| A Study on the Cross-Examination of Online Video Evidence in Patent Invalidity P |
With the development of 5G communications and Internet technologies, an increasing amount of information is disseminated in the form of online videos. In particular, with the popularity of various short-video apps in recent years, many new products are introduced and promoted online through short videos. In patent invalidation and administrative litigation proceedings involving products, short-video evidence has been increasingly used and has become a key focus of debate during cross-examination. Taking a utility model patent invalidation case handled by our firm as an example, this paper discusses how to reasonably and effectively respond to the cross-examination of online video evidence in patent invalidation proceedings. Case Introduction This patent relates to a plastic cup for beverages. The petitioner filed a request for patent invalidation on the ground that Claims 1 to 6 of the disputed patent lack inventiveness compared with the prior art and do not meet the inventiveness requirements stipulated in Article 22, Paragraph 3 of the Patent Law, and cited a Douyin video as the closest prior art. Evidence 1:Douyin video titled Bayberry Cup (URL: http://v.douyin.com/emDLxdo/), with a publication date of July 12, 2020. Analysis of the Authenticity and Publicity of Evidence 1
After analysis and research, it is found that although the upload time of Evidence 1 is earlier than the filing date of the subject patent, the scope of its disclosure is questionable. The reasons are as follows: (1) A large number of video works uploaded by the Douyin account that published Evidence 1 have been hidden. The number of video works on this Douyin account is inconsistent with the total number of likes received by the store. The Douyin account has 50 public video works, 848,000 followers and a total of 2.94 million likes for all works. Among them, the most liked video has 1,089 likes and the least liked one has 49 likes. The total number of likes for the 50 public video works adds up to 7,414, which is hundreds of times less than the total 2.94 million likes. From a quantitative perspective, an analysis based on the average number of likes for the public videos under the current account shows that the vast majority of videos remain undisclosed compared with the currently public ones. According to the calculation rule for the total number of likes of Douyin works—"If a Douyin account deletes a work, the likes received during the work's release period will also disappear; if a Douyin account hides a video work, the likes will still be included in the account's total likes, only the video work will not be disclosed to the public. Only when a video work is deleted will the corresponding likes be deducted from the total likes"—it can be inferred that more than 99.75% of the video works of the Douyin account that published Evidence 1 have been hidden. Therefore, there is sufficient reason to suspect that the scope of disclosure permissions for the video works of this account may have been altered after the works were uploaded.
(2) The timing of video uploads and publications on this Douyin account is unreasonable. A statistical analysis of the 50 currently public videos of the store reveals that their release dates are concentrated within an extremely short period from June 19 to September 6, 2020. For a Douyin store claimed by the petitioner to be engaged in promotional sales, uploading and publishing works only within a three-month window runs counter to the basic logic of sales promotion. (3) The public products in the store lack brand information and purchase links, which is inconsistent with the basic practices for promoting and selling products. (4) The Douyin account did not publish the works for the purpose of public disclosure. The homepage introduction of the Douyin account that released Evidence 1 states it is "a emerging home textile brand creating a new experience of comfortable sleep", yet its public videos cover a wide range of product categories with no fixed focus. It is reasonable to conclude that the account publishes works in a random manner and its purpose is not product promotion or sales. Therefore, Evidence 1 fails to meet the requirement of high probability that the video of the beverage cup has been in a non-private state since its release, as claimed by the petitioner, and its publication does not constitute public disclosure in the sense of the Patent Law. In addition, during the oral hearing, our team demonstrated the operations of deleting and hiding a published Douyin video with an original 8 likes on the spot. The demonstration results showed that deleting the video would result in the immediate disappearance of its likes; hiding the video—i.e., setting its disclosure scope to something other than "visible to all the public"—would retain the likes in the account's total like count. Opinions of the Collegial Panel
The probability that Evidence 1 was set to the disclosure scope of "Visible to All" at the time of publication has not reached the degree of high probability, and the existing evidence is insufficient to prove that the video’s release time is its disclosure time. Therefore, Evidence 1 cannot be regarded as the prior art of the subject patent. The main reasons on which the Collegial Panel based its opinion are the arguments stated by our team in the cross-examination opinions: The homepage introduction of the publisher of Evidence 1 states that it is "a new home textile brand creating a new experience of comfortable sleep", yet the videos it published cover a variety of commodities, and it cannot be confirmed whether these commodities are produced or operated by the publisher. No product brand information or purchase links are provided in the videos either. Based on the videos still publicly available at present, the Collegial Panel cannot determine the thinking guiding the compilation and selection of content for the videos published on this account (whether filmed and uploaded or downloaded and reposted), the purpose for which the account holder published these videos, or the interrelationship between these videos. Approximately 50 video works are publicly available on the homepage of the publisher of Evidence 1, with the total number of likes for these videos adding up to more than 7,400, while the total number of likes displayed on the homepage of the account stands at 2.94 million—a difference of hundreds of times. From a quantitative perspective, an analysis based on the average number of likes for the currently public videos under this account shows that the vast majority of videos remain undisclosed. Furthermore, in light of Point 1 above, the Collegial Panel cannot easily judge the rules followed by the account holder for selecting videos to publish publicly. Based on the existing evidence, such selection is highly random and uncertain. In particular, the vast majority of the currently undisclosed videos have experienced changes in their disclosure status. Thus, it is difficult to confirm the status of the currently public videos before the filing date of the subject patent, whether their status has changed, and the number of such changes (if any). There are no comments posted prior to the filing date of the subject patent below Evidence 1, the video cited by the petitioner. Conclusion Chapter 8 of the Guidelines for Patent Examination, Provisions on Evidence Issues in Invalidation Proceedings, requires the following for cross-examination of evidence:“During cross-examination, the parties shall challenge, explain and debate the relevance, legality and authenticity of the evidence, as well as the existence and weight of its probative force. “During cross-examination, the parties shall challenge, explain and debate the relevance, legality and authenticity of the evidence, as well as the existence and weight of its probative force.” The author holds that the authenticity and publicity of online video evidence should be determined by focusing on the following aspects: (1) The release mechanism of the video platform Video platforms for users to share and exchange information have different release mechanisms due to differences in their operation mechanisms and service focuses. For videos published on such platforms, it is necessary to examine the release mechanism and the publisher, and comprehensively assess the likelihood that the content shown in the evidence was in a public state at the time of publication. (2) The subjective intent of the video publisher On some video platforms, publishers have the right to choose whether to make their videos public. In such cases, it is necessary to infer the uploader’s subjective intent when publishing the video based on the video content. Even assuming the publisher intended to promote or market a product, further consideration should be given to the possibility that the publisher may hide, delete or modify the video content, to determine whether its publicity meets the standard of high probability. (3) Mutual corroboration among online video evidence A single online video evidence has relatively weak probative force on its own. However, if multiple platforms corroborate the same factual content, the likelihood of such evidence being accepted will be greatly improved. References： [1]《专利审查指南》（2010）知识产权出版社. [2] 无效宣告请求审查决定（第562965号）. [3]《以案说法-专利复审、无效典型案例汇编（2018-2021）年》知识产权出版社，2022年. [4]《以案说法-专利复审、无效典型案例指引》知识产权出版社，2018年. [5] 赵静，从复审/无效案例看网络证据的使用原则，专利代理 ,2020.8.15. |
