Unwired Planet v Huawei first technical appeal decision
16 August 2017
This appeal was brought by Huawei in respect of the decision of  EWHC 336 (Pat) in which it was held that Unwired Planet’s patent was both valid and infringed.
The patent is one of a series purchased by Unwired Planet from Ericsson relating to LTE 4G telecommunications; in particular, this patent concerned a method of polling a receiver for a transmitted-information status report after a predetermined amount of data, measured in Protocol Data Units (PDUs), has been transmitted. The appeal covered three main areas relating to the validity of the patent: priority, novelty and obviousness.
It was agreed by all parties that the document ‘Ericsson TDoc’ would be novelty-destroying for the patent if it were determined to be a part of the state of the art at the time of filing. Ericsson TDoc was uploaded to a Europe-based public server at 08:36CET on 08 January, while the priority document was filed at the USPTO fourteen hours later (16:59EST/22:59CET on 08 January).
Huawei argued that Ericsson TDoc was uploaded early enough that it was still the previous day (07 January) in some parts of the world, including parts of the US in the GMT-10 time zone, whereas the priority document was filed late enough that it was no longer 07 January anywhere in the world. Huawei submitted that as the TDoc was in principle accessible by residents of Hawaii on 07 January (and thus before the priority document’s filing date), the TDoc formed a part of the state of the art for the purpose of novelty.
The appeal judge held that ‘the date is determined in the time reference of that patent office’, and that the filing/priority date is the 24-hour period (ie, day) in that time zone. Therefore the fact that the document was available to some members of the public on 07 January was irrelevant, as no member of the public as measured in the time zone of the USPTO was able to access this document before 08 January.
While this decision may be detrimental to those who genuinely publish documents in these earlier time zones, as the published documents may not be citeable against competitor’s applications filed the next day, it appears to be a logical conclusion on the matter. To take the alternative view would allow the publication of documents after the filing of an application that could be treated as a part of the state of the art for assessing patentability.
In view of this decision, there is no real impact to filing strategy for new applications; as ever, the earliest filing date possible should be sought for time-sensitive applications. That said, it may be beneficial to file first at the EPO rather than the USPTO for these applications so as to secure an ‘earlier’ 24-hour window as the filing date. In this case, if the Ericsson TDoc had been uploaded three hours earlier then it could have been prior art as assessed at the USPTO (as the publication time would have been 23:36 EST on 07 January) but not at the EPO (as this was 05:36 CET on 08 January). Therefore for an application filed at 16:59EST/22:59CET on 08 January (as in this scenario), the TDoc would be prior art for an application filed at the USPTO (as it was uploaded on 07 January in the USPTO’s time zone) – but not if filed at the EPO (as it was uploaded on 08 January in the EPO’s time zone).
This was a question of how the skilled person would interpret the teaching of a priority document. The priority document disclosed the feature of ‘counting the number of transmitted data units’, which had been held at first instance to include embodiments in which the counting was performed before transmission.
Huawei argued that this disclosure should be limited to embodiments in which the counting was performed after transmission, on the grounds that for this to be performed the data units must have been transmitted (otherwise they would be data units to be transmitted). In response to this, Unwired Planet argued that the skilled person would not be concerned with such a level of detail, and would adopt the approach known from common general knowledge in the absence of specific teaching to the contrary (broadly in line with the opinion of the judge in the first instance).
The appeal judge noted that “the exercise of determining priority involves asking whether the invention is directly and unambiguously derivable from the priority document, not whether every possible embodiment is so derivable”. This position was seen as more reasonable than the alternative; to expect the patentee to describe every possible embodiment would not be reasonable.
The judge went on to state that the priority document should not be read in a vacuum, and that common general knowledge should be used to provide context for the disclosure. This approach seems reasonable, as the use of the common general knowledge at the priority date in interpreting a document cannot be seen to be adding to the disclosure in an inventive or non-obvious manner.
In summary, the appeal judge took the view that the skilled person would indeed interpret the priority document as having the broader meaning, as to ignore the well-known conventional approach would be to interpret the document in a vacuum. It was also emphasised that the threshold for entitlement to priority is significantly lower than that for determining the novelty of a claim.
This aspect of the appeal related to the judge’s use of secondary evidence in the first instance. A document (Motorola TDoc) which had been submitted to the 4G standards body for consideration was relied upon as prior art. Motorola TDoc disclosed a polling method that uses a trigger based on byte count, rather than a PDU count. Huawei argued that at the time of the priority date, the skilled person would be concerned with a problem that required the counting of PDUs and therefore would combine the teachings of Motorola TDoc with a PDU counter known from the 3G standard.
Considering secondary evidence, such as minutes of the meetings of the standards committee, the judge determined that the method of Motorola TDoc had already been dismissed at the time of the priority date and that the committee were moving away from the idea of using a PDU counter generally. Huawei argued that the primary evidence (Motorola TDoc and expert witnesses’ evidence) would lead to a conclusive determination of a lack of inventiveness, and that no amount of secondary evidence could override this.
The appeal judge noted that an appeal court is usually reluctant to interfere with any fact-based finding on obviousness at the first instance unless there is an error of principle (in line with MedImmune v Novartis  EWCA Civ 1234). The judge also went on to state that the arguments for a lack of inventiveness by Huawei (based on the primary evidence only) were not persuasive, as it was considered that there was a reliance on hindsight.
The use of hindsight was evidenced by the fact that only the byte counter was taken from Motorola TDoc, and as such the arguments from Huawei appeared to be more an identification of individual components from different sources rather than a coherent argument of obviousness. The judge also noted that the expert evidence appeared to take no account of a number of technical facts, and so was not persuasive.
The judge found that it was appropriate to turn to the secondary evidence in assessing the obviousness in this case; the judge at first instance had made it clear that they were aware of the evidential limitations of the secondary evidence, and had expressly warned against relying on this evidence too heavily. In any case, it was held that the secondary evidence suggested the same conclusion as the primary evidence and as a result the use of the secondary evidence was not unduly relied upon.
Huawei’s appeal was dismissed in respect of each ground, and therefore Unwired Planet’s patent was held to be both valid and infringed.