There is a persistent narrative in some quarters that you need to “balance” the needs of SEP holders and implementers[1] in order to encourage innovation and standards development. Their point seems to be that large numbers of innovative companies will not participate in standards development unless motivated by the ability to earn big bucks (or other currencies) from licensing their SEPs to implementers. This seems improbable: vanishingly few entities that do not plan to implement a standard have the expertise, interest and resources to participate for years (often decades) in the development of a standard that might, or might not, include their patented technology and which might, or might not, be successful enough to lead to large licensing revenue years down the line.
Compare this improbable narrative to the words of an individual (Andrew Myles) who has been participating in standards development for over twenty years:
The vast majority of companies that have their people participate in standard development make money by selling products that implement the standard. Companies are rewarded for having people participate and contribute by the ability to sell those products. For almost every company that participates and contributes, licensing is not the goal. Implementation is. Licensing does not drive people to contribute to standards at all in my experience. Interview with Andrew Myles - Part 2 (sepessentials.com)
I thought given the difference between what some commentators say about what is necessary to encourage standards development, and what those who actually participate in developing standards experience, it would be worthwhile to take a closer look at what types of companies have participated in standards development and see how those participants have changed over the years.
My next few posts will be a mix of history and data: history because context is important and data because evidence is important too. So, if you’re not a data person, focus on the history but bear with me because there’s a lot of interesting conclusions that result from the data and the history. I’m using minutes from some of the plenary sessions of the 802.11 working group because those minutes contain the names and some details about the attendees. This post focuses on the development in the late 1980s and into the 1990s.
To set the stage, the Internet as we now know it was made possible in the 1980s when the Transmission Control Program was coupled with the Internet Protocol to form the Internet protocol suite known as TCP/IP.[2] The Internet did not become fully and finally the "Internet" until ARPANET was decommissioned in 1990 and other interconnected networks were shutdown a few years later.[3]
On a personal note, when I was at graduate school at UCLA (which was one of the first ARPANET nodes so it already had connectivity when I started in 1987), I could browse FTP sites and email friends, but only those friends who were graduate students at some of the other universities that also had ARPANET sites. Not every university was connected and there were no pictures, no video, no blogs, no social media but a lot of alt. and other “sites” (which were just Listservs, not anything like a modern day website).
With a lot of forward thinking given how rudimentary the Internet was in those days, the 802.11 working group was started in around 1989 as an exploratory group. It had its first plenary meeting in 1990 when the Internet was just getting off the ground as a commercial endeavor.
In 1997, the first version of the WiFi standard, 802.11-1997, was released by the IEEE, but no one ever really implemented it. It turned out that the 1997 version of the standard was not sufficient to make products implementing that version interoperable.[4] In 1998, the 802.11 working group had gone back to the drawing board and was developing what eventually became two different versions of the standard (generally known as 802.11a and 802.11b), both of which were released the following year, in 1999. So, even in 1998, there was no functional Wi-Fi.
I started law school in 1990 after leaving graduate school. In 1998, I was a mid-level associate in a large law firm in San Francisco. My law firm used big IBM desktop computers that had a windows-based graphical user interface (“GUI”), but you still occasionally had to revert down to DOS for certain functionality. Light weight laptops were not yet ubiquitous: no one I knew had a personal laptop in 1998, although by then many of us had personal desktop computers at home. Those desktop computers generally communicated over the Internet via a wired connection to a very very slow (56kbit/s) dial-up modem.[5]
The world-wide web (“WWW”)[6] existed at that stage, but it was nothing like the current multi-media experience. We were still calling what we were using the Internet (the term the Web was only just beginning to come into common parlance) and doing large data transfer mostly via file-transfer-protocol (“FTP”) sites.[7] By 1998, people were transitioning away from browsing the Internet using Netscape’s Navigator Internet browser: Microsoft’s Internet Explorer had just recently become the most widely used browser. Yahoo was the largest Internet email service. I had a Yahoo account, but my mother (a university professor) was clinging to Qualcomm’s Eudora software for her email. The current web search platforms did not exist – Google was founded in late 1998 and Microsoft was just launching MSN Search (it had not yet been rebranded Bing). The Internet was still slow to load and hard to navigate (limited search functionality). The dot com bubble was in its infancy and the dot com bust was not on anyone’s radar yet.
Between 1990 and 1998 there were also a lot of transitions in the telecommunications and network equipment sectors. At the beginning of the decade, AT&T still had its equipment arm. Lucent Technologies, which had been AT&T’s equipment side, was not spun out of AT&T until 1996. A Canadian telecommunications company founded in 1895 began the 1990s as Northern Telecom. After a few acquisitions, including its 1998 purchase of Bay Networks, it became Nortel Networks. In 1998, the long-titled Alcatel Alsthom Compagnie Generale d’Electricite spun out its Alsthom side (which focused on railroads) and changed its name to Alcatel S.A. to focus on selling telecommunications and networking equipment. Cisco was no longer a startup having been founded more than ten years previously (in 1984), but still had only a limited product line mostly consisting of routers and switches, the backbone of the Internet.
I didn’t even know what a cell phone was in 1990. By 1998, only one person I knew (a law firm partner with whom I frequently worked) had a cell phone. But he had a fixed car mounted phone that you could not take out of the car. 2G (which did not carry data) still dominated the cellular space and Nokia, Ericsson and Motorola sold the most popular mobile phones.
Apple Computer started the decade out on a high note, selling popular personal desktop computers. But by 1997, Apple was on the brink of bankruptcy. In a last ditch effort to save itself, Apple fired its then CEO, brought Steve Jobs back as “interim” CEO and, in 1998, released its first iMAC computer. Apple's iPhone and iPad had not yet been created.
On the legal side, there were comparatively few patent lawsuits in general, and nearly all of them were between competitors. There were almost no non-practicing entity patent lawsuits throughout the 1990s.[8] The competitor suits included pitched patent battles between Qualcomm, Motorola, Ericsson and Nokia over who would control the underpinnings of 3G and what the licensing arrangements for 3G SEPs would be. Lucent had sued Cisco for patent infringement which began a series of cases between them. The Eastern District of Texas was (mostly) still just a twinkle in Texas Instruments’ eye. By 1998, it was beginning to host some cases, but all the patent cases I worked on in the 1990s were filed in the defendant’s or plaintiff’s home town or state of incorporation (so for me, almost all of my patent cases in those days were venued in California).
The competitor patent lawsuits and licensing of the 1990s also was starting to significantly change how companies grew and developed their patent portfolios and the size of those portfolios. In 1998, only a few companies had significant patent holdings, but this was rapidly changing.[9] Cisco for example started the decade with less than a handful of patents, but by the end of the decade was filing hundreds of patents a year and had acquired additional patents through acquisitions.
You could say that by the end of the 1990s, the Mondo 2000[10] era of the Internet was just ending, and the more corporate Wired[11] era had begun.
With that history in mind, what follows are the companies[12] that participated in the IEEE 802.11 Full Working Group Plenary Meeting that took place (a) in La Jolla, California on November 12-16 according to the Tentative Minutes of the IEEE P802.11 Working Group from November 12, 1990 and (b) in Irvine, California in March 9 - 11 according to the Tentative Minutes from March 9, 1998.
I’ve done my best to put each company into the category that best fits what they were doing in those years (there are explanations of my categories after the table), but sometimes the choice of category is a bit arbitrary. I have also done my best to try to group affiliated entities together (so for example, participants from both Clarion Co. Ltd. and its U.S. subsidiary, Clarion Corporation of America, are grouped under Clarion). I’ve also tried to indicate which companies were “startups” in the relevant period of time by which I generally mean a relatively recently formed company that had not launched many (or sometimes any) products at the time.
There are a few entities whose history is lost to time so if any of you know what they did in the relevant periods, please let me know and I’ll update the chart.
Of note, there are only a few companies that had representatives present at both the 1990 and 1998 plenary meeting. These are: AT&T, NEC, NCR (which for some of the 1990s was part of AT&T) and Symbol Technologies.
With that in mind, here are the attendees based on the Tentative Minutes which can both be found here: SUMMARY REPORTS & MINUTES OF 802.11 WG SESSIONS (ieee802.org). Note that a blank space means zero participants.
The categories can be described as follows:
· Component part supplier: companies that provide component parts (other than semiconductors) to equipment manufacturers. So, for example, AMP Inc. made connectors to electrically couple other component parts.
· Defense contractor: company whose primary business is making specialized products for the U.S. and other militaries. There were a lot of defense contractors in the working group in these years, although perhaps this is not surprising given that the Defense Advanced Research Projects Agency (“DARPA”), along with several universities and defense contractors, created ARPANET and many of the networking technologies which became the foundation for the Internet.
· Internet Service Provider or ISP: company whose primary business is providing access to the Internet (but which did not also provide significant cellular or wireline services).
· Networking equipment: companies whose primary business is equipment for networking.
· Semiconductor: companies whose business is solely or primarily to design and/or manufacture semiconductors. Often called “chip companies”
· NPE or “non-practicing entity”: an entity (a) whose primary business is licensing patents and (b) which never had a product business or which at one time had a product business but that product business is essentially kaput (this latter category I sometimes call “once were companies”). This one is for future reference – in 1998 there were no NPEs participating in the 802.11 development efforts.
· Organization: a non-university government or partially-government sponsored organization or non-governmental organization (“NGO”).
· Other: for companies that do not fit into any of the other categories
· Telecommunications service provider: company whose primary business is providing telecommunications services (which includes one or both of cellular and wireline telephone services, and might also include the provision of Internet access but, unlike the ISPs, is not limited to the provision on Internet access).
· University: just what it says.
· User device: a bit of a misnomer since many of the companies labeled user device provide enterprise level products, but this term is intended to denote a company that provides products that are usually used by one person at a time and often can be handheld. This category runs the gamut from phones to computers to audio equipment to RFID devices and handheld scanners. This is the broadest category and so I’ve tried to indicate the main types of devices sold by that entity in the relevant time period.
· WLAN or Wireless LAN: a specialized type of networking equipment company with a focus on equipment to wirelessly connect nodes within a local area network. Technically, 802.11 is a form of wireless LAN.
[1] This whole idea that there is a dichotomy between “implementers” and “SEP holders” just does not hold up. Although it is true that not all implementers are also SEP holders, until very recently, nearly all SEP holders were also implementers. This remains true of nearly all originating (e.g. the original owners) SEP holders, i.e. originating SEP holders are nearly uniformly also implementers.
[2] What became the Internet was, in large part, derived from the ARPANET which had been developed and mostly funded by the Defense Advanced Research Projects Agency (“DARPA”) of the U.S. government. The suite of networking protocols that helped turn the ARPANET into the Internet consisted of the transmission control protocol (TCP) which had been defined as a standard by the IETF (RFC 675 - Specification of Internet Transmission Control Program (ietf.org)), the User Datagram Protocol (which originally was part of the TCP but was later split out from it), also an IETF standard (RFC 768 - User Datagram Protocol (ietf.org)) and the Internet Protocol standard which had several experimental incarnations, Internet Experiment Note Index (rfc-editor.org), but its first real release in 1981, also as an IETF standard, RFC 791 - Internet Protocol (ietf.org). All of these networking protocols began as DARPA projects.
[4] To get technical, the 1997 version focused on standardization at the PHY layer of the Open Systems Interconnection (“OSI”) stack (the OSI model is itself an ISO standard which “provides a common basis for coordination of standards development for the purpose of systems interconnection” ISO/IEC 7498-1:1994 - Information technology — Open Systems Interconnection — Basic Reference Model: The Basic Model). By the time the 802.11-1997 version was released, it had become apparent that there needed to be standardization at the MAC layer as well as the PHY layer to ensure interoperability.
[5] Ethernet is defined by the IEEE 802.3 standard. See, IEEE 802.3 ETHERNET. Dial up modems used various ITU V. standards (the 56 kbit/s was V.90 : A digital modem and analogue modem pair for use on the Public Switched Telephone Network (PSTN) at data signalling rates of up to 56 000 bit/s downstream and up to 33 600 bit/s upstream (itu.int))
[6] The web uses the Hypertext Transfer Protocol (HTTP) which is an application layer protocol first adopted as a standard by the IETF, RFC 1945 - Hypertext Transfer Protocol -- HTTP/1.0 (ietf.org). Of course, in order to make the Internet and the Web function, numerous other standards for packetization, audio, video, etc. are needed.
[7] File Transfer Protocol is also a standard, IETF RFC 959, RFC 959 - File Transfer Protocol (ietf.org).
[8] The original “patent troll”, Jerome Lemelson was an exception to this. A Great Inventor, or a Big Fraud - Los Angeles Times (latimes.com). And there were a few other ones as well (Invention Is Often the Mother of Litigation - Los Angeles Times (latimes.com)) – but these early non-practicing entities were universally individuals who had patented their own ideas. There were not yet any large patent aggregation entities.
[9] In 2010, a young assistant professor at Santa Clara University School of Law by the name of Colleen Chien wrote a quite interesting article about the patenting arms race which discussed the patenting and assertion strategies of a number of the companies, including several that participated in the development of 802.11. From Arms Race to Marketplace: The Complex Patent Ecosystem and Its Implications for the Patent System (uclawsf.edu). Ms. Chien is currently a professor at my alma mater UC – Berkeley School of Law and has gone on to an illustrious career including stints in the Obama White House as a Senior Advisor, Intellectual Property and Innovation and more recently on the Transition Team and senior counselor to the Department of Commerce and Marian Coak distinguished scholar at the United States Patent and Trademark Office.
[10] In the 1980s and into the early 1990s, there were a lot of incredibly smart and somewhat crazy people creating and inhabiting what became the Internet and there were not a lot of other people populating it (except us graduate students and some researchers and professors - sometimes these were the crazy smart people). There were hackers and crackers, “phreakers, neo-hippies, artists, anarchists, ravers and cyberpunks.” It was still the era in which the dominant belief (at least in the Bay Area tech community) was that all things on the Internet should be free and open and unencumbered by corporate types or legal impediments: the “Information Wants to Be Free”, open standards, open networks era. See, http://www.law.upenn.edu/fac/pwagner/wagner.control.pdf and Long Live the Web: A Call for Continued Open Standards and Neutrality - Scientific American. Mondo 2000 harnessed that combination of technology and wild-west freedom. It was the penultimate Bay Area hacker/cyberpunk magazine from the early days of the public Internet, and it epitomized some of the Internet’s early ideals. See, SF Weekly: Mondo 1995 (archive.org).
[11] When Wired started out in 1993, it had some overlapping writers with Mondo 2000 and had some similar aspirations, although it was always more grounded in technology. See On Reading Issues of Wired from 1993 to 1995 | The New Yorker. But, by 1998, Wired had become focused on the business of technology, although it was still aiming for a hip audience. You can see the differences between the world of Wired and the Mondo 2000 world in the article Wired did about Mondo. Tech Time Warp of the Week: Before WIRED, There Was the Eccentric Mondo 2000 | WIRED
[12] I recognize that, at the working group level, the IEEE is an individual membership organization. Nonetheless, most of the individuals that participate in IEEE working groups are there because the company or organization they work for is interested in the development of the standard. Each person’s company “affiliation” is recorded in the minutes.
[13] NCR (originally National Cash Register) was an independent company until its acquisition by AT&T in 1991. In 1996, AT&T restructured which resulted in AT&T separately spinning out both NCR and AT&T’s networking equipment business and its Bell Labs research wing (these later two collectively became Lucent Technologies).