, we find the Examiner has failed
to establish a prima facie case of obviousness for claim 18.
The Examiner responds that the claimed means correspond to a
general purpose computer, and the cited portions of Livshits and Zeller
suggest using such general hardware (See Ans. 13-14). However, for
computer-related means-plus-function claims, the corresponding structure is
the described algorithm executing on the computer hardware. WMS Gaming
Inc. v. Int’l Game Tech., 184 F.3d 1339, 1349 (Fed. Cir. 1999). Appellant
points to the Specification, for example, page 8, line 18 through page 13,
line 7 (App. Br. 6), as describing the algorithm required by claim 18. The
Examiner has not interpreted claim 18 in light of this description to
determine the scope of the claimed means, and therefore has not shown the
combination of Livshits and Zeller discloses the invention as defined in
claim 18.
We are therefore constrained by the record to find the Examiner erred
in rejecting claim 18.

Appeal 2011-001984
Application 11/093,517

8
CONCLUSIONS OF LAW
Under 35 U.S.C. § 103(a), the Examiner erred in rejecting claim 18,
but did not err in rejecting claims 1-17.

DECISION
For the above reasons:
The Examiner’s decision rejecting claims 1-17 is affirmed.
The Examiner’s decision rejecting claim 18 is reversed.
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). See 37
C.F.R. § 41.50(f).

AFFIRMED-IN-PART

llw