`
`
`
`
`
`
`EDWARD G. POPLAWSKI (SBN 113590)
`epoplawski@wsgr.com
`OLIVIA M. KIM (SBN 228382)
`okim@wsgr.com
`TALIN GORDNIA (SBN 274213)
`tgordnia@wsgr.com
`STEPHANIE C. CHENG (SBN 319856)
`stephanie.cheng@wsgr.com
`WILSON SONSINI GOODRICH &
`ROSATI
`Professional Corporation
`633 West Fifth Street, Suite 1550
`Los Angeles, CA 90071
`Telephone: (323) 210-2901
`Facsimile: (866) 974-7329
`
`Attorneys for Defendant
`QUALYS INC.
`
`RYAN R. SMITH (SBN 229323)
`rsmith@wsgr.com
`CHRISTOPHER D. MAYS (SBN 266510)
`cmays@wsgr.com
`WILSON SONSINI GOODRICH &
`ROSATI
`Professional Corporation
`650 Page Mill Road
`Palo Alto, CA 94304-1050
`Telephone: (650) 493-9300
`Facsimile: (650) 493-6811
`
`
`
`IN THE UNITED STATES DISTRICT COURT
`
`FOR THE NORTHERN DISTRICT OF CALIFORNIA
`
`OAKLAND DIVISION
`
`
`FINJAN LLC,
`
`
`Plaintiff,
`
`
`
`v.
`
`
`QUALYS INC.,
`
`
`Defendant.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`CASE NO.: 4:18-cv-07229-YGR
`
`DECLARATION OF DR. SYLVIA D.
`HALL-ELLIS IN SUPPORT OF
`QUALYS’S OPPOSITION TO
`FINJAN’S MOTION FOR SUMMARY
`JUDGMENT
`
`)
`)
`)
`)
`)
`)
`)
`)
`)
`)
`)
`
`
`
`CASE NO. 4:18-cv-07229-YGR
`
`
`
`DECLARATION OF DR. SYLVIA D. HALL-ELLIS
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 2 of 10
`
`
`
`I, Dr. Sylvia D. Hall-Ellis, declare as follows:
`
`1.
`
`I have been retained as an expert by Wilson, Sonsini, Goodrich, & Rosati (“WSGR”),
`
`1
`
`2
`
`3
`
`on behalf of Defendant Qualys Inc. (“Qualys”). I make this declaration in support of Qualys’s
`
`4
`
`Opposition to Plaintiff Finjan LLC’s (“Finjan”) Motion for Summary Judgment (D.I. 192). I have
`
`5
`
`personal knowledge of the facts set forth in this declaration and, if called as a witness, could and
`
`6
`
`would competently testify thereto.
`
`7
`
`2.
`
`I am currently an Adjunct Professor in the School of Information at San José State
`
`8
`
`University. I obtained a Master of Library Science from the University of North Texas in 1972 and
`
`9
`
`a Ph.D. in Library Science from the University of Pittsburgh in 1985. Over the last fifty years, I
`
`10
`
`have held various positions in the field of library and information resources. I was first employed as
`
`11
`
`a librarian in 1966 and have been involved in the field of library sciences since, holding numerous
`
`12
`
`positions.
`
`13
`
`3.
`
`I am a member of the American Library Association (“ALA”) and its Association
`
`14
`
`for Library Collections & Technical Services (“ALCTS”) Division, and I served on the Committee
`
`15
`
`on Cataloging: Resource and Description (which wrote the new cataloging rules) and as the chair of
`
`16
`
`the Committee for Education and Training of Catalogers and the Competencies and Education for a
`
`17
`
`Career in Cataloging Interest Group. I also served as the founding Chair of the ALCTS Division’s
`
`18
`
`Task Force on Competencies and Education for a Career in Cataloging. Additionally, I served as
`
`19
`
`the National Chair for the ALA Office of Diversity’s Committee on Diversity, a member of the
`
`20
`
`REFORMA National Board of Directors, and a member of the Editorial Board for the ALCTS
`
`21
`
`premier cataloging journal, Library Resources and Technical Services. Currently I serve as a Co-
`
`22
`
`Chair for the Library Research Round Table of the American Library Association.
`
`23
`
`4.
`
`I have also given over one-hundred presentations in the field, including several on
`
`24
`
`library cataloging systems and Machine-Readable Cataloging (“MARC”) standards. My current
`
`25
`
`research interests include library cataloging systems, metadata, and organization of electronic
`
`26
`
`resources.
`
`27
`
`28
`
`
`
`CASE NO. 3:18-cv-07229-YGR
`
`
`
`1
`
`DECLARATION OF DR. SYLVIA HALL-ELLIS
`
`
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 3 of 10
`
`
`
`
`1
`
`5.
`
`I have reviewed Finjan’s motion for summary judgment and disagree with their
`
`2
`
`assertions regarding the authenticity of the scanned copy of Dr. Solomon’s Antivirus Toolkit for
`
`3
`
`Windows and DOS by Alan Solomon (“DSAVT”) I analyzed in this case.
`
`4
`
`6.
`
`I previously provided an expert report regarding DSAVT in another case, Finjan,
`
`5
`
`Inc. v. Juniper Networks, Inc., U.S. District Court, Northern District of California, 3:17-cv-05659-
`
`6
`
`WHA (“Juniper Case”). A true and correct copy of the expert report I provided in that case is
`
`7
`
`attached to the Declaration of Christopher Mays (“Mays Decl.”) as Exhibit 37. The opinions I
`
`8
`
`rendered about DSAVT in the Juniper Case are materially the same as the opinions I rendered about
`
`9
`
`DSAVT in this matter.
`
`10
`
`7.
`
`In preparing my expert report in the Juniper Case, I obtained from Juniper’s counsel
`
`11
`
`a physical copy of DSAVT as well as a scanned copy. At that time, I compared the physical copy
`
`12
`
`of DSAVT with the scanned version and determined that the two copies were substantially the same.
`
`13
`
`8.
`
`I also prepared an expert report regarding DSAVT (and other references) in this case.
`
`14
`
`A true and correct copy of the expert report I provided in this case is attached to the Mays Decl. as
`
`15
`
`Exhibit 34. In preparing my expert report for this matter, I received a scanned copy of DSAVT
`
`16
`
`from Qualys’s counsel. I reviewed the scanned copy of DSAVT that I received from Qualys’s
`
`17
`
`counsel and compared it to the scanned copy of DSAVT that I received from Juniper’s counsel in
`
`18
`
`the Juniper Case. To the best of my knowledge, the two versions appear identical.
`
`19
`
`9.
`
`To the best of my knowledge, both expert reports are true and correct to my
`
`20
`
`knowledge. I stand behind my analysis in both expert reports under the penalty of perjury.
`
`21
`
`10.
`
`Additionally, I have been informed that the scanned copy of DSAVT I received in
`
`22
`
`this case was provided to Qualys’s Counsel by Juniper’s Counsel.
`
`23
`
`11.
`
`As I explained in my expert reports in both the Juniper Case and this matter, it is my
`
`24
`
`opinion that DSAVT was published and accessible to the public no later than April 3, 1996. See
`
`25
`
`Exhibit 37 at ¶¶ 58-61; Exhibit 34 at ¶¶ 56-61. I base my opinion on the DSAVT reference itself,
`
`26
`
`the MARC record attached to my expert report as Attachment 2a, and my experience as a librarian.
`
`27
`
`28
`
`
`
`
`
`12.
`
`As explained in my expert report in this matter, ¶ 56 states:
`
`Document 2 is a manual, Dr. Solomon’s Antivirus Toolkit for Windows and
`DOS, by Alan Solomon (hereafter, “Solomon”) and issued by S&S Software
`
`2
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 4 of 10
`
`
`
`
`International in 1995. Exhibit 1002 is a true and correct copy of the title page,
`title page verso, table of contents, and contents. I obtained this book via counsel
`and Exhibit 1002 is a copy of the document submitted in this matter
`(QUALYS00002805 through QUALYS00003045). Specifically, the text is
`complete; no pages are missing, and the text on each page appears to flow
`seamlessly from one page to the next; further, there are no visible alterations to
`the document. Exhibit 1002 is a true and correct copy in a condition that creates
`no suspicion about its authenticity.
`
`The DSAVT reference itself indicates that it was issued by S&S Software International in 1995.
`
`For example, DSAVT lists a copyright date of 1995:
`
`
` D.I. 192-3 at QUALYS00002807. DSAVT also indicates that it is “Edition 3.0” with a stated date
`
`of “November 1995”:
`
`
`
`
`
`Id. In my personal experience, statements such as these from within a reference itself are generally
`
`reliable indications of approximately when a reference was published, and in this case these
`
`statements concur with my ultimate opinion that DSAVT was published and publicly accessible no
`
`later than April 3, 1996.
`
`
`13. My opinion that DSAVT was published and publicly accessible no later than April
`
`3, 1996 is further supported by the MARC record that I located for this reference. MARC is an
`
`
`
`3
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 5 of 10
`
`
`
`acronym for Machine-Readable Cataloging. As I noted in my expert report at ¶ 34, since the early
`
`1
`
`2
`
`1970s and continuing to the present day, MARC has been the primary communications protocol for
`
`3
`
`the transfer and storage of bibliographic metadata in libraries. Almost every major library in the
`
`4
`
`world is MARC-compatible. MARC records indicate publication dates for references as I detailed
`
`5
`
`in ¶ 36 of my expert report. Moreover, the MARC record creation date reflects the date on which,
`
`6
`
`or shortly after which, the item was first acquired or cataloged. See ¶42 of my expert report.
`
`7
`
`14.
`
`The MARC record for DSAVT indicates that it was first published by S & S
`
`8
`
`International PLC in 1995 in Aylesbury, England. D.I. at 192-4. The MARC record “CTRY” field
`
`9
`
`lists “enk,” which is the MARC abbreviation for England. Line 260 of the MARC record lists
`
`10
`
`Aylesbury, Bucks., a reference to the town of Aylesbury (which is in Buckinghamshire, England).
`
`11
`
`Line 260 also lists S & S International PLC as the publisher and a publication date of 1995. See
`
`12
`
`Attachment 2a to my expert report.
`
`13
`
`15.
`
`The MARC record for DSAVT is consistent with the information in DSAVT itself.
`
`14
`
`For example, line 245 of the MARC record lists the title of the document as “Dr. Solomon’s anti-
`
`15
`
`virus toolkit for Windows and DOS [manual].” See Attachment 2a to my expert report. This is
`
`16
`
`consistent with the title on DSAVT itself (see D.I. at 192-4 at QUALYS00002805) as well as the
`
`17
`
`statement in DSAVT that it is a manual that “describes Version 7.5 of Dr Solomon’s Anti-Virus
`
`18
`
`Toolkit for Windows and Dr Solomon’s Anti-Virus Toolkit for DOS.” See D.I. at 192-4 at
`
`19
`
`QUALYS00002807. Likewise, the back cover of DSAVT lists S & S International PLC as the
`
`20
`
`publisher for DSAVT in the United Kingdom with a location of Aylesbury, Buckinghamshire in
`
`21
`
`England, the same as the MARC record. See D. I. at 192-4 at QUALYS00003045.
`
`22
`
`16.
`
`During my deposition, Finjan’s counsel asked me questions about two entries in the
`
`23
`
`DSAVT MARC record. First, that line 300 of the MARC record lists twenty pages (noted by the
`
`24
`
`roman numerals “xx”) plus 222 pages (for a total of 242 pages), while the DSAVT reference itself
`
`25
`
`is 241 pages (including covers, tables, and indices). However, in my opinion such minor
`
`26
`
`discrepancies between a published document and its MARC record are common and do not alter my
`
`27
`
`opinion that the MARC record refers to DSAVT. As I stated while testifying at my deposition, the
`
`28
`
`MARC record appears to have been created based on the United Kingdom printing of DSAVT. It
`
`
`
`
`
`4
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 6 of 10
`
`
`
`is not unusual in my experience for page numbers to slightly deviate between different printings of
`
`1
`
`2
`
`a book, particularly when dealing with United States-based reprintings of English versions of books.
`
`3
`
`For example, the MARC record indicates a printing size of “23 cm,” and a printing size in the United
`
`4
`
`States that differs from 23 cm would impact the number of pages without changing the content of
`
`5
`
`the printing in any material way. In such a circumstance, it would not be unusual for United States
`
`6
`
`librarians to rely on the U.K. MARC record even for the United States printing of a publication. As
`
`7
`
`I discuss in my expert report at ¶ 43, “Once one library has cataloged and indexed a publication by
`
`8
`
`creating a MARC record for that publication, other libraries that receive the publication do not create
`
`9
`
`additional MARC records—the other libraries instead rely on the original MARC record.” This
`
`10
`
`practice “allows librarians around the world to know that a particular MARC record is
`
`11
`
`authoritative.” Id. Thus, the fact that the copy of DSAVT that I reviewed is 241 pages while the
`
`12
`
`MARC record indicates 242 does not alter my opinion that the MARC record refers to DSAVT,
`
`13
`
`particularly given the consistency of the remaining information in the MARC record that I discussed
`
`14
`
`above. At most, it indicates to me that the original MARC record for DSAVT was created for a
`
`15
`
`U.K. printing of the book while the copy I reviewed was for a U.S. printing of the same book.
`
`16
`
`17.
`
`During my deposition, Finjan’s counsel also asked me about line 500 of DSAVT’s
`
`17
`
`MARC record which states “On cover: “Complete virus Protection: Prevention, Detection and
`
`18
`
`Repair.” The copy of DSAVT that I reviewed may not list this phrase on the cover, although it is
`
`19
`
`difficult to tell given that the copy I reviewed is in black and white only and does not clearly depict
`
`20
`
`every aspect of the cover. Even if the phrase is not on the cover of DSAVT, however, my opinion
`
`21
`
`that the MARC record refers to the same DSAVT book that I reviewed remains the same. As I
`
`22
`
`testified at my deposition, it would not be unusual for cover elements to change between printings
`
`23
`
`of the same book in different countries. Thus, while the U.K. version of DSAVT (on which the
`
`24
`
`MARC record appears to have been created) may have included the phrase listed in field 500 on its
`
`25
`
`cover, the U.S. version may not have included that phrase on the cover. In such a case, it is not
`
`26
`
`unusual for librarians to rely on the original MARC record notwithstanding such minor differences,
`
`27
`
`because the book itself remains the same even if certain cover aspects change. Critically, this means
`
`28
`
`that the date of publication also remains the same. As I testified, this potential discrepancy between
`
`
`
`
`
`5
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 7 of 10
`
`
`
`the MARC record and DSAVT itself is minor, not unusual, and supports my opinion that the MARC
`
`1
`
`2
`
`record simply refers to the U.K. printing of DSAVT while the copy I reviewed in this case was the
`
`3
`
`United States printing of the same book.
`
`4
`
`18.
`
`During my deposition, I was asked whether I had seen the physical book that is
`
`5
`
`identified in the MARC record I discuss above. I testified that I had not seen a copy of that book
`
`6
`
`and could not obtain one. To clarify, I was referring to the work in preparing my expert report for
`
`7
`
`this case, and not the work I performed in preparing my expert report in the Juniper Case. As I
`
`8
`
`noted above, while preparing my expert report in the Juniper Case I was provided a physical copy
`
`9
`
`of DSAVT, which I compared to the digital version I opined on in that case and confirmed that they
`
`10
`
`were the same document.
`
`11
`
`19.
`
`I also understand that Finjan is challenging the authenticity of the Mounji and
`
`12
`
`Thomson references analyzed in this case. As to the authenticity of Mounji, as discussed in my
`
`13
`
`expert report in ¶ 103:
`
`I downloaded the Mounji technical paper from the CiteSeerx website. Exhibit
`1008 is a true and correct copy of the technical paper (pages 1-37).
`Specifically, the text of the article is complete; no pages are missing, and the
`text on each page appears to flow seamlessly from one page to the next;
`further, there are no visible alterations to the document. Exhibit 1008 was
`found within the custody of a well-known and recognized digital repository
`– a place where, if authentic, a copy of this technical paper would likely be.
`Exhibit 1008 is a true and correct copy in a condition that created no suspicion
`about its authenticity.
`
`As to the authenticity of Thomson, as discussed in my expert report in ¶ 116:
`
`I downloaded the Thomson technical paper from the Semantic Scholar
`website. Exhibit 1010 is a true and correct copy of the technical paper (pages
`133-146). Specifically, the text of the article is complete; no pages are
`missing, and the text on each page appears to flow seamlessly from one page
`to the next; further, there are no visible alterations to the document. Exhibit
`1010 was found within the custody of a well-known and recognized digital
`repository – a place where, if authentic, a copy of this technical paper would
`likely be. Exhibit 1010 is a true and correct copy in a condition that created
`no suspicion about its authenticity.
`
`Given the completeness of the Mounji and Thomson references I retrieved from CiteSeerx and
`
`Semantic Scholar, respectively, I have no reason to believe that these references are not authentic.
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`
`
`6
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 8 of 10
`
`
`
`Both CiteSeerx and Semantic Scholar are authoritative repositories, where an authentic copy of the
`
`reference would be.
`
`20.
`
`Also, I understand from reviewing Finjan’s motion for summary judgment that
`
`Finjan takes issue with my reliance on three document repositories discussed in my report; namely,
`
`CiteSeerx, ResearchGate, and Semantic Scholar. As I stated in my expert report in ¶19, “In preparing
`
`this report, I used authoritative databases, such as the OCLC WorldCat, the Library of Congress
`
`Online Catalog, the Internet Archive (Wayback Machine), CiteSeerx, IEEE Xplore, ResearchGate
`
`Semantic Scholar, Google Scholar, and Scopus, to confirm citation details of the various
`
`publications discussed.” These repositories are widely used by librarians to locate publications as
`
`well as information about them. Among librarians, these repositories are considered “industry
`
`standard,” reliable, and authoritative.
`
`21. As I stated in my expert report at ¶ 20, CiteSeerx is:
`An evolving and scientific digital and search engine focused primarily on the
`literature in computer and information science. CiteSeerx aims to improve the
`dissemination of scientific literature and to provide improvements in the
`functionality, usability, availability, cost, comprehensiveness, efficiency, and
`timeliness in the access of scientific and scholarly knowledge . . . CiteSeerx has
`developed new methods and algorithms to index PostScript and PDF research articles
`on the World Wide Web.
`
`In fact, CiteSeerx was created because use of its predecessor, CiteSeer (created in 1997), “grew to
`
`index over 750,000 documents and served over 1.5 million requests daily, pushing the limits of the
`
`system’s capabilities,” thus demonstrating how respectable and widely CiteSeer, and consequently,
`
`CiteSeerx, was and is used by professionals in library sciences. Id.; D.I. 192-6. While these three
`
`repositories were not created for the sole purpose of indexing references and providing dispositive
`
`dates for when the reference was published, library science professionals do regularly make use of
`
`the information in the repositories to determine when a reference was published.
`
`22.
`
` As another example, ResearchGate is an authoritative repository that is the largest
`
`academic social network in terms of active users that scientists and researchers use to share papers,
`
`ask and answer questions, and find collaborators. A European social networking site for scientists
`
`and researchers, ResearchGate requires that individuals who wish to join and participate in activities
`
`have an email address at a recognized institution or to be manually confirmed as a published
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`
`
`
`
`7
`
`
`
`Case 4:18-cv-07229-YGR Document 195-3 Filed 05/10/21 Page 9 of 10
`
`
`
`researcher in order to sign up for an account. Scientists and researchers use the ResearchGate site to
`
`1
`
`2
`
`share papers, ask and answer questions, and find collaborators. Participants in ResearchGate may
`
`3
`
`post information about their work, abstracts, and full-text documents. A researcher or scholar who
`
`4
`
`created a document independently or in partnership with colleagues knows when the work was
`
`5
`
`prepared and when it was published, and in the case of conference papers, where and at what time
`
`6
`
`a public presentation took place. As discussed in my deposition testimony, ResearchGate is a
`
`7
`
`repository for researchers to give evidence of what we have done and what we are working on to
`
`8
`
`our colleagues.
`
`9
`
`23.
`
`During my deposition, I was asked how I could determine when an article from
`
`10
`
`ResearchGate was publicly accessible. I testified that ResearchGate is not necessarily a source of
`
`11
`
`public availability that is reliable as other means. Although ResearchGate is not the most reliable
`
`12
`
`method for determining when a document is publicly accessible, it is a resource that can be used to
`
`13
`
`determine when a reference was available to the public since the author or the individual uploading
`
`14
`
`it to ResearchGate has to input such information. Determining the exact citation for a document
`
`15
`
`may require checking a digital repository, the curriculum vitae of one or more authors, or similar
`
`16
`
`documentation to determine the exact title, a publication citation, or conference presentation
`
`17
`
`information. In selected cases, information posted in a scholarly repository by the author can be
`
`18
`
`reliable and easy to find. Researchers and scholars know what they write, when they do so, and
`
`19
`
`where their work is presented and published. In the case of the Thomson document, ResearchGate
`
`20
`
`was confirmation from the author about the document and its date of public accessibility. As
`
`21
`
`discussed in ¶ 117 of my report, I based my opinion that Thomson was publicly available no later
`
`22
`
`than May 31, 1997, on the ResearchGate opening page, which indicated that Thomson was
`
`23
`
`published in May 1997.
`
`24
`
`24.
`
`During my deposition, I was asked how I determined when an article from CiteSeer
`
`25
`
`was publicly accessible, and I testified that you cannot necessarily tell from CiteSeer alone. This is
`
`26
`
`consistent with my testimony that dates listed in digital repositories alone are something one should
`
`27
`
`accept at face value without the appropriate authentication and checking. I also testified that
`
`28
`
`CiteSeer’s information about publication dates is dependent on whether the author or authors decide
`
`
`
`
`
`8
`
`
`
`2
`
`3
`
`4
`
`5
`
`look at the face of the reference for a publication date. As I stated in my report at 1] 104, “The cover
`
`of the Mounji technical report has a date of May 27, 1994. The CiteSeer‘r indicates a ‘1994’ date.
`
`For these reasons, it is my opinion that Exhibit 1008 was published and accessible to the public no
`
`later than May 27, 1994, or shortly thereafter.” My report makes clear that my opinion that Mounji
`
`6 was publicly available by May 27, 1994 is based on the fact that face of the Mounji reference listed
`
`7
`
`8
`
`9
`
`10
`
`1]
`
`12
`
`13
`
`14
`
`a May 27, 1994 date and the CiteSeer“ landing page for Mounji listed a 1994 date. This indicates
`
`that I did not rely on CiteSeer" alone to determine the date of public accessibility for Mounji. The
`
`fact that CiteSeert indicates a publication date that is consistent with the date of publication on the
`
`face of the Mounji reference supports my opinion that Monnji was published and accessible to the
`
`public no later than May 27, 1994.
`
`25.
`
`It does not matter that the repositories were not in existence when some of the
`
`references at
`
`issue were published because if a reference is in the repository,
`
`the industry
`
`understanding is that these repositories only house authentic copies of a reference. For example,
`
` to include that information when publishing their article on CiteSeer, otherwise, one would have to
`
`15 ResearchGate allows authors who meet the criteria for participation to post abstracts full texts of
`
`16
`
`their writings, or a mechanism for readers to send a request for a copy of the full text.
`
`I declare under the penalty of perjury under the laws of the United States of America that each of
`
`the above statements is true and correct. Executed on May 10, 2021, in Denver, CO.
`
`
`
`