Appeal 2007-2127
Reexamination Control No. 90/006,621
whereby during execution of the program each thread may have at
various times direct access to the same program address space, and
with at least one thread invoked by a periodic clock-activated
interrupt service routine which upon each activation asynchronously
and preemptively takes control of the central processing means away
from an executing thread at a repetition rate sufficiently fast so that
even where the system contains only a single central processor the
concurrent threads appear to execute effectively simultaneously and
are so perceived by the user. [Emphasis added.]
Therefore, Patent Owner argues, he "specifically defined 'multithreading' to
state that 'at least one thread [is] invoked by a periodic clock-activated
interrupt service,'" (Br. 35) and "[n]othing in this definition, or elsewhere in
the specification, claims, or prosecution history, requires more than one
thread to be interrupted" (Br. 35).
The issue is how the terms "threads" and "multithreading" are defined
in the '604 patent, not in the '603 patent. The '604 patent requires that at
least a plurality of threads in a "preemptive multithreading" system are
interruptible. The definition of "multithreading" in the '603 patent was
added by amendment after its filing date and is not intrinsic evidence of
what the term meant as of the 1990 filing date. See Phillips v. AWH,
415 F.3d at 1313, 75 USPQ2d at 1326 ("We have made it clear, moreover,
that the ordinary and customary meaning of a claim term is the meaning that
the term would have to a person of ordinary skill in the art in question at the
time of the invention, i.e., as of the effective filing date of the patent
application."). Patent Owner stated numerous times during the prosecution
of the 1990 application that the terms "threads" and "multithreading" were
intended to have their ordinary meaning in the art. To the extent the
59
Page: Previous 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 Next
Last modified: September 9, 2013