Retroactive Correction of Patent ClaimsBack in March, I discussed the degree of change that can be secured from the USPTO via a Certificate of Correction (COC). As a reminder, one of the more limited methods to address errors in an issued patent is via COC as defined by 35 U.S.C. § 255. The statue provides that mistakes of a clerical or typographical nature, or of minor character may be corrected upon request. Such mistakes must have been made in good faith, and not constitute new matter or require reexamination.Once issued, a valid COC is only effective prospectively. For this reason, such certificates are best secured at the earliest possible date.Of course until claims are very closely scrutinized, the need for minor correction may not be so apparent. In cases where the COC is not secured for years, is there a mechanism available to retroactively correct a patent claim as to such minor errors?This issue was recently explored in Fujitsu Limited v. Tellabs Operations Inc., and Telllabs Inc. (NDIL).In Fujitsu U.S. Patent 5,386,418 was at issue. Claim 1 recited the language “and flag data;” this was a mistake. Claim 1 should have recited “a flag data.”Although the ‘418 Patent originally issued in 1995, a COC was not obtained until 2007, immediately prior to the Tellabs litigation.With respect to the claim language issue, Tellabs advanced two theories on summary judgment: (1) The COC was invalid for broadening the claims; and (2) claim 1, as issued, was indefinite.In ruling on these motions, the court explained that the COC corrected a minor issue, thus the COC was valid. For similar reasoning the court rejected argument (2).Fujitsu also sought summary judgment of its own, arguing that that the claim correction should be made retroactively, based on the doctrine of judicial correction. Tellalbs argued that the combination of a COC and judicial correction for the same patent was improper. The court agreed with Fujitsu, explaining, (full decision here):Although a valid certificate of correction is only effective prospectively, Southwest Software, Inc. v. Harlequin Inc., 226 F.3d 1280, 1295 (Fed. Cir. 2000) (“[F]or causes arising before its issuance, the certificate of correction is not effective.”), the district court can retroactively correct certain errors in a patent’s claims if “(1) the correction is not subject to reasonable debate based on consideration of the claim language and the specification and (2) the prosecution history does not suggest a different interpretation of the claims.” Novo Indus., L.P. v. Micro Molds Corp., 350 F.3d 1348, 1354, 1357 (Fed. Cir. 2003). These “determinations must be made from the point of view of one skilled in the art.” Ultimax Cement Mfg. v. CTS Cement Mfg., 587 F.3d 1339, 1353 (Fed. Cir. 2009).  .           .           .           .           .           .           .           .           .           . Neither Hoffer nor Ultimax suggests that the district court’s authority to correct a patent is constrained by the presence of a certificate of correction, and the court declines to impose such a proscription against judicial correction in this case. Consequently, the court agrees with Fujitsu that the Novo standard is the controlling standard for judicial correction, and the ‘418 Patent’s Certificate of Correction does not prevent this court from correcting the same error in claim 1.           .           .           .           .           .           .           .           .           . Patents provide an important notice function to the public and, as a result, post-issuance judicial corrections which have a retroactive effect must be carefully scrutinized. In this case, viewing the intrinsic evidence from the perspective of one of ordinary skill in the art, as the court must in determining whether a judicial correction is supported under Novo, this court finds that the only reasonable correction of claim 1 replaces “and flag bit data” with “a flag bit data.” A person of ordinary skill in the art, having reviewed the specification, drawings, and prosecution history, would have immediately recognized this error and understood what was claimed. The court accordingly corrects claim 1 of the ‘418 Patent to replace “and flag bit data” with “a flag bit data.”This case was brought to my attention by the great Docket Navigator.

Back in March, I discussed the degree of change that can be secured from the USPTO via a Certificate of Correction (COC). As a reminder, one of the more limited methods to address errors in an issued patent is via COC as defined by 35 U.S.C. § 255. The statue provides that mistakes of a clerical or typographical nature, or of minor character may be corrected upon request. Such mistakes must have been made in good faith, and not constitute new matter or require reexamination.

Once issued, a valid COC is only effective prospectively. For this reason, such certificates are best secured at the earliest possible date.

Of course until claims are very closely scrutinized, the need for minor correction may not be so apparent. In cases where the COC is not secured for years, is there a mechanism available to retroactively correct a patent claim as to such minor errors?

This issue was recently explored in Fujitsu Limited v. Tellabs Operations Inc., and Telllabs Inc. (NDIL).

In Fujitsu U.S. Patent 5,386,418 was at issue. Claim 1 recited the language “and flag data;” this was a mistake. Claim 1 should have recited “a flag data.”

Although the ‘418 Patent originally issued in 1995, a COC was not obtained until 2007, immediately prior to the Tellabs litigation.

With respect to the claim language issue, Tellabs advanced two theories on summary judgment: (1) The COC was invalid for broadening the claims; and (2) claim 1, as issued, was indefinite.

In ruling on these motions, the court explained that the COC corrected a minor issue, thus the COC was valid. For similar reasoning the court rejected argument (2).

Fujitsu also sought summary judgment of its own, arguing that that the claim correction should be made retroactively, based on the doctrine of judicial correction. Tellalbs argued that the combination of a COC and judicial correction for the same patent was improper. The court agreed with Fujitsu, explaining, (full decision here):

Although a valid certificate of correction is only effective prospectively, Southwest Software, Inc. v. Harlequin Inc., 226 F.3d 1280, 1295 (Fed. Cir. 2000) (“[F]or causes arising before its issuance, the certificate of correction is not effective.”), the district court can retroactively correct certain errors in a patent’s claims if “(1) the correction is not subject to reasonable debate based on consideration of the claim language and the specification and (2) the prosecution history does not suggest a different interpretation of the claims.” Novo Indus., L.P. v. Micro Molds Corp., 350 F.3d 1348, 1354, 1357 (Fed. Cir. 2003). These “determinations must be made from the point of view of one skilled in the art.” Ultimax Cement Mfg. v. CTS Cement Mfg., 587 F.3d 1339, 1353 (Fed. Cir. 2009).  

.           .           .           .           .           .           .           .           .           . 

Neither Hoffer nor Ultimax suggests that the district court’s authority to correct a patent is constrained by the presence of a certificate of correction, and the court declines to impose such a proscription against judicial correction in this case. Consequently, the court agrees with Fujitsu that the Novo standard is the controlling standard for judicial correction, and the ‘418 Patent’s Certificate of Correction does not prevent this court from correcting the same error in claim 1

.           .           .           .           .           .           .           .           .           . 

Patents provide an important notice function to the public and, as a result, post-issuance judicial corrections which have a retroactive effect must be carefully scrutinized. In this case, viewing the intrinsic evidence from the perspective of one of ordinary skill in the art, as the court must in determining whether a judicial correction is supported under Novo, this court finds that the only reasonable correction of claim 1 replaces “and flag bit data” with “a flag bit data.” A person of ordinary skill in the art, having reviewed the specification, drawings, and prosecution history, would have immediately recognized this error and understood what was claimed. The court accordingly corrects claim 1 of the ‘418 Patent to replace “and flag bit data” with “a flag bit data.”

This case was brought to my attention by the great Docket Navigator.