One Case a Day | China: Addressing the Pain Points of Using Online Public Disclosures as Prior Art – "Purifier" Case, Invalidation Decision No. 569325 (2024)

Case Introduction

Due to the technical nature of online evidence, which is easily modified, deleted, or retrospectively edited, there is often significant controversy regarding the determination of its publication date. Additionally, the admissibility of evidence from overseas websites in terms of authenticity, completeness, and methods of evidence preservation also faces certain challenges in practice.

Internet Archive (the Internet Archive) automatically and periodically crawls and archives publicly accessible websites on the internet, forming and preserving historical versions of web pages. It is an independent non-profit organization that provides public access services in the form of a digital library. The platform has been operating continuously for over 20 years, accumulating a vast amount of historical web page records. Its independence, non-profit nature, and automated crawling mechanism without human intervention provide an important technical means for objectively fixing the public state of a specific URL at a specific point in time, and also offer practical possibilities for proving the publication date of online information.

The case shared today originates from the "Compilation of Key Points from Typical Patent Reexamination and Invalidation Case Decisions 2024" published by the Reexamination and Invalidation Department. It specifically points out:

For prior art evidence based on web page archives from the "Internet Archive," the publication date should be determined in conjunction with the website's mechanism.

First, the Internet Archive assigns a URL to each crawled and archived web page file, and the specific numeric field in the URL represents the crawling and archiving time of the web page.

Second, elements such as images within the web page can be crawled multiple times separately, with their respective crawling times recorded. Therefore, the crawling time of a web page archive and the elements it contains may differ.

Finally, based on the systematic and consistent nature of the website's URL assignment for archived files, the URLs for archived web pages and images on this website should follow the same naming rules and format.

Therefore, in the absence of contrary evidence, it can be determined that the numeric field in the image URL that reflects date and time information represents the crawling and archiving time of that image, i.e., the publication time of that image on the Internet Archive website.

Case Information

• Application Number: 201611223250.7
• Invention Title: Multifunctional Purifier
• Invalidation Request Information: 4W116322, 4W116768
• Invalidation Decision Information: No. 569325
• Decision Date: June 24, 2024

Controversial Focus

The involved patent protects a multifunctional purifier. By using a gas-liquid exchange filter element and a spraying device in an air purifier, it utilizes a solution to adsorb or decompose air pollutants, achieving a self-cleaning function. This solves the problem of high costs caused by frequent filter element replacement and improves purification effectiveness and equipment lifespan.

The invalidation requester cited Evidence 2 (Notarial Certificate (2023) Jing Fang Yuan Nei Jing Zheng Zi No. 20439, whose notarized content is product information about gas-liquid filter screens on the web.archive.org website) as prior art to challenge the inventiveness of the involved patent.

The controversial focus of this case is: How to determine the authenticity and publication date of Evidence 2.

Views of the Parties

Regarding the authenticity of Evidence 2:

The requester argued that Evidence 2 was accessed through the notary office's dedicated network, and the process was notarized, meeting legal requirements.

The patentee argued that the Internet Archive website is a foreign website that requires specific methods to access, and ordinary people cannot log in to it.

Regarding the publication date of Evidence 2:

The requester argued that the publication time of the web page on the web.archive.org website is also the publication time of the images on that web page. They also argued that the numeric field in the URLs of both the web page and its images on this website represents the archiving time of those archived files, all of which predate the filing date of this patent. Therefore, it can be used as prior art for this patent.

The patentee argued that the archiving time of a web page and the archiving time of its images may not be consistent, so the archiving time of the web page cannot be used to determine the archiving time of its images. They also argued that if the numeric field in the URL is considered the archiving time of the archived file, then the numeric field in the URL of the image on page 10 of Evidence 2 shows a time earlier than the archiving time of the corresponding web page, which is illogical. Additionally, the URLs on pages 10 and 11 have an "im" suffix after the numeric field, and its meaning is unclear. Furthermore, Attachment 2's examination decision also did not recognize the publication time of web pages linked by the Internet Archive's archive.org crawl.

Regarding Authenticity

First, according to the provisions of Article 22, Paragraph 5 of the Patent Law, prior art refers to technologies known to the public at home and abroad before the filing date. Internet Archive (the Internet Archive) is a non-profit digital library organization outside China, providing permanent free storage and access to digital data such as websites, music, moving images, and millions of books. Its data is automatically collected by its own web crawlers. The Wayback Machine is one of the most important services of the Internet Archive, allowing people to search and access its web page archives. Although the content of the Internet Archive website is currently inaccessible through ordinary user terminals in China, this does not affect the use of the website's publicly available content by the requester to prove prior art before the filing date of this patent, which complies with the above provisions of the Patent Law.

Second, the dispute in this case lies in how to obtain the web page content of this website. According to Evidence 2's notarial certificate, the notary personnel of Beijing Fangyuan Notary Office accessed the website through the notary office's dedicated network and saved screenshots of the relevant content. According to Article 36 of the Notary Law: "Notarized civil legal acts, facts with legal significance, and documents shall serve as the basis for determining facts, unless there is contrary evidence sufficient to overturn the notarization." At the same time, Section 4.3.4 of Chapter 8, Part IV of the Examination Guidelines stipulates: "When one party submits a notarial certificate as evidence, the facts proven by a valid notarial document shall serve as the basis for determining facts, unless there is contrary evidence sufficient to overturn the notarization." Additionally, Article 39 of the Notary Law stipulates: "If a party or an interested party in a notarial matter believes that there is an error in the notarial certificate, they may apply to the notary office that issued the notarial certificate for review. If the content of the notarial certificate is illegal or inconsistent with the facts, the notary office shall revoke the notarial certificate and make an announcement, and the notarial certificate shall be invalid from the beginning; if there are other errors in the notarial certificate, the notary office shall make corrections."

In this case, from a formal perspective, Evidence 2's notarial certificate does not have obvious defects. From the notarization process recorded in the notarial certificate, the notarization act is legal and compliant. The patentee has neither submitted evidence sufficient to overturn the notarization, nor has the notarial certificate been revoked or corrected by the notary office. According to the above provisions of the Notary Law and the Examination Guidelines, the facts notarized by Evidence 2 shall serve as the basis for determining facts in this case. On this basis, the collegial panel recognizes the legality of Evidence 2's notarial certificate. Consequently, the authenticity of the notarized content in Evidence 2's notarial certificate is recognized.

Regarding Publication Date

The collegial panel holds: Based on an understanding of the Internet Archive's archiving mechanism, it uses web crawler software programs to browse the internet and automatically save backups of website files. It is a website that regularly searches, archives, and provides public access to information from various public websites in the form of a digital library, with over 20 years of web page archives available for access. The Internet Archive assigns a URL to each web page file it crawls and archives, recording the web page's crawling and archiving time. However, elements such as images within the web page can be crawled multiple times separately, with their respective crawling times recorded. Therefore, the crawling time of a web page shown by the Internet Archive and the crawling time of the images and other elements it contains are not necessarily consistent. At the same time, Attachment 1 (Zhejiang Provincial High People's Court Civil Judgment (2022) Zhe Min Zhong No. 117) determined that on the Internet Archive website, archived files are assigned a URL in Format 1, and the "[year in yyyy][month in mm][day in dd][time code in hh:mm:ss]" information in the URL address is the specific archiving time of the archived file.

In this case, the notarization process recorded in Evidence 2's notarial document shows that by accessing the URL "web.archive.org," entering www.szhengtong.com/htswl_5.htm, selecting the corresponding timeline, and further clicking on specific time nodes such as "03:23:41," the corresponding web page is obtained. From this notarization process, it can be seen that the archiving time of the web page opened by Evidence 2 is the time on the timeline displayed in the upper right corner of the web page. That is, the archiving time of the web pages shown on pages 5-9 of Evidence 2 is February 4, 2005; the archiving time of the web pages shown on pages 13-16 is February 15, 2008; and the archiving time of the web pages shown on pages 20-23 is January 17, 2010. These archiving times are consistent with the times reflected in the numeric fields "20050204032341," "20080215011457," and "20100117050340" in URL 1, URL 2, and URL 3 displayed on the above web pages. This aligns with the determination in Attachment 1 that "the '[year in yyyy][month in mm][day in dd][time code in hh:mm:ss]' part of the URL address is the specific archiving time of the archived file." The collegial panel recognizes that the "[year in yyyy][month in mm][day in dd][time code in hh:mm:ss]" part of the URL address on the Internet Archive website is the specific archiving time of the archived file.

Furthermore, pages 10-11, 17-18, and 24 of Evidence 2 are the corresponding images opened in new tabs from the above three web pages. The URLs of these images are URL 4, URL 5, URL 6, URL 7, and URL 8, respectively. The requester claims their archiving times are December 14, 2004; December 11, 2005; December 14, 2004; December 14, 2004; and December 31, 2010, respectively. The patentee argued that if the numeric field in the URL is considered the archiving time of the archived file, then the numeric field in the URL of the image on page 10 of Evidence 2 shows a time earlier than the archiving time of the corresponding web page, which is illogical. Additionally, the URLs on pages 10 and 11 have an "im" suffix after the numeric field, and its meaning is unclear. In this regard, the collegial panel holds: Regardless of whether the above URLs have an "im" suffix after the numeric field, and regardless of the meaning of this suffix, based on the systematic and consistent nature of the Internet Archive website's mechanism for assigning URL addresses to its archived files, the URLs of the above three web pages and the URLs of the images within these three web pages should adopt the same URL address naming rules and format. Otherwise, it would lead to confusion in the addresses of archived files within the same website. Therefore, the requester's claim that the corresponding numeric field in the above image URLs represents the crawling and archiving time of those images aligns with the general understanding of website management mechanisms. In the absence of contrary evidence from the patentee sufficient to overturn the requester's above claim, the collegial panel recognizes it. As for the patentee's argument that the archiving time of an image should not be earlier than the archiving time of the web page it is on, the collegial panel holds that the above times are the crawling and archiving times of the web page and its images, respectively, not their generation times. According to the Internet Archive website's archiving mechanism, elements within a web page (images, videos, etc.) may be crawled multiple times. Therefore, it is consistent with normal website crawling and archiving logic for the archiving time of a web page and the archiving time of its images to be inconsistent. The patentee's above opinion is not supported by the collegial panel.

Therefore, the crawling and archiving times of the relevant web pages in Evidence 2... all predate the filing date of this patent, and their content can be used as prior art for this patent to assess inventiveness.

...

Feng Shangjie (Gasoll Feng)