Opinion
Civil Action No. 16-453-RGA Civil Action No. 16-454-RGA Civil Action No. 16-455-RGA
04-10-2018
ACCELERATION BAY LLC, Plaintiff, v. ACTIVISION BLIZZARD, INC. Defendant. ACCELERATION BAY LLC, Plaintiff, v. ELECTRONIC ARTS INC. Defendant. ACCELERATION BAY LLC, Plaintiff, v. TAKE-TWO INTERACTIVE SOFTWARE, INC., ROCKSTAR GAMES, INC., AND 2K SPORTS, INC. Defendants.
MEMORANDUM ORDER
Presently before me is Defendants' Motion for Clarification of the Court's Claim Construction Opinion and Order (No. 16-453, D.I. 302; No. 16-454, D.I. 275, No. 16-455, D.I. 271) and related briefing (No. 16-453, D.I. 318; No. 16-454, D.I. 286; No. 16-455, D.I. 281).
All farther citations are only to the docket in No. 16-453.
For the reasons that follow, IT IS HEREBY ORDERED THAT Defendants' Motion for Clarification is GRANTED as to both Term 4 and Term 18.
I. Term 4 ("means for connecting to the identified broadcast channel") ('344/13, '966/13)
For Term 4 ("means for connecting to the identified broadcast channel"), my construction for U.S. Patent No. 6,701,344 (the "'344 patent") is:
A processor programmed to perform at least one of the algorithms disclosed in steps 801 to 809 in Figure 8 and described in the '344 Patent at 17:67-19:34, 19:66-20:44, 21:4-53, 22:61-24:6, and Figures 9, 11, 13, 14, 17 and 18, or Figures 3A and 3B and described in the '344 Patent at 5:33-55, which involves invoking the connecting routine with the identified broadcast channel's type and instance, connecting to the broadcast channel, connecting to a neighbor, and connecting to a fully connected state.(D.I. 287 at 3) (emphasis added). Term 4 has the same construction for U.S. Patent No. 6,714,966 (the "'966 patent") as for the '344 patent, with the only differences being column and line citations.
In their initial briefing, Defendants requested that I clarify my construction by deleting "at least one of" and changing "or" to "in combination with." (D.I. 302 at 4). Defendants argued that Figures 3A and 3B of the '344 patent do not provide "sufficient structure" for performing the recited function of "connecting to the identified broadcast channel." (Id.). More specifically, Defendants argued that Figures 3A and 3B and lines 5:33-55 are not an algorithm on their own and therefore cannot be "one of the algorithms" the structure requires. (Id. at 3). Rather, argued Defendants, Figures 3A and 3B and lines 5:33-55, along with Figure 8 and lines 17:67-19:34, 19:66-20:44, 21:4-53, 22:61-24:6, together comprise an algorithm that provides structure for performing the recited function. (Id.).
Plaintiff disagreed, arguing that either of the two alternative embodiments—"portions of Figure 8 and corresponding specifications or Figures 3A and 3B and corresponding specifications"—provides function and supporting structures sufficient for the means term's validity. (D.I. 318 at 1-3).
The arguments in the parties' initial briefs focused on whether Plaintiff had ever argued that Figures 3A and 3B alone constitute an algorithm. (D.I. 302 at 3; D.I. 318 at 5). However, the parties did not point to any evidence about whether Figures 3A and 3B actually constitute an algorithm.
I note that Plaintiff did identify alternative algorithm embodiments. Paragraph 57 of Plaintiff's expert Dr. Nenad Medvidovic's declaration identifies a single "algorithm" described in Figure 8 and its corresponding descriptions in the '966 and '344 patent specifications. Paragraph 58 then separately refers to a single "algorithm" found in Figures 3A and 3B and corresponding descriptions. Paragraph 59 then refers to these "algorithms" in tandem, stating that a "person of ordinary skill in the art would understand that the steps set forth in these figures and the related portions of the specification disclose [both] algorithms for performing the function of 'connecting a participant to an identified broadcast channel.'" (D.I. 191-1, Exh. F at ¶¶ 57-61).
Accordingly, in response to Defendants' motion, I directed the parties to submit additional briefs
on the issues of (1) whether there is a substantive difference between the algorithm/"process of a new computer Z connecting to the broadcast channel" of Figures 3A and 3B and corresponding specifications and the algorithm/"processing of the connect routine" of Figure 8 and corresponding specifications, and (2) if there is a difference, whether Figures 3A and 3B and corresponding specifications constitute a separate algorithm.(D.I. 332). They did. (D.I. 340, 345, 354).
As to the first issue, I found that the parties "seem to agree that the Figure 3A/3B algorithm and the Figure 8 algorithm are describing the same algorithm, but at different levels of detail." (D.I. 388 at 2).
As to the second issue, "Defendants argued that Figures 3A and 3B and corresponding specifications are a 'black box' and do not provide an independent algorithm for 'connecting.'" (Id. at 3 (citing D.I. 340 at 4-6, D.I. 354 at 2-6)). "Plaintiff, on the other hand, argued that Figures 3A and 3B and corresponding specifications do in fact provide an independent algorithm for 'connecting,' citing a new declaration from Dr. Medvidovic (D.I. 346)." (D.I. 388 at 3 (citing D.I. 345 at 9-10)). I ordered the parties to produce expert witness testimony to resolve the second issue. (D.I. 388 at 3; Noah Sys., Inc. v. Intuit Inc., 675 F.3d 1302, 1313-14 (Fed. Cir. 2012) (holding that the issue of whether structure is "sufficient . . . requir[es] consideration of what one skilled in the art would understand from [the] disclosure, whether by way of expert testimony or otherwise")).
I held a hearing on January 29, 2018, at which Plaintiff's expert Dr. Michael Mitzenmacher and Defendants' expert Dr. John Kelly testified. (D.I. 437 ("Tr.") 5:2-4, 63:17-19).
Dr. Kelly testified that Figures 3A and 3B and lines 5:33-55 are not relevant to the claims in which Term 4 appears. (Tr. 76:21-77:17). Both experts agree that Figure 3A depicts a "complete" graph, where the number of participants (5) is one more than m (4). (Tr. 30:5-9, 18-31). The "plurality of broadcast channels" in the claims, however, require that "the number of participants is at least two greater than m thus resulting in a non-complete graph." (Tr. 29:24-30:21; '344 patent, claim 13; '966 patent, claim 13). Thus, say Defendants, "[Figures 3A and 3B] depict a computer connecting to a small regime 'complete' graph, whereas Term 4 is directed to connecting to a channel 'of interest' from among a plurality of [large regime] 'non-complete' channels." (D.I. 456 at 2). Defendants assert that this mismatch means Figures 3A and 3B are "not relevant to the claims at issue and cannot be used as corresponding structure for Term 4." (Id.).
Plaintiff does not address this argument in letter briefing submitted after the hearing. (See generally D.I. 457). At the hearing, Dr. Mitzenmacher testified that Figure 3B depicts a "non-complete" graph, and therefore "meets the language of the claims." (Tr. 30:14-17). However, Term 4 requires "means for connecting to the identified broadcast channel." The "identified broadcast channel" refers to the graph being joined, not the graph that results from joining. (Tr. 29:5-17). Figure 3A depicts the graph being joined. Figure 3B, on the other hand, depicts the graph that results from joining. Figure 3B therefore does not depict an "identified broadcast channel." Dr. Mitzenmacher's point is unavailing.
I find that Figures 3A and 3B and lines 5:33-55 are not relevant to the claims in which Term 4 appears. Because Figures 3A and 3B and lines 5:33-55 of the '344 Patent specification are not relevant to the claims at issue, they cannot be used as a corresponding structure for Term 4. I will therefore grant Defendants' Motion as to Term 4. I need not reach the parties' arguments about whether the Figure 3 A/3B algorithm provides sufficient structure. (D.I. 456 at 2-4; D.I. 457 at 1-6). I adopt the following construction for Term 4:
This conclusion is consistent with my finding that that the parties agreed "that the Figure 3A/3B algorithm and the Figure 8 algorithm are describing the same algorithm, but at different levels of detail." (D.I. 388 at 2). Figure 8 is not limited to connecting to a "complete" graph. --------
A processor programmed to perform the algorithms disclosed in steps 801 to 809 in Figure 8 (described in the '344 Patent at 17:67-19:34, 19:66-20:44, 21:4-53, 22:61-24:6), and Figures 9, 11, 13, 14, 17 and 18, which involves invoking the connecting routine with the identified broadcast channel's type and instance, connecting to the broadcast channel, connecting to a neighbor, and connecting to a fully connected state.
II. Term 18 ("m-connected" and "m-connected network") ('634/1, 19)
For term 18 ("m-connected" and "m-connected network"), my construction for the '344 patent is:
A state that the network is configured to maintain, where the network may be divided into disconnected sub-networks by the removal of m participants in a steady state.(D.I. 287 at 5) (emphasis added). Defendants request that I strike "in a steady state" from this construction to be consistent with the construction for the "m-regular" terms. Plaintiff does not really oppose this request. (D.I. 318 at 7-8).
For the "m-regular" terms, I did not include Plaintiff's requested language "in a steady state" in my construction. Plaintiff expressed concern with my construction on grounds that it might read out a configuration where the network is not in a "steady state," defining a "steady state" as existing when the network is configured such that each participant is connected to exactly m neighbors. Plaintiff explained that there will be "points where [the] network . . . is configured to maintain where each participant [is] connected to exactly m neighbors, but . . . also allows for . . . flexing in and out of that configuration." (D.I. 219 at 70:25-71:16). Plaintiff urged that Defendants' proposed language of "at all times" would improperly disallow this "flexing" from the "steady state" because it would require the network to have each participant be connected to m neighbors at all times. But by eliminating Defendants' "at all times" language, my construction requires only that the network is configured to have each participant be connected to m neighbors. In other words, it is fine that the network does not have each participant connected to m neighbors so long as, when appropriate, the network tries to get that configuration. (D.I. 275 at 14-15).
Here, as is the case in my construction for the "m-regular" term, the network is not "configured to maintain" an "m-connected" state "at all times." Indeed, for that reason, I struck "at all times" from the Defendants' proposed construction. (D.I. 275 at 16). As a result, to the extent that Plaintiff's concerns were alleviated for the "m-regular" term by my striking Defendants' proposed "at all times" language, they should be alleviated here, as well, even absent the "in a steady state" language. Thus, to be consistent with the "m-regular" construction, I will grant Defendants' Motion as to Term 18 and strike "in a steady state" from the construction.
Entered this 10 day of April, 2018.
/s/_________
United States District Judge