concurring in part, dissenting in part.
I join the majority in affirming the district court’s dismissal with prejudice of claims 1 through 11. I join the majority, too, in affirming the district court’s construction of claim 12 insofar as that construction requires the transmitted gateway message to be received by user terminals. I part company with the majority and the district court in two respects.
First, I disagree with their understanding that receipt of the gateway message by the user terminals entails that the message received be completely unchanged, bit by bit, from the message initially sent. On its face, claim 12 requires the messaging system “to transmit at least one gateway message to a plurality of the user terminals.” '428 patent, col. 30, lines 31-33. As the *968majority concludes, the natural meaning of that language here is that the transmitted gateway message is received by user terminals, especially given that the claim requirement of a “second message” received by user terminals, in order to have an antecedent basis, must refer to the “gateway message.” But that conclusion does not exclude all possible alterations between transmission by the messaging system and receipt by the user — for instance, an alteration of address information without alteration of the text expressing the content being communicated. Nothing in the claim language, the specification, or the prosecution history compels such a strict exclusion.
The practice of “blind copies” (Bcc) of email messages supplies an example of a common understanding that the same “message” may be considered to be received even when the copies differ in certain respects, specifically, in the address information in the header field of a message. The proposed Internet standard document, Request for Comment (RFC) 2822, concerning “Internet Message Format,” discusses blind copying in section 8.6.8. See Internet Engineering Task Force, Internet Message Format, RFC 2822, § 3.6.3 (April 2001), available at http://www.rfc-editor.org/rfc/rfc2822.txt. In one implementation,
recipients specified in the “To:” and “Cc:” lines each are sent a copy of the message with the “Bcc:” line removed ..., but the recipients on the “Bcc:” line get a separate copy of the message containing a “Bcc:” line. (When there are multiple recipient addresses in the “Bcc:” field, some implementations actually send a separate copy of the message to each recipient with a “Bcc:” containing only the address of that particular recipient.)
Id. The “Bcc:” recipients thus get “a copy of the message” that is received by the “To:” and “Cc:” recipients, but the address information in the several copies is changed. This example points strongly toward an accepted usage, at least in the area of electronic messaging, under which the address portion of a transmitted multi-cast message may be changed for delivery to different recipients, yet all recipients are understood to receive copies of the same “message.” Similar usage may apply to software that transforms a group email into a group of individually addressed emails. I see no basis for departing from that intuitive, common-sense understanding in construing the requirement of claim 12 that the message transmitted by the messaging system arrive at the user terminals.
The majority opinion rests its contrary, strict-exclusion-of-all-changes construction to a large extent on the understanding that (i) there must be a difference in scope between the transmission limitations of claims 1 and 12 and (ii) the difference would be erased unless what is received by the user terminals in claim 12 is identical, bit by bit, to what left the messaging system. The second of those steps seems to me wrong. The reason is that claim 1 permits a vast range of differences between the gateway message and the message received by the user terminals-differences that could never be encompassed within a notion that the same “message” (the gateway message) is received, as claim 12 requires. Allowing some changes in the claim 12 messages thus does not collapse claim 12 into claim 1. Nor does it contradict anything the examiner said during reexamination about the distinction between claim 1 and claim 12, because the examiner, in observing that the latter requires the gateway message to arrive at the user while the former does not, said nothing about what that requirement en*969tails, such as whether it excludes all changes, including address changes, between sending and receipt.
Specifically, claim 1 recites a “first message,” a “second message corresponding to the received first message,” “converting the first message to one or more gateway messages” (through format changes that allow the gateway to understand the first message), and “transferring ... gateway messages to one or more corresponding communication gateways for distribution of the second message” — the “second message” being the message the user receives. '428 patent, col. 29, lines 22-46. Critically, claim 1 does not define the relationship between the gateway message and the second message, except to say that the gateway message acts as a kind of trigger for the distribution of the second message, which must in some undefined way “corre-sponde ]” with the first message that gets reformatted into the gateway message. In no sense does claim 1 require the gateway message (or first message) to be received by user terminals, ie., to be the same as the received message. Under claim 1, every word or character in the user-received message, including in the text expressing the content being communicated (body), could differ from what is in the gateway message or first message: the body of the first and gateway messages could read, “thermometer number 7 reads 140 degrees,” and the second message (programmed into the gateway for distribution to users) could read, “FIRE ALERT; EVACUATE BUILDING CALMLY BUT QUICKLY.” Such examples are plainly covered by claim 1, but they could not reasonably be covered by claim 12 according to an ordinary meaning that allows a much more limited range of changes, such as in addressing. The key premise of the no-change-at-all construction — the need to avoid making claims 1 and 12 identical in scope — thus seems to me not to support that construction.
Without the rigid no-change-at-all construction, the district court’s summary judgment of no infringement — even of no literal infringement — cannot stand. That is so without regard to any supplemental reports submitted by Cooper (which, as the majority holds, the district court could properly reject). We do not have before us a district court ruling, or an evidentiary and legal presentation by the defendants, that acknowledges the ordinary-meaning possibility of some changes along the transmission path and then focuses on what changes, in type or magnitude, would be understood by a person of ordinary skill in the art to be consistent with the requirement that the “gateway message” gets received by user terminals and whether the accused systems make more than such changes. Without examining those questions, a judgment of no literal infringement on claim 12 and its dependent claims cannot be entered. I would remand for further proceedings on those questions, leaving to the district court whether further claim construction should occur or whether the matter should proceed directly to the application-of-construction stage of an infringement analysis.
Second, even under the no-change-at-all construction adopted by the majority and the district court, I believe that Cooper pointed to enough evidence — without regard to the supplemental reports it sought to add to the record — to create a genuine issue of fact on infringement of claim 12 (and its dependent claims) under the doctrine of equivalents. While it is undisputed that in the accused systems some changes are made between sending out and user receipt, as the majority explains, Cooper consistently argued that, and submitted expert evidence that explained in simple terms why, the changes the accused *970systems make in the gateway messages make no substantial difference to the role of the claim 12 transmission/receipt limitation.
The defendants criticize the explanation as insufficiently expansive, citing the requirement of particularized testimony and linking argument for equivalents proof. See Texas Instruments Inc. v. Cypress Semiconductor Corp., 90 F.3d 1558, 1566-67 (Fed.Cir.1996). But I see no need for more explanation, at least at the summary-judgment stage, where the point is so simple, namely, that in the accused systems there is no change in the communicative text, but at most only in certain addressing or formatting information. Moreover, adopting that explanation of equivalents would not vitiate the claim limitation by leaving the scope of claim 12 the same as if the limitation were not present; nor would it override any representation or action by Cooper during prosecution. I would accordingly hold that Cooper said enough to preclude summary judgment of no equivalents infringement even under the strict construction of claim 12 adopted by the district court and, now, by the majority.
For the foregoing reasons, I would reverse in part the district court’s claim construction and the order granting summary judgment of non-infringement. In those respects, I respectfully dissent.