`FOR THE NORTHERN DISTRICT OF ILLINOIS
`EASTERN DIVISION
`
`Case No. 04 C 4240
`
`Hon. Samuel Der-Yeghiayan
`
`) ) ) 3 )
`
`)
`
`) ) ) ) )
`
`Windy City Innovations, LLC,
`a Delaware Compan 3/,
`
`Plaintiff,
`
`V.
`
`_
`
`America Online, Inc.,
`a Delaware Corporation,
`
`Defendant.
`
`EXPERT REPORT OF BRUCE M. MAGGS
`
`1.
`
`My name is Bruce Maggs. I have been retained by the defendant in this action,
`
`America Online Inc. (‘‘AOL’-‘) to consult on technical issues pertaining to this lawsuit and to
`
`prepare a report that provides a summaiy of the testimony that I am prepared to give at trial, if
`
`called to testify. This document constitutes my expert report on the validity and enforceability of
`
`U.S. Patent 5,956,491.
`
`2.
`
`In summary, frist, this report explains my opinion as to why the ’49l patent is
`
`invalidated by the prior art. It also explains my opinions that the patent fails to disclose the
`
`claimed i.nvention’s “best mode.” Finally, the report indicates my opinion that the Grail;
`
`software, co-authored by the named inventor prior to the inventioll. but DOT diSC10S€d $0 1115
`
`patent ofiice, is non-cuniulati vs.
`
`3.
`
`My curriculum vitae is attached hereto as Exhibit A. In Sl.lII'l.lTlElI'}’, in academia, I
`
`am a tenured Professor of Computer Science in the School of Computer Science at Carnegie
`
`Mellon University.
`
`I joined the faculty as an Assistant Professor in January 1994, was promoted
`
`to Associate -Professor in July 1997, was given tenure in July 1999, and was promoted to (full)
`
`Facebook's Exhibit No. 1017
`Page 1
`
`
`
`Professor in 2004.
`
`I also held the position ofVisiting Associate Professor in the Electrical
`
`Engineering and Computer Science Department at the Massachusetts Institute of Technology
`
`from September 1998 through January 1999.
`
`4.
`
`With respect to my industry experience, I helped launch Akatnai Technologies in
`
`1998. Alcamai provides content delivery services for many of the world’s most popular websites.
`
`I served as a Senior Research Scientist for Akamai from January 1999 through March 1999, and
`
`as Vice President for Research and Development from April 1999 through December 1999. I am
`
`currently the Vice President for Research at Akarnai and have held this position since January
`
`2000. In addition, Iwas a Research Scientist at NBC Research Institute, Inc., from September
`
`1990 through January 1994.
`
`5.
`
`I received my Doctorate degree in Computer Science from the Massachusetts
`
`Institute of Technology in 1989, my Masters of Science degree in Electrical Engineering and
`
`Computer Science from the Massachusetts Institute of Technology in 1986, and my Bachelor’s
`
`of Science degree in Computer Science from the Massachusetts lnstitute of Technology in 1985.
`
`6.
`
`Additional information concerning my teaching experience, publications, surveys,
`
`manuscripts, distinguished-lecture-series speeches, keynote addresses, invited lectures, awards,
`
`grants/contractsffellowships, committee service, technical advisory boards, and the patents for
`
`which I am a named inventor, is set forth in my Curriculum Vitae.
`
`7.
`
`I was also employed as a computer programmer at the University of Illinois at
`
`various times between 1979 and 1983. At the University, I wrote numerous programs for the
`
`PLATO computer system, including educational programs and recreational programs. One of
`
`these recreational programs was a rnulti-player “dungeons and dragons" game (or “MU'D”)
`
`called Avatar. Avatar, among its other features, included communications functionality that
`
`Facebook's Exhibit No. 1017
`Page 2
`
`
`
`allowed for messaging to a number of users simultaneously as well as messaging between two .
`
`individuals. These messages could include both text and graphical images. PLATO is now
`
`known as NovaNET. I have played Avatar running on NovaNET over a public TCPFIP network.
`
`I am familiar with many computer communications programs, including ruunerous “chat”
`
`and messaging systems. I am familiar with e-mail standards and protocols such as SMTP, POP,
`
`D\rl'.A.P, and MIME, and have taught courses at Clai-negie Mellon University on these standards
`
`and protocols.
`
`I am familiar with and have used other PLATO programs preceding Avatar,
`
`including “empire,” “tal.komatic," and “term talk," which provided similar communications
`
`fiinc-tionality.
`
`8.
`
`I have testified before as an expert witness in the lawsuit captioned Lax-rnark
`
`International. Inc. v. Static Control Components. Inc, No. 02-S71-KSF, United States District
`
`Court for the Eastern District of Kentucky.
`
`'.\
`
`9.
`
`I am being compensated at the rate of $300 per hour for my work in this case.
`
`10.
`
`In preparing this report, I have thoroughly reviewed a number of documents and
`
`other materials, and have otherwise prepared for the report as discussed below. The pertinent
`
`documents have been attached as exhibits or are included in the attached CD-R and DVD-R.
`
`The documents that I have reviewed include:
`
`I U.S. Patent 5,956,491, and the documents that comprise the “file history" of this patent
`
`(including the references cited therein) (Ex. l).
`
`The source code appendix to the ’491 patent. (Ex. 2).
`
`An America Online service called "Road Trips.” I reviewed versions l.3, 1.30, 1.64, 2.0,
`
`and 2.1 of the prirnary source code file for Road Trips, which was called “tour.c." (EX. 3),
`
`I also reviewed CVS logs for the files tour.c (Ex. 4) and tou:r2.<: (Ex. 5), and a set of
`
`printed ‘‘screenshots‘‘ of the forms used by Road Trips. (Ex. 6).
`
`I also reviewed AOL 2.5
`
`client software (Ex. 7?) and a list of forms and form creation dates. (Ex. 7).
`
`Facebook's Exhibit No. 1017
`Page 3
`
`
`
`Certain Netscape Cornrnunications software known as "Netscape Chat” version 1.0.1.8
`(32-1310 and 1.01 €16-bit), and Netscape Navigator version 1.2.2 (16-hit), (Ex. 8) and
`
`I
`
`associated source code and design specifications. (Ex. 9). I also compiled and
`
`configured an IRC server, called ircd, from Undernet, the source code for which is
`
`attached. (Ex. 10).
`
`Network Worldng Group Request for Comments (RFC) 1459, “Internet Chat Relay
`
`Protocol”, by J. Oikarinen and D. Reed, May 1993.
`
`(Ex. 11).
`
`Sun Microsystems‘ Hotlava Browser, the applet viewer fiorn Sun's JDK version 1.0, and
`
`Netscape Navigator version 2.0 (Ex. 12).
`
`Certain CompuServe software entitled Cor_npuServe Producer, as well as CompuServe’s
`
`“WIl\TCBV[.E}CE” client program, and various associated source code files. (Ex. 13).
`
`Several versions of software known as Gtaik, including versions 1.6.8, 1.6.6, and 1.6.4
`
`for the Unix operating system; version “1.9z1.4" for DOS, and version 2.2.3 for OSIZ,
`
`which I understand vvas prepared in part by Daniel Marks, the named inventor of the ‘£191
`
`patent, or were derived from software written by Mr. Marks, and “G'I'U.X,” another
`software program Ga. 14).
`
`The “Gtallc Owners Manual” dated July 14, 1995. (Ex. 15).
`
`An article by Prof. Judith Donath and Niel Robertson, entitled “The Sociable Web,“
`
`posted on the World—Wide Web in October of 1994 (Ex. 16) and The following related
`
`docurneriratioo (Ex. 16):
`
`l DXO l 7
`DX013
`DX019
`' DX021
`
`I DXO24
`f DX086
`
`Sociable Web Article
`‘ Sociable Web Article gno pictures)
`hlet
`1' World Wide Web Conference a
`Sociable Web Article HTML source code
`
`I World Wide Web Conference pamphlet
`READIVIEIXT file
`
`.
`
`|
`
`Online vmions of “The Sociable Web,” found at
`
`hm;:£"s1'ng.media.git.edtI1’geo_nle!'J udith/SocialwebfSociable-W ebhtml and at
`
`ht_rg:lfarchive.ncsa.uiuc.edu/SDGz'1'1'94fProceedings.fCSCwfclonathfsociablew eb.l1t1nl
`
`(Ex. 17),, and the HTML source files for each page (Ex. 18), and also online files found
`
`in the web directory h_ttp;/r'srna.niedia.mit.flpeo1_3_1eijudithi’Social Web;’Pi;,»_' (Ex. 13).
`
`A transcript of the May 25, 2005, deposition of Niel Robertson. (Ex. 19).
`
`The Mosaic User Authentication Tutorial,
`
`hL.E'.m'100h00.DCS3.UiuC.Bd1.lJ'd0Cs/t‘LJ'tD1'ia1S)'u5aT\ .hnnl, dated 9-27295.
`
`(Ex.20_1.
`
`Facebook's Exhibit No. 1017
`Page 4
`
`
`
`Upgrading NCSA H'I'I"Pd, imzz‘,-’hooboo.ncsa.uiuc.eduJdocs/U_ggade.ht1nl, dated 08-01-
`
`95 fEx. 21).
`
`NCSA Mosaic Version History,
`
`h
`
`flfwww.ncsa.uiuc.edu!Divisio
`
`bl'cAfi“airsfMosaicHistorw‘historv.hrm1 (Ex. 22).
`
`A log entry frorn November 18, 1993, on the NCSA Mosaic Website, indicating that
`
`Mosaic 2.0 was availableon that date,
`ht_Ipflarchive.nc§a.uiuc.edufSDG!Sofiwa.rea’MosaiC./Docsfold-whats-newfwhats-n ew-
`
`l 193.htn1l (Ex. 23).
`
`An article by Markus Sohlenkarnp and Greg Chewlos entitled “Integrating
`
`Commimicafions, Cooperation, and Awareness: The DIVA Virtual Office Environment."
`Proceedings of the gonfeficc on Qonsputer Cooperative Worlc October 22-26, 1994.
`
`(Ex.'24).
`
`U.S. Patent 5,880,731 (a Microsoft patent) (Ex. 25).
`
`To the extent not encompassed in the above, the contents of the production CD5 that bear
`
`production numbers DM 50-53, and WCI 001589, 002859, and 002860, which include
`
`additional source code not included in the patent. These documents have not been
`
`attached to this report per the protective order.
`
`I have otherwise prepared for this report as follows:
`
`I reviewed and analyzed the C progratnnling language source code contained in
`
`ll.
`
`12.
`
`die ‘£191 appendix. I obtained an electronic copy of a transcription of this code and prepared
`
`executable software based on this code. When compiled, the source code produces two
`
`executable files, a “server” program called “uc" and a “client” program called “ucc”. Both of
`
`these programs are meant to execute on the same computer. {The client program connects to a
`
`server progam running on a machine named “locaJ.host”, which refers to the same machine.) I
`
`also connected to this computer running the executable software over a public TCPXIP network
`
`connection using the telnet application running on a different computer, and observed the
`
`behavior of this compiled software.
`
`I studied the client and server executables running on a
`
`computer with the Red Hat Linux 6.2 operating system, and also on a computer with the Linspire
`
`Linux operating system.
`
`Facebook's Exhibit No. 1017
`Page 5
`
`
`
`l_3.
`
`I also performed tests in which connections were made to the server program
`
`compiled from the C source code appended to the patent using a client program implemented as
`
`a Java applet found on the CD bearing the production number DM52, and executed on a different
`
`computer. The client program connected to the server program using a TCP/lP connection.
`
`a. All tests with the Java applet were performed on an isolated private
`
`network located in the offices of Banner & Witcoff consisting of a desktop
`
`computer running the Linspirem Five-Om distribution of the Linux
`
`operating system and a laptop computer rtn-ining the Windows XP
`
`operating system. The two computers were both connected to a Linksys
`
`router, model number BEFSR411 ver.3.1
`
`.
`
`Java code is run using what is known as a “Java virtual machine” (NM).
`
`The Java code on the laptop was run using a number of different JVMS.
`
`First, I used the Java applet viewer provided in Sun Microsystems Java
`
`Development Kit (JDK), version 1.0. Next, I installed the Apache web
`
`server on the Linspire server, and prepared an html document that
`
`instructed a web browser to fetch and then execute the Java applet from
`
`the same web server. -The Java applet was then tested by “downloading”
`
`both the html document and subsequently the applet using both Surfs
`
`HotJava brovgser, and also Netscape Navigator, Version 2.0.
`
`When using the Java client, it is not necessary for the client executable
`
`generated from the C code appended to the patent to also execute on the
`
`server machine. However, to test the interoperability of the Java client
`
`Facebook's Exhibit No. 1017
`Page 6
`
`
`
`and the-C client, I connected to the server simultaneously using 3 telnet
`
`session and using the Java client.
`
`14.
`
`I reviewed AOL Road Trips.
`
`a.
`
`Itested America Onli.ne's “Road Trips” service, version 2.0, and discussed
`
`I the operation of this service with one of the creators of Road Trips, lay
`
`Elinsky. To experiment with Road Trips, I registered an account with
`
`America Online (AOL), hrucemrnaggs@aol.com, and then connected via
`
`TCPIIIP to an AOL server using version 2.5 of the AOL client program,
`
`running on a Pentium 75 MHz desktop with the Microsoft Windows 3.11
`
`for Workg-roups operating system.
`
`. The installation of the AOL client software version 2.5 was creaied using
`
`a copy of the contents of an AOL CD-ROM containing installation
`
`software provided by AOL. I did not have a copy of an original AOL CD-
`
`ROM. The CD-ROM version, which has a much higher storage capacity
`
`than a floppy disk, installed a copy of IWENGDLL with a creation date
`
`SE30! 1995. Its length was 700KB.
`
`I also experimented with different versions of version 2.5 of the AOL
`
`client program on various computers, and saw no differences in behavior.
`. The first version was installed on a desktop computer (the participator
`
`computer) using an original AOL floppy disk titled “America Online FOR
`
`WINDOWSTM Version 2.5 America Online produced these floppy disks
`
`in 1995. The disk contains a single file called “SETUP.EXIE”, which,
`
`when executed, installs the client software, which consists of several files.
`
`Facebook's Exhibit No. 1017
`Page 7
`
`
`
`The creation dates of the files copied from the disk are all 6/27/1995 or
`
`earlier and the copyright notice on the disk label reads "©1993-1995
`
`America Online, Inc.’’. The ilistailation program SETUP.EXE also creates
`
`several new files and directories (e.g., mainjdx) whose creation dates are
`
`set to the date of. the installation, e.g., 6:23;’2005. None of these files are
`
`executable code or modify the operation of the software. The computer
`
`was running the Windows X? operating system.
`
`The floppy disk did not install two files, “IWENG.DI..L” and
`
`"TOOL/WWW.AOL” that are required in order for the browser/chat
`
`window of Road Trips to operate correctly on the client computer. These
`
`files were-not included in SETUP.E}CE'. on the floppy dislc because
`
`IVVENCLDLL is large (698 kilobytes) and would not fit on the disk with
`
`the other files, even when compressed.
`
`(T136 Capacity Df 3 fl°PP3’ disk is
`
`only 1.44 megabytes.) AOL client version 2.5, when installed from this
`
`floppy disk in 1995, would instead automatically retrieve the files fi-om
`
`AOL when they first used the software to access AOL using a dial-up
`
`connection or using TCP/IP. Today, however, while it is possible to
`
`connect to AOL using version 2.5 of the client software, version 2.5 is no
`
`longer fully supported, and I could no longer download these files. For
`
`this reason, Jay Blinsky provided me with copies of these files with a
`
`creation dates of 6/27f1995. He indicated that the files were taken from an
`
`old laptop computer on which AOL client software version 2.5 had been
`
`installed around 1995. I then installed these files in the same directory as
`
`Facebook's Exhibit No. 1017
`Page 8
`
`
`
`the other AOL software and then accessed AOL by starting the program -
`
`AOLEXE.
`
`I also tested the AOL client software version 2.5 installation that was
`
`copied in its entirety from Mr. Elins!cy’s old laptop computer. The latest
`
`creation date on any of these files was 6!'27x'1995.
`
`.
`
`In some cases, I made one change in a configuration file called
`
`“TCP.CCL”. This file specifies the host name and port number to connect
`
`to when accessing AOL using TCPKIP over the Internet, which was
`
`possible using version 2.5 of the client software. The host name in the
`
`original file is “AmericaOn1ine.ao1.co1n”. I was able to access AOL and
`
`use Road Trips using this hostname. The Road Trip service is installed on
`
`a server that is accessible via TCPJTP connectivity to a “BBB?” server,
`
`which sometimes is not assigned when connecting to
`
`arner-icaon]i.ne.aol.com. To ensure connectivity to a BERP server, I
`
`modified the line
`
`Netconnecl:
`
`1
`
`5190
`
`3.0 AmericaOnline.a.o1.c:n-n
`
`by changing it to
`
`Netconnect
`
`1
`
`5190
`
`10
`
`berp—nz01.clial.aol.com
`
`Jay Elinsky intfonned me that he had installed version 2.0 of the Road
`
`Trips software on today‘s production AOL system. The software is run on
`
`an AOL server. All of the code that is used to implement Road Trips is
`
`identical to the code version 2.0 as written in 1995. Mr. Elinsky added
`
`one line of source code to assist the software in operating in the current
`
`AOL environment. Specifically, the following single line was added to
`
`Facebook's Exhibit No. 1017
`Page 9
`
`
`
`tours in function “Initia1ize_A.pp1icau'on”:
`
`m*I_set_Per_Use:_sr.ream_Ins
`
`(/ *SID_CfHJ-‘.'I‘__LD*/ 0x2 0051:,
`
`/*SID_CEEAT_1‘-II*/ cxc4cca.b) ,-
`
`This line does not change the functionality of the Road Trips code.
`
`Instead, this line assists the Road Trips code in operating in the current
`
`I AOL environment.
`
`.
`
`In addition to source code, AOL services make use of “forms”. A form is
`
`a graphical script that controls the graphical user interface provided by the
`
`AOL client software. A form is interpreted by the AOL cheat software by
`
`the participator computer. It specifies, for example, where a button should
`
`appear on-"the screen and what should be transmitted to an AOL server
`
`when the user clicks on the button. A fonn also specifies ‘where a browser
`
`window should appear, where a field for entering text should appear, and
`
`where a scrolling text field should appear. An AOL server can send an
`
`“atcm" to an element of a form in order to change its appearance or
`
`behavior. For example, an AOL server can send an atom to a button
`
`created by a form indicating that it should no longer appear on the form.
`
`Road Trips used various forms. Some of these forms are depicted in
`
`printed screenshots in Ex. 6. Jay Elinsky informed me that he installed
`
`these forms on today’s AOL production system so that they could be sent
`
`to the AOL client sofiware when users today access Road Trips. Eiinsky
`
`also indicated that all of the Road Trips forms used today are dated before
`
`August of 1995. A list of forms and form creation dates which confirms
`
`this was given to me by Elinsl-ty and is attached as Ex. 7.
`
`Facebook's Exhibit No. 1017
`Page 10
`
`
`
`After accessing AOL via a TCPIIP connection over the public global
`
`TCPIIP series of networks, using the version 2.5 of the client software, I
`
`then pressed control-k, which caused an AOL “form” to appear on my
`
`screen that allowed me to enter an AOL keyword.
`
`I was provided with an
`
`AOL keyword that led to Road Trips. Jay Elinsky also connected to Road
`
`Trips at the same time.
`
`Jay Elinsky and I then participated in several tours, and exercised the
`
`features of the software. One difference between my account and Mr.
`
`E]insky’s account is that mine is a “member” account, whereas his is an
`
`“internal” -account. Hence his account has certain privileges that mine
`
`does not.
`
`Elinsky was able to create an “AOL tour”,_ a tour that a user
`
`with a member account cannot create, but that users with “iI1terna.l”_ or
`
`“overhead” accounts can create.
`
`I accessed this “AOL tour" and observed
`
`the behavior of Road Trips during this tour.
`
`I created a “member” tour
`
`and participated in this tour with Mr. Elinsky, and observed the behavior
`
`of Road Trips during this tour. Another type Road Trips tour is a “private
`
`tour." Mr. Elinslcy demonstrated the creation of a “private tour.” I joined
`
`a private tour created by Mr. Elinsky and observed the behavior of Road
`I
`
`Trips.
`
`I later created a video record which demonstrates the features ofRoacl
`
`Trips while using computers built with parts available in 1994 (in
`
`particular, systems based on the Intel P75 processor) and running the
`
`Facebook's Exhibit No. 1017
`Page 11
`
`
`
`Windows 3.11 for Workgroups operating system. A copy of this video
`
`record (on DVD) is enclosed as Ex. C.
`
`15.
`
`-I reviewed Netscape Chat.
`
`3.
`
`I used “Netscape Chat” version 1.01 (16-bit) on the same Pentium 7'5
`
`MHz computers running Microsoft Windows 3.1 1 for Workgroups
`
`operating system. Netscape Chat was installed using an original Netscape
`
`“Power Pack” CD—ROM. This CD—ROM contains a pro gram called
`
`Netscape Power Packm (Powerpaeleexe) that allows a user to install
`
`certain software, including Netscape SI!12lI'1It1aIkSTM, Netscape Chatm
`
`(version 1:01 ), Adobem Acrobatm Reader (version 2.1), App1e®
`QuickTi.me® version 2.0, and Realfitudiom Player (version 1.0.0). Power
`
`Pack can be used to install either the 32-bit version of Netscape Chat,
`
`version 1.0.1.8 (for use with Microsoft Windows 95 and later versions of
`
`Windows), or the 16-bit version, 1.01 {for use with Microsoft Windows
`
`3.1). Netscape Navigator version 1.22 (l6—bit version) was also installed
`
`on the same Pentium 75 MHz computers.
`
`I examined the behavior of the
`
`Netscape Chat program {and simultaneously the Netscape Navigator
`
`program)-by making connections from both machines using a TCPKIP over
`
`the public Internet to an [RC server that I set up.
`
`.
`
`I later created a video record in which I demonstrate the features of
`
`Netscape Chat while using the same computers and operating syst. To
`
`perform these experiments, I downloaded the Undernet IRC chat server
`
`software ircu2.9.19 from ftp.undernet.ore. A copy of this software is
`
`Facebook's Exhibit No. 1017
`Page 12
`
`
`
`attached on a CD-R as Ex. 10. The IRC chat server ircd was then
`
`configured, compiled and run on a Sun Microsystems 'SparcStatior1 4
`
`workstation, using the Solaris 1.1.2 operating system, which was installed
`
`from an original Sun Microsystems CD-ROM. Both the workstation and
`
`the operating system were available in 1994. To compile the IRC server, I
`
`made two syntactical changes to con.figu.1'e the code to compile on the
`
`Solaris C compiler; in 1.994 and 1995, different C programming language
`
`compilers supported slightly different syntaxes, and this change would
`
`have been normal to a prograrnrner of that time in order to allow this
`
`software to compile on Solaris.
`i._ added a comma at the beginning of line 1704 in file ircdfs_bsd.c.
`
`ii.
`
`added a comma at the beginning of line 726 in file ircdfs_user.c
`
`To configure the IRC server with a resolvable host name, and to verify
`
`that the connections between the IRC server and the IRC client computers
`
`were made through the Internet, Imoved it to a location remote from the
`
`Pentium computers and connected the server to a different Internet Service
`
`Provider (ISP). The [RC server was placed behind a firewall whose public
`
`name was i.rc.mooreusa.ne‘r_., and whose public IP address was
`
`64.81.139.232.
`
`.
`
`I also configured the JRC server by creating _a file ircd.conf from
`
`cxaJ:nple.con.f, a file sample confi guration provided with the IRC server
`
`sottware. Imodified example.co1:|fto change the name, port number, and
`
`Facebook's Exhibit No. 1017
`Page 13
`
`
`
`access password for the server, and to remove a number of non~rnandato1_-y
`
`configuration lines.
`
`I also reviewed the C-I-F source code for Netscape Chat and associated
`
`design documents and specifications, as well as the Internet Relay Chat
`
`(IRC} Specification, Request for Comments (RFC) 1459.
`
`16.
`
`I reviewed the Compus erve Producer 2’ Viewer system and CompuServe‘s CB
`
`Conferencing system.
`
`a.
`
`I tested "CompuServe Producer” V.198C, Copyright 1996, using the
`
`foliowing hardware configuration. The Producer software, program
`
`csprod.exe, was run on a Packard Bell personal computer. A sticker on
`
`the computer indicates that its model is “LEGEND 2150 ISOMI-Iz DXXSO
`
`486 processor PC with Microsoft Windows operating system." The
`
`operating system installed on the machine was Microsoft Windows 3.1.
`
`A fact sheet that accompanied the computer also listed the model as
`
`“LEGEND 2150 MULTI-MEDIA". The fact sheet gave a "Test Date” of
`
`10.’20f93 01:26:2'?PM. The fact sheet indicated that the sound card was an
`
`SGPRO-16, but I noticed that the Windows software called the sound card
`
`an MM 16 PRO. A video capture card that did not come with the
`
`computer had,been installed in the computer. The original box containing
`
`the video card indicated that it was “PCVDIOOO Intel Smart Video
`
`Recorder for Indeom Video“. There was a shipping date on the box that
`
`read “6:’10f94”. The specifications of the card are listed on the back of the
`
`box. They indicate that the card is a full ISA board, that it has one RCA
`
`Facebook's Exhibit No. 1017
`Page 14
`
`
`
`and one S-VI-IS video inputjacl; and that it accepts NTSC or PAL analog
`
`video-composite YKC (S-VHS). Finally, the video capture card was
`
`connected to a Mitsubishi VCR HS-U5 90 video cassette recorder machine.
`
`In particular, the VIDEO OUT (yellow) jack from the VCR was connected
`
`to the S-VHS video input jack on PCDVD1000 using an RCA video cable.
`
`The audio output jacks on the VCR were not used. Instead, a microphone
`
`was connected to the 3 .55mm miniature input jack labeled “MIC" on the
`
`sound card. The seiial port on the computer was connected to an external
`
`dial-up modem.
`
`.
`
`I executed‘ a program called csprod.exe. This program immediately
`
`brought up a Window labeled “CS Producer". I then selected “GO” from
`
`the “Session” drop-down menu.
`
`I then entered “CATHOLIC” for service.
`
`This is a current service (a place for users with shared interests to gather
`
`and chat) on CompuServe. Ihad the option of pressing “Set Nickname" to
`
`choose a nickname other than my Compus ewe user ID (which is just a
`
`number}, so I chose “Maggs.” I then choked on “Go". At this point the
`
`Producer sofiware dialed in to CompuServe using a modem, making a
`
`connection was made to the Compus erve server that hosts the
`
`“CATHOLIC; service.
`
`Once Iwas connected, a “Room Selection" Window came up. I selected
`
`“9 — Music Room”, and a “CompuServe Control — Room 9” window
`
`popped up. This window had two sections, one labeled “Image Control”
`
`and the other labeled “Audio Control."
`
`Facebook's Exhibit No. 1017
`Page 15
`
`
`
`d.
`
`included in the Image Control section was a button for “Send Image
`
`Snapsho ” There were also check boxes for setting the video quality to
`
`low, medium, high, and very high. There were more check boxes for
`
`“Auto Send Images” and grayed out (not active) “S end Closed Caption."
`
`Under “Audio Control", there were buttons for “Record”, “Play”, and
`
`“Send". There were also check boxes for setting the audio quality to low,
`
`medium, good, and radio.
`
`The "CompuServe Control - Room 9" window also provided buttons for
`
`“Chat. ..", “Users...", “Change Room. . .”, and “Seieet Handle...“
`
`Pressing Chat pops up a “CS Producer — Chat Window” box, with a
`
`scrolling dialog box and text entry field. Pressing Users pops up a “User
`
`List” window, which shows the other participants in the same room.
`
`Pressing Change Room pops up a window with a list of other rooms in the
`
`same service. Finally, pressing Select Handle pops up a window that
`
`allows a user to his or her nicltname (apparently the terms “handle” and
`
`“niclcnanie” are used interchangeably.
`
`At the same time that the “CompuServe Control ~ Room 9” Window came
`
`up, the title of the “CS Producer” window changed to “CS Producer (on
`
`line)", and the video playing from the VCR appeared in this window. The
`
`window provided three pulldown menus, “Session”, “Options”, and
`
`“Help”. Under Options the choices were “Video Format. and “Video
`
`Source..." I selected "Video Format. . .”, and this opened a “Video
`
`Fonnat" window. Here there were pulldown menus for “Video
`
`Facebook's Exhibit No. 1017
`Page 16
`
`
`
`Compression Method?’ which I set to “Intel Indeo[TIvf] R3.l Video”, and
`
`“Si.'ze:" which I set to 160x120, indicating 160x120 pixels.
`
`I then selected
`
`“Video Source. .” and a window popped up. In this window, for “Input
`
`Source” there were check boxes for “Composite” and “S-Video (YEC).
`
`I
`
`' selected composite because that is the format provided on the VCR’s
`
`VIDEO OUT jack. Under “Input Type" there were check boxes for
`
`“NTSC” vs. “PA_I..." I selected NTSC, as that is the type of the signal on
`
`the 'V'CR’s VIDEO OUT jack. Finally, I clicked a separate button labeled
`
`VCR.
`
`. At this point, on another computer, a desktop running the Windows 2000
`
`Server operating system, I started the “CompuServe Infonnation
`
`Manager” (Cllxd) software, by executing a file called W'D~ICIM.E'XE.
`
`CIM is the standard client software run by CompuServe users.
`
`W]NC]M.EX;E includes executable code called Viewer that implements
`
`the client side of the Producer I Viewer system on the participator
`
`computers that are not running the Producer software.
`
`I clicked on the
`
`green traffic light “go” button, and was asked for a room. I was asked to
`
`select a service, entered “CATHOLIC”, and was logged into CompuServe
`c
`
`via TCP/TP with a different user name.
`
`11. Once logged in, I began by pressing the “Who‘s Here” button. A ‘'Who‘ 5
`
`Here” window popped up, and I was able to list all users in the service or
`
`all users in any particular room. I observed that Maggs was in Room 9.
`
`Next I pressed the “Enter Room” button, and selected Music (9). The
`
`Facebook's Exhibit No. 1017
`Page 17
`
`
`
`rooms have both names and numbers, and room 9 is also known as the
`
`Music room. A “Music Room“ window came up, which contained a
`
`scrolling dialog chat window with a line for entering messages. When this
`
`Music Room window appeared, I saw my other user name (which I had
`
`arbitrarily chosen to be “Phoebe") appear in the list for Room 9 in the
`
`“Who’s Here" Window.
`
`On the producer side, as user “Maggs," I pressed "Users..." and a window
`
`labeled "User List (2)” popped up, showing two users in room 009.
`
`(“Maggs", running the Producer software, and “Phoebe”, running client
`
`software, ‘ClZM). As user “Maggs,” I pressed “Change Room.. .," which
`
`brought up same “Room Selection” window seen before. It listed nine
`
`diiferent rooms associated with “CATHOLIC” service. User “Maggs”,
`
`however, did not change rooms. As the user "Maggs," I then pressed
`
`“Select Handle...” and chose “Bruce" as a new handle. I noticed that the
`
`name changed on the user list. From this point forward, any chat
`
`messages sent by the producer were labeled “BrL1ce>” rather than
`
`“Maggs>.”
`
`As user “Bruce,” I then pressed “Send Image Snapshot”, and a “Bruce
`
`Image” window immediately appeared on “Phoehe’s” screen, showing
`
`snapshot of Video that was being played by the VCR. This message was
`
`sent by the Producer software to a Compuserver server running the CB
`
`Conferencing system and from there to the Viewer software on user
`
`“Phoebes”’s computer. Phoebe" then received a text message from
`
`Facebook's Exhibit No. 1017
`Page 18
`
`
`
`“Bruce” in her chat window. As user “Phoebe,” I then sent a text message
`
`to the chat room, and “Bruce" received it. As user “Phoebe," I then
`
`pressed the “Ignore" button, and an “Ignore. . window came up.
`
`I
`
`selected “Bruce” from list of Room 9 users. As user “Bruce,” I then
`
`entered a text message. This time it did not appear in “Phoebe‘s “chat
`
`window. As user “Phoebe,” I then sent a message to the chat room. It did
`
`appear in “Bruce’s" chatlbox. User “Phoebe” was-ignoring “Bruce”, but
`
`“Bruce” was not ignoring “Phoebe”.
`
`k. As user “Bruce,” I then pressed “Record” in the “CompuServe Control 4
`
`Room 9” window, and recorded a brief message. Ithen pressed “Play”
`
`and heard the recording, then pressed Pressed “Send” but did not hear it on
`
`“Phoebe’s” computer because it did not have a sound card.
`
`17.
`
`Icompiled and ran Gtalk versions 1.6.4, 1.6.6, and 1.6.8 for Unix. The Gtalk
`
`source code produces two executable programs, a server program called
`
`and a client
`
`program called “gtclient”.
`
`I ran and studied the software with both the server and client
`
`software nmnjng on a computer with the same Red Hat Linux 6.2 operating system. I connected
`
`to this computer over a public TCPKIP network using the telnet application, and I studied the
`
`behavior of each of these versions of GTALK.
`
`18.
`
`Ireviewed The Sociable Web.
`
`a.
`
`I examined the two html versions of “The Sociable We ” paper by Donath
`
`and Robertson found at
`
`hm :2’?Sm 2.media.rnit.eduJp eoplefJ udith!S ociaJWeb/Sociablew eb.h1:m.I and
`
`at
`
`Facebook's Exhibit No. 1017
`Page 19
`
`
`
`hgpzfiarchjve.ncsa.uiuc.edu!SDG!lT94fProeeedi.ngsr'CSCWz‘donath!Sociab
`
`leWeh.htm.I (Ex. 17) and the HTML source files for each page (Ex. 18)
`
`and also online files found in the web directory
`
`hr_tg:/fsmg.media.mit. ed;,1[_oeoplefjug'thfSocialWebfPix/ (printout at Ex.
`
`13).
`
`i
`
`. When I first examined the first document (Ex 1?), hosted on the server
`
`smg.media.mit.edu, the embedded graphical images in the document (gif
`
`files), such as the one specified by the link below (found in the HTML
`
`source file (Ex. 18) <img src =
`
`"h
`
`:/r"ud,ith.vNvw.rnedia.mjt.edu.I‘Socia.lWebl'PixfWho0nlineText.'
`
`'
`
`'>,
`
`the images did not appear in my browser because the gif files were not
`
`hosted on the server judith.wWw.media.mit.ed1:. The images