Chapter 4: Business Cooperation and the Business Politics of Regions in the Information Age

Chapter 4:
Business Cooperation and the Business Politics of Regions in the Information Age

In early 1992, Silicon Valley faced an economic crisis where job growth since the mid-80s had been lagging behind the national average. Fed by both defense cutbacks and a sense of foreign and domestic competition against its high-tech products, business leaders in the region created a new organization in 1992 called Joint Venture: Silicon Valley, its goal "a community-wide effort...to construct a rational blueprint for the continued economic vitality of Silicon Valley."

With the publication of a commissioned report, An Economy at Risk, along with a jammed 1000+ conference of technology leaders by the new organization in the summer of 1992, the organization became the focus for business revitalization within the region. Belatedly, government leaders were also encouraged to become involved, with San Jose mayor Susan Hammer becoming co-chair of the organization in 1993. Not only would Joint Venture work to strengthen existing high-tech organizations such as the San Jose Chamber of Commerce, the Santa Clara Valley Manufacturing Group, and the American Electronics Association, it also unleashed a host of new working groups and a range of new incorporated organizations. One of these "spin-off" organizations was Smart Valley, its vision being to "create an electronic community by developing an advanced information infrastructure and a collective ability to use it."1

Smart Valley, in turn, would become the catalyst for a host of initiatives and new organizations focused on using advanced communication technology to strengthen the regional economy. For most in Silicon Valley in the early 1990s, the Internet was not seen as a focus of commerce itself but rather as a new tool for regional revival, of the creation of electronic communication bonds that would create more efficient production within the region. In doing so, Smart Valley was following the model of Singapore, a "smart island", and the "science cities" being planned throughout Japan, themselves in turn trying to improve on the model of regional technology development in Silicon Valley itself.2

Two major initiatives of Smart Valley, CommerceNet and the Bay Area Multimedia Technology Alliance (BAMTA), were in turn incorporated as separate organizations with more yearly funding than their parent (thereby adding to the explosion of interlocking business organizations spawned by Joint Venture). CommerceNet, initially chaired by John Young of Hewlett-Packard and involving almost every major computer company in the region, was established initially to make business-to-business networking over the Internet more feasible, while BAMTA saw its mission as uniting the diffuse and often non-cooperating artistic, software and hardware talents of the Bay Area together in creating new multimedia technologies over networked technology. All of these initiatives seemed to clearly be fulfilling the original mandate of Joint Venture to strengthen Silicon Valley vis-a-vis other regions.

However, both of these spin-off initiatives rapidly went through transformations that downplayed their specifically regional nature in favor of allegiance to a national and even global perspective on the industry. CommerceNet had allowed a few non-Bay Area companies and federal agencies in its membership from early on. Such outside members proliferated so quickly that CommerceNet began talking of creating regional subgroupings around the country, which was followed by the acceptance of foreign member companies, including Japanese companies, a rather startling change from the nationalist anti-Japanese stance of the Valley in the 1980s. While the bulk of members still hailed from the Bay Area, the organization began to publicly downplay any specific regional focus. Similarly, BAMTA very quickly expanded its membership nationwide and globally and soon officially wiped the words "Bay Area" from its name, preferring "Broad Area Multimedia Technology Alliance."

What does this transformation mean? For companies who established the network and remained heavily involved, it did not mean necessarily a loss of loyalty to their home base in Silicon Valley. However, it did reflect the technological, social and economic forces buffeting the high technology economy and these institutions in turn played real roles in reinforcing those same globalizing economic forces. Further, these changes reflected the specific changes that led the Internet from being seen mostly as a high-tech tool for regional revival back in 1992 to being perceived as a global industry unto itself by 1994 and 1995. And it reflected new dynamics of high technology where global initiatives are almost impossible to separate from the regions from which those initiatives are launched. It is in the latter sense that, paradoxically, the abandonment of regional language by these Bay Area initiatives reflects the intense importance of the region, not its absence.

What is a Region in the New Economy

As a regional space, it is clear that Silicon Valley and the Bay Area are very specific products of national forces, especially the investments of the federal government in technology in the post-war period. Yet even as some of those investments are withdrawn (although not to the extent that some believe), regions like Silicon Valley continue to loom large in the geographic space of the global economy. Despite the odes to "virtual corporations" and the rhetoric that global communication technology made where one did business irrelevant, real companies in the specific regional economy of the Bay Area are playing a disproportionate role in guiding not only the new technology of the information age but in leveraging that technological dominance into global power over finance, media and even politics in the new economy.

Yet even as regions loom as large as ever, if not larger than in the past, in shaping business organization, globalization of work for poorer, less skilled workers continues to accelerate. The irony is that those families most bound by poverty and community to specific spaces are the ones most at the mercy of the ravages of globalization, while the most cosmopolitan engineers and professionals find solid community in the nexus of business and living spaces in regions like Silicon Valley. Business leaders find that far-flung global alliances are most easily built on the foundations of trust in such regional spaces, the key being that those least dependent ultimately on a specific space end up with the most power within a region. The fact remains that the economic action of technology innovation is overwhelmingly local but the power of corporations to choose their venues is global and outside the control of the local actors who desperately try to negotiate with these global partners.

As described in the first chapter, a range of theorists (Saxenian, Piore and Sabel, Best) have explained this regional agglomeration by citing the economic tradition of Alfred Marshal and the idea of industrial districts. Instead of the traditional neoclassical economic view that firms can be treated as isolated units seeking the lowest cost production area, these new views identified the milieu of production to be as important as the individual firm. Such industrial districts create a shared employment pool along with a flexible set of production partners that can be easily rearranged as needed for new innovation or changes in the market.

However, analysts like Bennett Harrison and Robert Reich emphasize that even in such industrial districts, multinational corporations are the driving force in economic decisions. Corporate webs use the flexibility in regions to maximize the effectiveness and profitability of their local offices and factories while using local knowledge learned to enhance global empires. They promote the collaboration of symbolic analysts in local regions to create the key information assets for global competition, while outsourcing low skill production jobs to the array of regions, which have low labor costs.

Yet there is a third factor shaping regional economic space that is becoming increasingly important, namely the shaping of the agreements on information standards that are increasingly crucial for economic growth. The creation and maintenance of standards is essential to capitalist accumulation, but how to achieve those standards create tensions that shape economic geography in distinct ways. At the most basic design of technology, government standards have become increasingly crucial, but the dynamics of corporate power inevitably pushes for privatization of those standards.

On one hand, the preference of individual corporations is often for proprietary standards they themselves control, since that route allows the company controlling the standards to collect monopoly rents, the primary example today being Microsoft. However, such proprietary standards narrow competition and innovation, while often leaving incompatible technologies littering the landscape. The alternative is for the government to promote open standards supported broadly by a range of private sector companies, who while foregoing monopoly rents on proprietary standards gain by a much broader expansion of the overall economic pie due to the innovation around open standards.

Creating such standards can open up whole new markets that grow more quickly and explosively if consumers of technology products have faith that they are not buying into a dead-end proprietary technology. But such open standards, which require continual interaction between all partners in a rapidly changing technology field, will fail unless the social networks among companies are strong enough to assure each one that they will get a fair piece of the overall economic returns from cooperation. Such social networks can theoretically be built between companies in widely disparate locations, but empirically their success is much greater when companies have the multiple interactions and the shared social capital derived from sharing the same regional space.

The kicker is that while open standards, such as those around the Internet, may have a strong regional character in their origin, they must also generate globalized marketplaces and globalized systems of alliances for ultimate success. This is one of the roots of the contradictions between the localized cooperation highlighted by analysts like Saxenian and the multinational corporate direction observed by Harrison and Reich. And in the absence of government involvement to continually replenish the local social capital and legitimize the standards, it is unclear whether those open standards can long hold out against the opportunism of global corporations defecting from cooperation in favor of proprietary or semi-proprietary approaches.

This chapter will use the development of the Smart Valley initiatives to explore, first, the process by which the Internet changed from being merely a tool for "smart networking" in a region to being a focus for a whole new industry, even a whole new kind of marketplace, that would radically modify the mission and goals of regional actors and regional development in the Bay Area. Not only was the Internet a new marketplace for innovative goods produced in the region for global consumption but the Internet itself began to modify past industrial practices and industrial links dependent on pre-Internet modes of communication. In this way, the projects initiated by Smart Valley would help undermine the original regionally based goals of integration envisioned in its creation. At the same time, these developments revealed the basis for regional cooperation in the new information economy: the collaboration of top-level engineers and their corporate leaders in setting standards that lock-in economic dominance for the region.

The next part of the chapter will step back to put these developments in the broader context of the politics of business collaboration, both across the nation and historically in California. Even as specific production relationships globalize at an ever more rapid pace, there has been a countervailing trend of ever increasing business consortia relationships within regions, most often with government assistance to help stabilize the relationships between businesses. Precisely because of that globalization and the increasing disconnect between firm self-interest and the need to create the "public goods", both technical as in the case of basic research or social as in the case of standards, multi-firm business collaboration has become crucial within the new high technology economy. And beyond specific public goods, the emergence of the Internet as a radically new marketplace has required (and strained) new forms of business cooperation and trust within regions to assure its proper functioning and stability. The chapter will examine the crucial role of standards in the economy and discuss whether the open collaborative standards around the Internet developed in the Bay Area region can stand up in the battle with Microsoft. In regards to the Bay Area itself, collaboration around the Internet has begun to highlight how this new technology is allowing the region to leverage global power in a whole range of economic spheres from finance to politics.

Lastly, this chapter will outline how this new global form of cooperation at the regional level leaves out many communities in the region. In an increasing "politics of abandonment", the focus on the regional needs of high technology increasingly focuses the provision of public goods narrowly for the elite while leaving public squalor for those communities on the periphery. This reflects the reality that the cross-class collaboration of industrial production that had stabilized regions in the past has given way to the new politics of business where collaboration between engineers and corporate leaders is the new basis for regional development.

The Politics of Regional Revival: Conceiving a "Smart Valley"

When Joint Venture was launched in 1992, the high-tech executives involved saw the Internet as merely one of a range of networking tools available to spark a regional revival. Initially, the program was focusing as much on private Local Area Networks (LANs) and other proprietary networks as on use of the Internet in order to link businesses, schools, governments and health care facilities in the region. With plans for dedicated fiber optic lines to run this local technology, the focus was in many ways on uses of the technology to "lock-in" regional relationships, much as dedicated technology had locked in relationships between many individual companies.3 The main vision of electronic networking was, in the original view of Joint Venture in launching the Smart Valley project, that "[The] foundation of communications can enable innovations in regional services that will assist Silicon Valley to improve its appeal and capacity to retain and grow enterprise. The Valley has the ability to become a leader in telecommunicating and in high speed, high bandwidth networking of companies, suppliers, and individuals."4

In this, Smart Valley was like a number of regional information networking projects around the country. In Arizona, a collaborative project centered around Arizona State University, Digital Corporation and the Times Mirror corporation sought to reinforce the same kind of regional collaboration of local industry using Times Mirror cable technology along with Digital's Ethernet networking technology. "We've tried to work with some major manufacturing companies and their suppliers to begin to look at what utility these companies can gain by being connected across a metropolitan area network," said Dave Rosi, Digital's director of video and interactive services, back in 1994. "If provided with the right tools, businesses will look to this type of network, embrace it, and pay for it." Defense contractors like McDonnell Douglas were the first customers in the region who saw an advantage at that time in encouraging smaller contractors to conform to a single regional standard.5 Obviously, both Digital and Times Mirror cable saw advantages in supporting making that standard tied to their own technology.

The Route 128 region followed up on Silicon Valley's initiative with an on-line network called the Commonwealth Exchange (originally called CommerceNet East until Silicon Valley folks objected) which was explicitly designed to avoid having Silicon Valley get a jump on Boston's high-tech corridor. Funded jointly by the state of Massachusetts and the Massachusetts Telecommunications Council (MTC), an organization of technology companies in the Boston area, the Commonwealth Exchange was envisioned as a system to recruit employees, place orders, and collaborate regionally. One initial focus was using the Boston-based Lotus Corporation's Notes' 'groupware' software, popular on proprietary networks for collaboration, to connect government agencies, schools and businesses on new projects.6 Like the initial conceptions of Smart Valley, use of the Internet was merely an option among more proprietary possibilities that would reinforce regional standards (and not coincidentally assist specific regional companies promoting those technologies).

Other initiatives in this period of the early 90s focused less on specific regions than on whole industries in order to make US production more efficient in international competition. The American Textile Partnership (AMTEX) was created in 1994. AMTEX was organized to link different parts of the textile industry, from fiber to retail, with 11 Department of Energy laboratories in developing technologies to recapture global market share -- through better manufacturing processes, electronic commerce and quick response to consumer demand. About 80 industry partners are now involved and funding for 1995 was pushed to $50 million annually, provided jointly by industry and government. Almost half of that funding was devoted to electronically networking the industry.7

It was in these industry-specific networks that the emergence of Internet standards as a business standard became more and more obvious. The auto industry had been an early user of electronic networking with its suppliers using proprietary networks. Seeking to escape the costs and limits of such networks (often called Value-Added-Networks or VANs in industry parlance), Ford Motor company announced in March 1995 that they would begin using Internet-based IP standards over a secure private version of the Internet managed by Ameritech Corporation to exchange automotive design files with contractors. This followed the determination in 1994 by the Automotive Industry Action Group, the technical advisory organization for the automobile industry, to adopt IP as the communications protocol for communicating with parts suppliers. With parts suppliers often working with multiple companies all over the country, using a standard set of protocols became more and more important. Ford soon was working with General Motors and Chrysler to develop what was soon called the Automotive Network Exchange (ANX) to allow contractors to use Internet connections to link to the private network with encryption and security for sensitive data being exchanged.8

So it was in the context of competing regional networking initiatives, often using proprietary or company-specific technology, and national industrial networking initiatives turning more and more to Internet standards that Smart Valley and its spin-off projects were developed. Partly, the Smart Valley initiative sought to ride the new federal interest in information networking that emerged out of the 1992 Presidential campaign and Bill Clinton's and Al Gore's call for creation of a national "Information Superhighway."

Coming out of Joint Venture (itself partly funded by federal and local government), the Smart Valley initiative was funded primarily with private corporate contributions, but it used its position to help leverage millions of dollars in federal funds for its other incorporated initiatives. Michael McRay, the Program Manager at Smart Valley who worked extensively on helping to develop the CommerceNet initiative, described Smart Valley's role as, "We are venture capitalists without capital. We help consortia find resources." After that point, Smart Valley stepped back to offer strategic advice, but each initiative would take on a life of its own.9 In the case of CommerceNet and BAMTA, that life would end up being very different from its sponsors' initial conception.

CommerceNet emerged out of the initial Smart Valley idea of creating local networks between firms to enhance production in the region. Backed by Silicon Valley firms, Stanford's Center for Information Technology and the Bay Area Regional Research Network (the NSF-sponsored regional Internet coordinator that was just about to be sold off to the BBN), Smart Valley solicited an $6 million federal grant in late 1993 from the new Clinton administration Technology Reinvestment Program.10 This would be supplement with $6 million in corporate and local government support over the initial three years. Sponsoring corporations would pay anywhere from $5000 to $35,000 per year to join CommerceNet, with many federal and local government agencies paying those fees as well.11 Through this public-private partnership, the federal government hoped to see commerce on the Internet grow while the regional companies hoped to foster more regional growth.

Most of the staff involved in Smart Valley held to the view that the region would trump the global effects of the Internet. Michael McRay at Smart Valley sees the cooperation around CommerceNet as part of a long tradition of rivalry and market competition in the area. "Silicon Valley has a wonderful history of competition and collaboration." McRay explicitly cites Saxenian as a guide in Smart Valley's thinking about regional economics: "There's definitely a regional advantage. Geographically defined clustered industries have a real need to keep on their toes. They are blessed in having a cluster in the electronics industry. I don't see going on-line losing that regional advantage. It goes to that issue of trust. These very close business relationships are fostered by having a geographic proximity. In Silicon Valley, people are loyal to technologies, not firms. This isn't as much stigma. So there is much more cross-fertilization and crossover. The Internet has played a key role in this cross-fertilization. The proof is that both Silicon Valley and California have a high connection to the Net."

McRay sums up the effects of the Internet this way: "While the Internet allows you to find out about things around the world, sometimes the most important function is to find out what is happening in your own backyard." In this view, the social networks are just as important in determining the future of a region as their communications network. And if the Internet strengthens the social networks, it can only strengthen regional advantages. McRay notes that CommerceNet will allow the further disintegration of companies to subcontract functions that are not within their "core competencies" but he expects that this will happen more regionally than internationally.12

Randy Whiting became the first chair of CommerceNet's Sponsored Projects committee, the day-to-day governance body for the varied projects initiated at CommerceNet. Whiting, a Hewlett-Packard manager at the time who would later become CEO of CommerceNet, described the initial mandate of CommerceNet as looking "to automate the value chain and supply chain of electronics in the Silicon Valley. At the high end, we (at Hewlett Packard) are more of a user than producer of electronics, so the idea was to bring together the main high-end consumers, and then we could deal with the whole wealth of contractors "downstream"--fab, consultants, etc. If we could increase the efficiency of the supply chain, prices would be lower and the companies here would be more competitive. We would be using the technology to improve the efficiency of that region. In a business-to-business relationship, the Internet was a good solution because there isn't a generalized infrastructure [of specific networking technology] in place. We had a regional imperative."13

BAMTA had a similar initial focus with the addition of tying in many of the graphics and software companies often located in San Francisco's "multimedia gulch" that had often been more peripheral to day-to-day business in Silicon Valley. Additionally, BAMTA had the extra impetus of NASA's Ames Research Center pushing to commercialize multimedia technology in order to justify its continued existence during a period when NASA was looking to cut costs. BAMTA was formed largely because in late 1994, Dan Golden, the chief administrator of NASA, approached the leaders of Smart Valley with the idea of establishing an alliance with local industry to develop multimedia technology over computer networks. No group was bringing together the disparate parts of the multimedia industry in the region, so they decided to create BAMTA to deal with the cradle-to-grave production of networked multimedia products, from the creation of content to the tools for managing that content, to delivery to the consumer. Bringing these different parts together would be a challenge, since, in the words of Karen Greenwood, a Smart Valley Project Coordinator who helped launch BAMTA, "People in content creation areas are not used to working with people at UB Networks. You have the longhaired people in their Tevas not talking with the people in their suits and ties. Until you had the bandwidth and the machines to have networked multimedia, there wasn't much reason to talk. Say you're Broderbund, you're creating a children's book software item, you're not thinking about the company that created the machines and since you didn't have the networks, you didn't think about the intersection."14 NASA agreed to fund the project with an initial grant of $5 million over two years with a commitment that the money would be matched five to one with private funds and in-kind donations. BAMTA quickly grew to forty-three companies focusing on a range of areas from the delivery of health care information over networks to creation of a "Collaboratory" to test multimedia products in combination in a facility with state-of-the-art equipment, a special boon to smaller companies. NASA specifically supported three large projects aimed at air traffic control, reconstructive surgery assistance over networks, and networking high technology laboratories.

NASA framed its involvement squarely in economic terms, both for its own bureaucratic self-interest and in public interest goals. NASA wanted to make Ames a model for how NASA could partner with local industry in order to commercialize its technology. "The fundamental reason is that if you commercialize technology," argued Paul Kutler, the NASA director assigned to oversee the agency's involvement in BAMTA, "you can create economic growth and new jobs. The new mandate for NASA and other agencies is to commercialize their technology. Some have been in classified areas with state-of-the-art technology which were not in the public domain and now can be declassified and used to spur the economy, especially in heavy Department of Defense areas of the country." As to why NASA decided to invest so heavily in this kind of venture when they did, Kutler stated, "Fundamentally, it gets down to that there is world economic competition. In aerospace, it used to be McDonnell Douglas versus Boeing versus Lockheed versus Martin versus Hughes, and now it is Boeing and McDonnell Douglas versus Europe." In stark terms, NASA and other agencies see sparking regional economic growth in new technologies as a nationalist economic strategy against competition from other country's leading industries.15

As federal investment in both CommerceNet and BAMTA show, high technology centers still can attract more resources to reinforce their lead. And even if they were to lose the investment advantage in tangible dollars, many involved in Smart Valley see its initiatives as helping consolidate local advantage by reinforcing the intangible political and social connections regionally that help the Bay Area thrive. Bill Davidow, an old Intel executive who has become a prominent venture capitalist and was an early financial supporter of Smart Valley, sees building "the intangible asset base in Silicon Valley" as the greatest accomplishment of the initiatives. Whether through ongoing projects are through Smart Valley lectures, he sees new ideas exchanged locally that the Internet cannot quite duplicate:

The Net can facilitate a lot of things, but there's something that happens in face-to-face interchange that isn't going to happen over the network; I may be wrong in that since I was in a generation that grew up without the network, but I suspect that over hundreds of years, mankind has evolved skills that are oriented to physical interaction that won't exist for the network. Maybe in the next million years, we'll have it down pat.16

What all these comments show is that the originators of the Smart Valley initiatives assumed that electronic networking and the initiatives spawned by Smart Valley would facilitate local interaction to increase productivity and increase production links within the region. However, while some of that no doubt occurred, what would become clear is that the globalizing effects of the Internet would be more dominant and the role of regional interaction would end up being much different than envisioned by Smart Valley's creators.

CREATING A GLOBAL ELECTRONIC MARKETPLACE

It was partly a testament to the rapid changes in the Internet, partly a testament to the recovery of the Bay Area economy (and the two were not unrelated), but the Smart Valley initiatives rapidly abandoned their focus on using networking technology to enhance production within the region and instead began to focus on using the Internet technology for marketing to customers, especially for marketing Internet products themselves. With a global market to conquer, most Bay Area firms saw an increasing need for global alliances, and CommerceNet and BAMTA became prime vehicles for facilitating the alliances that would stabilize the Internet marketplace and create a bigger pie for all involved in the exploding networking industries. CommerceNet and BAMTA would within a year of their establishment officially abandon their focus on Bay Area firms and local collaboration, welcoming membership from companies all over the country and the world.

Many of the company executives recruited into Smart Valley projects ended up being quite skeptical of claims of regional production links. Mack Hicks, a Bank of America Vice President and the second chair of CommerceNet's Sponsored Projects board, leaned heavily to the view that the Internet spells the end of the limitations of geography. "We have to watch out here, since we think we have Silicon Valley-but there is no region per se." Hicks sees the whole CommerceNet project less as a regional industrial policy than as a font of experimentation whose effects will benefit all business.

To quote Mao, let a thousand flowers bloom. Some people will use [electronic commerce] to contact others in the area. Others will use it to take from the area to go out into the world. What it will do is break down the whole of the business... Any thought of boundaries--state, religious, age, sex--all go away. It is the ultimate promise of marketplace and personal diversity...In an industrial society, cities were appropriate because of the regional nature of business and the delivery of services, in the information age there is no boundary for the delivery of services so there is no reason for regions. You want to compete in the international region."17

Hicks noted that Bank of America's chief regional rival, Wells Fargo, had been using the Internet to transform itself from being a large regional bank into a major international player. With a good Web site and new strategies and services, Wells Fargo could have an international presence with no costs for new branches in London and New York. So even as local banks like Wells Fargo have merged and downsized employment (partly with the help of information technology), that same technology is helping them find new markets worldwide to substitute for less profitable customers previously served regionally. (Note that the next chapter will deal carefully with the effects of the new networking technology on banks and other non-computer companies like utilities with a traditional regional focus).

For most of the companies involved in CommerceNet, reaching customers in the global marketplace became a much higher priority than working more closely with suppliers in the region. Randy Whiting, Hicks predecessor as sponsored projects chair and later executive director of the organization, argues that where CommerceNet started out with the mission to be a "regional marketplace" to make local companies more competitive, the companies involved quickly decided that "the real value was not that I could deal with my suppliers more effectively but how could I reach my customers more effectively." With the emphasis on marketing, the focus went much more to relationships outside Silicon Valley, helping to push forward membership in the consortium for companies outside the region and internationally. 18

BAMTA went through a similar transformation, although with a later start than CommerceNet, its transformation to a global consortium was even quicker. Mason Myers, the Smart Valley staff member who helped get BAMTA up and running, argues that it became obvious quite soon that the Bay Area focus was more important to NASA than most of the businesses involved: "NASA is government and they feel bad about Moffett (the air base that closed down in the early 90s) and they worried that Ames might be closed down...so the more that was happening in the Bay Area, the more likely NASA could save Ames." Bay Area businesses didn't want the words "Bay Area" in the name and soon had it dropped, leaving only the acronym. Not that those business leaders had no vision of Silicon Valley's role in the multimedia future, but they argued that a "Bay Area-centric" approach, as Myers derisively calls it, would miss the global business opportunities in multimedia networking. Still, as Myers sees it, in the end, "The Bay Area will do very well: you have everything you need, you have the technical folks, the creative folks and the ad folks. The only other area that has it is New York, which doesn't have the technical folks, but that will be less important in the future." 19 But doing well means hooking up with multinationals like Kodak and others who are key to the multimedia future.

Whiting is a bit sardonic in seeing the whole Smart Valley effort as a case where "It started with the academics at Smart Valley, EIT [a consulting firm that helped get CommerceNet off the ground], and Stanford who didn't come from the business side...We came in and said screw it, let's look at the selling side." This led to big banks and multinationals joining in the rest of the effort rather than many of the small contractors originally envisioned. In many ways, the whole Joint Venture/Smart Valley project was set-up according to all the academic models (including explicit reference to Saxenian) of encouraging more "industrial district" style collaborations, but in the end the business folks in practice saw less need to strengthen local production and more opportunity in exploiting local advantages in skills and technology to sell to the global market. In practice, the reality was much more in line with Bennett Harrison's model of the withering of industrial districts under pressure from global competitors skimming the skills and advantages of local regions.

The very expansion of the Internet helped fuel this focus on global sales since Silicon Valley suddenly found itself at the center of the most hyped technology since the personal computer. Job growth around computer networking and related software exploded in the region with employment at the top five computer networking firms in Silicon Valley growing nearly 300% between 1991 and 1996.20 Selling the idea of the Internet to a global market became a primary goal of those involved in the Smart Valley projects. Networking projects within the region came to be seen as less about improving productivity than in showcasing the technology for the global media. Networking firm 3Com's CEO Eric Benhamou became a key director on Smart Valley's board and became heavily involved in efforts to wire the public schools for the Internet. As Smart Valley staffer Kathy Blankenship noted, "People look at Eric Benhamou who is a poster child for civic involvement, but when you look at 3Com and its work in remote networking and the projects he is involved with, Smart Valley has helped him to be seen as Mister Networking."21 Benhamou himself sees the regional networking as a launchpad for his company's global expansion. While he expects 3Com's employment to expand in the region over the years, he expects it to be a diminishing percentage of the company's payroll as 3Com moves into new world markets.22

Beyond the hardware of networking, Silicon Valley saw an explosion of new software firms linked to the Internet, with Netscape leading the pack of new firms. Throughout the early 90s, even as defense and most hardware-based firms were losing jobs in the steep California recession, software continued to expand rapidly. Semiconductor and other hardware employment staged a recovery by 1994, but software job growth significantly outpaced new jobs in manufacturing in the mid-90s as new Internet software firms exploded in the region and older software firms like Oracle expanded with new Internet-related strategies.23 The culture of innovation mattered for these new software successes but direct production links as originally envisioned by Smart Valley mattered much less than expanding the global appetite for the region's new Internet software products.

INTERNET ITSELF UNDERMINING LOCAL SUPPLY NETWORKS

Beyond the disinterest in using the Internet for regional supplier networks, the specific projects that CommerceNet and BAMTA emphasized actively undermined much of the old basis for the need for closer supplier-customer relationships. Partly, this reflected the abandonment of private electronic communication networks which themselves had served in the 80s and early 90s to lock-in relationships because of the costs of infrastructure investment. With networks based on public IP protocols, networking costs were significantly lowered and this allowed looser, fluid and more market-driven relationships with vendors. This in turn has allowed, with the strong support of CommerceNet as an organization, the emergence of an electronic marketplace for subcontracted goods to replace buyer-supplier relationships that had existed in an environment of high investment needs to establish contact and gain information on alternatives.

At the most prosaic level, the Internet has become a cheaper alternative for corporate networks. Private VANs (Value Added Networks) have been based on metering prices per kilobyte of data sent. Companies found that the $25,000 to $30,000 per year it might cost for an Internet connection was less than using a VAN for just ten hours of typical data use. More crucially, any private VAN connection limited a company's connections to other businesses and virtually excluded connections to individuals. Added costs for private high-speed "toll road" Internet access might push up costs, but using the public network still ended up being one-hundredth of the cost of older proprietary networks. 24

Software costs for maintaining internal "Intranets" or business-to-business "Extranets" are also less than older proprietary networking systems because of the shared standards and flexibility of using Internet compatible servers and support software. This led to explosive growth in Internet-based software; by the end of 1996, sixty-four percent of the top 1,000 companies had Intranets, a 30% increase from just six months earlier.25 With standardization around IP protocols, the costs of making new supplier and buyer connections tied to electronic communication significantly dropped, thereby lessening the "lock-in" of relationships around proprietary software.

What is remarkable is how quickly the US network infrastructure had abandoned the proprietary network choices of the late 80s and early 90s which, in the words of researcher Francois Bar in a 1990 study, had created "the fragmented US infrastructure [which] risks inhibiting flexibility in productive arrangements among firms."26 As noted earlier in the chapter, many larger companies like the auto industry were quite willing to forego any monetary advantages of proprietary networks for the flexibility of public standards. This opened up access to communication networks to a much wider range of companies than could have afforded to lock themselves into proprietary network systems. Similarly, large Silicon Valley firms like Hewlett-Packard had invested large amounts of resources in the 1980s in a private electronic network for suppliers. While it prized its private network for its security, it began to push for a more public system to more easily deal with its ever-changing production needs. And as that public Internet system took form, as HP's Randy Whiting notes, the company found that instead of merely creating smoother supplier-buyer relationships, the Internet was opening up the possibility of marketizing many transactions. This opened up new vistas for direct electronic purchases from suppliers and new opportunities for electronic sales to HP's own customers .

Information technology has increasingly been used to coordinate global production networks, hold inventories down and promote just-in-time delivery. The continued growth of FedEx in the 1990s was largely a product of this globalized information infrastructure where it used online tracking to increase customers' confidence in ordering parts from far-flung areas; by mid-decade it was calling itself less a transportation company than an information-technology company. In one example, Silicon Valley's National Semiconductor Corp. was using FedEx to slim its global warehousing system and to cut its total logistics costs from 3% to 1.9% between 1993 and 1996.27

On the selling side, one key project at CommerceNet became the "Cataloging Working Group" which focused on how to best put company catalogs on the Internet. If specifications for specialty products could be precisely and flexibly presented to customers, relations between buyers and sellers would need less time-consuming human massaging to make deals. Instead, more subcontracting relationships could be "marketized" rather than depending on a tight supplier-relationships to deliver precise goods. CommerceNet organized a large group of company executives focused on this area, since the pay-off would be especially dramatic in high-technology area where electronically cataloguing the tight specifications needed for products could expand from whom a company buys. The effect would be a dramatic decrease in the need for companies to be located in the same region if a company could electronically scan the product lines of parts vendors worldwide and quickly narrow down which ones met the company's needs. Decreases in the transaction costs of finding such information could allow the substitution of more market relations for structured relationships with suppliers.

The key for the working group was to create catalog systems that presented information with complex specifications in an easy to understand way. Instead of just seeing a long list of possible parts with each set of specifications, buyers could interactively test for the availability of combinations of specifications. As a customer tests each specification, they would automatically see a narrowing of subsequent choices and options that had been discounted based on previous choices. Mark Masotto, the CommerceNet program manager who worked extensively on organizing the Cataloguing Working group offers the analogy to bed buying: "Say you're buying a bed, you can get a king, queen, twin, etc. You can get different colors, different bed posts--so you want to buy a bed, you have these options of bedsprings, you have options of number of bedposts, you have a choice of sizes. So you click on a no box springs option, and you then see that a no box springs model comes in only certain colors and you see the choices narrowed to fit the first choice." With precision high technology specifications, the choices and evaluation of lost options is much more technical, but the idea remains the same.28

By allowing much quicker searches through a variety of parts makers, such interactive catalogs will decrease the advantages of long-term relationships with subcontractors and also decrease the need for person-to-person meetings to work out detailed specifications. From a salesperson perspective, Randy Whiting sees the Web interface as operating remarkably like how transactions have traditionally occurred through the intermediary of sales people. What sales people traditionally do goes beyond merely presenting products. They present the company itself in a perspective that makes sense to the buyer. That is the basis for long-term relationships for which the Web is beginning to electronically substitute itself. Explains Whiting:

In business relationships, very rarely does a customer align perfectly with a vendor suppliers' organizational structure. They may align well in many factors, but there may only be one factor how a customer wants to look at me. You do it for R&D, product line, economies of scale, etc. A customer looks at a supplier only based on products and customer needs. Those typically don't match up. Sometimes they do, but in large structured corporations, that's not usually true. What you have is sales representatives who are good at understanding customers needs and readjusting the presentation of the corporations for the customer needs. I give you a picture of the Hewlett Packard structure that more aligns with your needs. I won't tell you about the medical lines if that is not your needs, since that may confuse you and may incite questions I can't answer. The gist is that technology on the Internet is highly conducive to maintaining those kinds of business relationships. It's the first time we have technology that supports such a buyer approach.

The Internet is a non-linear methodology of presenting content and navigating through a buying process. And most buying processes are non-linear. The idea that if I present a product, very seldom can I present you with a book that says if you read from a to z in this part of the book, you'll buy my product. Instead, you'll discover things about my company serendipitously, say 'great, you do color at five pages per minute.' Never before have we had a generalized technology with a ubiquitous infrastructure that matches that relationship between a customer and a sales representative.29

Even for more routine purchases, the advantages of buying locally diminish with many of the advantages of the Web interface and smart catalogs. Bruce Lowenthal, a vice-chair of CommerceNet and Tandem Computers manager for Electronic Commerce, notes that traditional local warehouses often have limited availability and little hard information on the differences between parts. "I go to Fry's Electronics," notes Lowenthal," and take what they have. I can't get a good understanding of the quality between their parts. Now, those purchases can be more and more through the networks instead of just taking off the shelf because they're there and I have no choice." Lowenthal envisions a growing electronic ratings service need, a "Siskle and Ebert function" that will provide companies with simple comparisons between parts purchased from the whole range of global vendors. Given the ease of purchasing manufactured parts over the Net, the need for local supply sources is disappearing and Lowenthal states flatly that "Manufacturing is leaving" the region.30

The result of this disdain for local supply networks was that the first major appearance of electronic cataloguing appeared not in Silicon Valley but most prominently in places like Harrisburg, Pennsylvania where AMP Inc., a leading supplier of electrical and electronic connectors and interconnection systems, launched an electronic catalog system called AMP Connect in January 1996. AMP had traditionally spent $8 million to $10 million per year on over 400 separate paper catalogs of its products, most of which quickly became out-of-date and had to be redone every 24 months. With its new Web service, it quickly signed up over 30,000 users, two-thirds of which were engineers, from over 100 countries. Other companies quickly followed with similar products.

However, Silicon Valley's CommerceNet partners were key to these new cataloguing projects. They had a prime role in AMP Connect's and other electronic catalogs' success. AMP's catalog system was built on a Sun Microsystems SPARC 1000 server using Oracle's Oracle 7 databases and using specialized catalog software developed by Sunnyvale-based Saqqara Systems Inc., another member of CommerceNet. AMP's director of the company's on-line division, Jim Kessler, was just following the experience of CommerceNet partners when he promoted the company's electronic catalog by arguing, "To serve the business-to-business market, companies need to create database-driven dynamic content that can be searched by requirements."31 By supporting the expansion of Internet commerce, Silicon Valley firms rode an exploding market for their services, even as traditional local production is undermined by the technology they promoted. A company like General Electric was doing over $1 billion of business by 1997 over the Web with 1400 of its suppliers, an amount more than all consumer-based Web purchases in the same year. With estimates of $160 billion in total supplier-contractor business over the Net by the year 2000, there was a new market for software exploding like a bomb.32 An expanding pie for business Internet-products meant that any diminishment in local production-based commerce in Silicon Valley was being far overshadowed by the new global Internet market's contributions to local employment in the region.

Standards and the Post-Industrial Politics of Regions

The example of electronic catalogs is just one part of the explosion of electronic commerce promoted out of the furious networking surrounding Smart Valley, CommerceNet, BAMTA and all their associated projects. What tied them all together was the creation of the new marketplace of cyberspace and the support for a variety of public goods necessary to create the trust in that new area of commerce: from the development of successful technologies to facilitate commerce such as the electronic catalogs to the technology standards that create an open, innovative environment for new companies to enter the electronic marketplace.

The erosion of local production relationships at the exact same time as the appearance of new business associations like CommerceNet is not a contradiction, but actually highlights key features of the new post-industrial politics of regions in the information age. It is precisely because natural production relationships do not naturally produce open standards and public goods in a globalizing economy that new political structures between businesses are necessary. While such political structures do not necessarily have to be based in physical regions, they are facilitated by having a core of business leaders who share such regional interests and have the backing of governments who operate politically within fixed geographic space.

Business politics is much like ordinary politics--it is a way to pool resources for endeavors that would be too expensive for individual businesses to attempt or would likely generate too little monetary returns if any individual business had to do its own investments. What CommerceNet delivers to its members is a testbed for the new electronic marketplace where, as Mack Hicks argued, "If every firm was to figure out how to provide service...by themselves it would cost a lot of money and they'd make the same mistakes over and over." By working together, companies can experiment to find out what tools work, what practices work, and what standards will really facilitate a strong marketplace. The inclusion of national and multinational firms facilitates creating practices that are applicable globally (which is important if companies eventually want to sell to that global market). At its core, though, CommerceNet was built on the relationships fostered in the environment of the Bay Area and backed by government participation that helped legitimize CommerceNet's results and practices.33

The reason that politics and regions matter is that market forces tend toward proprietary systems. "A lot of these technologies depend on widespread acceptance of standards and open non-proprietary systems," noted Smart Valley's Michael McRay. "Left to themselves, an IBM and others will offer a closed system. These consortia will offer open systems. In a nutshell, they provide a system of collaboration for competing firms to work to their mutual advantage. Without this, there will be a greater tendency towards proprietary systems...We've hit it on the head with the importance of standards for the industry and the Valley. On one hand you fight with your competitor for market share; on the other hand, you work with your competitor to expand your market." McRay cited the example of the budding conflict around security for financial transactions over the Internet that evolved between Netscape and EIT. When it looked like two different systems of security were evolving that would have required two different kinds of browsers, the CommerceNet consortium helped them hammer out a joint deal to create a common standard.34

Similarly, as the basic language of the World Wide Web, the HTML markup language, began to push the limits of its capabilities, competition over its successor, called XML, threatened to create new proprietary divisions as companies like Netscape and Microsoft sought to control their implementation. A strong open XML language was crucial for strengthening the ability of business groups - whether stockbrokers, travel agents or pornographers - to create agreed upon codes embedded in Web pages to ease buying, selling and communicating over the Net. CommerceNet developed an XML-based electronic commerce standard, called eCo, to help prevent the splintering of the Web into mutually incompatible commerce systems and codes. The federal government stepped in with a $5 million grant from its Advanced Technology Program to encourage CommerceNet to form a consortium with a number of other businesses to promote its standard.35

For companies outside the Bay Area, Smart Valley's projects have offered a structured way to create new strategic relationships and participate in the standards creation process. In the BAMTA project, Eastman Kodak invested in a shared technology-testing site (called the Collaboratory) where combinations of technology can be tested for interoperability and performance. For smaller companies, a public space like BAMTA gives start-ups a chance to, in the words of Smart Valley's Karen Greenwood, "partner with these big companies without giving their companies away."36

With manufacturing and production relationships increasingly going international and multinationals joining in regional relationships in organizations like CommerceNet and BAMTA, Bennett Harrison is correct in describing industrial districts like Silicon Valley as part of "a global system populated by big companies perpetually on the prowl for new profitable opportunities [where] the very success of a district can itself bring about changes that give rise to its opposite" in global hierarchies.37

However, at the same time, the region itself becomes an arena for negotiations on public goods like standards that in turn create restraints on proprietary competition between multinationals. Even as multinationals seek to harvest the benefits of regional collaboration, their investment in that collaboration in turn shapes a new politics between business that constrains some of their opportunistic activity.

It is this tension that makes consortium like CommerceNet more and more ubiquitous in the new information economy. Why regions matter for these standards and other public investment have to do with the nature of innovation; electronic communication facilitate greater and greater dispersion of production from corporate headquarters with a whole variety of strategic production relationships, but higher-level innovation depends on physical proximity. Not only does this reinforce the traditional concentration of engineers and innovators in places like Silicon Valley, it makes such locations a natural point for negotiations over standards and the development of shared practices to facilitate new marketplaces.

Tandem's Bruce Lowenthal sees little future for manufacturing in the region, but argues, "The fact is that people have and will work together better if they do personal interactions. So I work a lot better in Tandem with people who are physically close to where I am. This is true of all companies. So if I go out to dinner with people, I learn a lot more about them and work better with them than if I just talk with them over the telephone. I don't think that will be supplanted by electronic interaction...I've seen a lot of people move away and try to keep up with hat's happening in Silicon Valley. But their productivity declines after half-a-year. They try but they usually fail. It's an observation. When you're looking at science, the more interactions the better, engineering is next, down the food chains."38 If anything, the electronic mode of communication supplements and enhances other forms of local interaction by allowing those casual contacts to be followed up with more methodically through e-mail which can be dealt with at a more leisurely pace.

For collaboration between companies, the fact that innovators and decision-makers interact with one another makes the region stronger as a hub for standards agreements. Regional organizations help to facilitate those human interactions and channel them. BAMTA's importance is less about hooking up companies than about bringing minds together in developing the framework for networked multimedia. "I'm sure that jobs in the industry will be more evenly spread across the industry over time," notes BAMTA's Myers, "But there also has to be a center, and because of the number of people in the Bay Area thinking about this, it attracts more people. So I think the Bay Area will continue to lead. It'll come from the people in the region more than the companies."39

Whether this will change is an open question with the new technology. While NASA made its initial BAMTA investments in the Bay Area because it was where the technology people were and where Ames was located, there was also interest in the Bay Area economy in general, at least partly because of the economic downturn in the early 90s that hit California so hard. At least part of NASA's interest in multimedia networking is to facilitate technology-based economic development around the country, especially in areas without the already existing technology infrastructure of the Bay Area. NASA's Kutler cites the example of a Fairmont, West Virginia laboratory that was created because Senator Robert Byrd, then head of the Appropriations Committee, pushed for its creation to assist West Virginia's economic growth. Such facilities are hard to manage remotely from NASA headquarters but with networked multimedia and the creation of what BAMTA is calling a Virtual Research Center, it would be much easier. "With this sort of technology," noted Kutler, "you wouldn't have to locate facilities in high tech areas. This could all lessen the need for regional location of technology centers together."40 However, while this may facilitate the expansion of high-tech spending in new areas without previous government or corporate technology investment, it is clear that the critical mass of innovators in Silicon Valley will continue to reinforce its regional dominance. The creation of consortia like Smart Valley, CommerceNet and BAMTA are instruments to regularize interactions that take advantage of that proximity.

The Role of Consortia in the New Economy

Consortia are relatively new kinds of organizations in the US and their existence seems to have grown with the rise of the information economy. To understand the role of CommerceNet et al, it is worth considering their predecessors in the electronics field. While Japan has officially encouraged the creation of consortia since 1960, most business consortia in the US have appeared in the 1980s and 1990s. An overwhelmingly large number of these have been in the electronics industry, including the Microelectronics Center of North Carolina founded in 1980 by the state of North Carolina and 70 company members, the Semiconductor Research Corporation (SRC) founded in 1982 by eleven semiconductor companies (with eventual participation by government agencies an 13 other companies), the Austin-based Microelectronics and Computer Technology Corporation (MCC) launched in 1982, and SEMATECH launched in 1987.41

Much of this growth was facilitated by the passage in 1984 of federal legislation to regularize the anti-trust rules around shared research between companies (largely because of questions arising out of the establishment of MCC in Austin). By 1994, over 350 US-based R&D consortium would be registered in the United States. These consortia have been a mix of national industry-based endeavors combined with region-specific endeavors often driven by local government needs. And like Smart Valley, most have had the experience of both industry and region having a complicated relationship given the need for local collaboration. What is clear is that the growth of such new forms of business association has become a fixture of the new economy.

The Microelectronics and Computer Technology (MCC) in Austin illustrates many of the lessons, positive and negative, that inspired the creation of Joint Venture, Smart Valley and the whole array of new regional business-to-business initiatives that emerged in the early 1990s. The success of Austin in attracting MCC and the SEMATECH consortia and using them to boost technology development in that region was cited in the original 1992 Joint Venture publication "An Economy at Risk" as illustrating the political infrastructure which was lacking in the Bay Area region.42

In turn, MCC had largely been inspired by the success of the Japanese VLSI (Very Large Scale Integration) Project consortia from 1976-79 which had allowed a group of Japanese companies to produce the manufacturing technology to challenge US computer and semiconductor manufacturers. Ten of those US electronics firms, including Silicon Valley firms like AMD and National Semiconductor, would launch the MCC consortia in 1982 and would eventually recruiting twenty-two participants by 1993. The companies involved were geographically spread across the country and saw the need to focus cooperative research on developing new technologies to compete, both internationally and with IBM.

Where to locate the consortia was the largest issue facing MCC's founders. Texas was selected not only because of its large economic subsidies of the project (over $70 million in leased property and low-cost loans) but also due to Texas's massive upgrading of the University of Texas at Austin's microelectronics research program. This included tripling the program with the establishment of thirty new endowed professorships in electrical engineering and computer science. The MCC founders had wanted to avoid the top electronics areas where their project would get little attention, yet were able to work with a newly world class electronics faculty--the best of both worlds in the view of the MCC companies.

With investments of $500 million over ten years of operation, the consortia would file for 117 patents, license 182 technologies and publish over 2400 technical reports. The Austin region gained new recognition for its already established base of companies and company branches (Tracor, IBM, Texas Instruments, Motorola, AMD) and twenty major companies would locate sites in the city over the next decade. And by the 1990s, new firms were spinning off MCC Ventures Inc. (a commercializing subsidiary of MCC) and a new Austin Technology Incubator (funded by the city government and local chamber of commerce) had helped create forty additional new startups by 1994. By the late 1990s, Austin and its suburbs were home to over 900 software companies and the largest concentration of microchip plants outside Silicon Valley.

LESSON 1: REGIONAL SPINOFFS TRUMP CORPORATE TIES

Yet the companies who created MCC were not happy with the results. Their goal had been to create cutting edge technology breakthroughs for their own companies, not boost Austin's economy. Technology transfer back to corporate headquarters and manufacturing facilities outside Austin largely failed. The creation of MCC Ventures to commercialize MCC's technology was actually an admission of that failure and an attempt to recoup some of the funds invested in the consortium. What this highlighted was that local collaboration on research and local conversion of that research into commercial products and companies still overwhelmed internal corporate communication about technology mediated over geographic distance.

When SEMATECH was formed in Austin in 1987-88 by the federal government and top semiconductor firms, almost half of them from Silicon Valley, it applied those lessons. It concentrated on a clear development program with some funds (nearly $200 million per year in the early 90s) being directed both to outside research centers beyond Austin and much of the research being done in collaboration with company sites located in Austin in order to gain the benefits of the applied research program. One prime example was Silicon Valley-based Applied Materials Inc., the top American semiconductor equipment supplier, which created a design site in Austin specifically to be near SEMATECH. Other companies followed suit in expanding office sites in the area to take advantage of SEMATECH's results.

Many of these firms became backbones of the Joint Venture consortium; Applied Material executive Tom Hayes in fact became Joint Ventures first chairperson, and all the companies involved would take advantage of the lessons of MCC and SEMATECH in highlighting the gains to local regions of the existence of consortia established in a region. And the lessons of consortia like MCC seemed to show that far from fearing that the inclusion of national and even international participation would hurt regional actors, the inclusion of far flung participants was more likely to support the local region than the other way around. This was one reason why the leaders of Smart Valley initiatives so confidently could welcome national and international participants knowing that despite electronic communication, technology transfer would still be biased towards those closest geographically to the consortium work.

LESSON 2: THE NEED FOR SOCIAL LEVERAGING OF RESEARCH INTO COMMERCE

For Austin and other regional economies, another lesson had been learned. Political coordination of social institutions in a region were crucial for attracting technology resources to the region and, in turn, ongoing social coordination was crucial to making sure that those technological resources would be turned into businesses that would feed employment in the region. Despite the presence of MCC in Austin from 1982 onwards, almost no business spinoffs were created from its work until institutional support was built in the form of MCC Ventures Inc. and the Austin Technology Incubator. Turning technology into regional economic growth followed a step-by-step institutional process that helped create the institutional ties, the "social capital", that could translate into technology exchange and economic development.

When Joint Venture was formed, many business leaders felt that the drop-off in technology leadership by the region in the late 80s had strong institutional sources. In evaluating both the failures of the 1980s and the recent successes in Silicon Valley of the mid-1990s, Joint Venture's publication The Joint Venture Way: Lessons for Regional Rejuvenation would argue "Social capital is the glue that holds together a successful economic community. Before Joint Venture, the stock of social capital across the public and private sectors in the region was low...Joint Venture has repeatedly demonstrated that collaborative processes can create social capital."43

Silicon Valley's success over the years can be tied directly to an ongoing rejuvenation of its social institutions and relationships that in turn could attract investments in technology and then, in turn, commercialize those investments into economic development in the region. Its origin came largely from the Southern Pacific fortune of Leland Stanford with that fortune itself the result of the close cooperation of the "Big Four" Southern Pacific partners of Collis Huntington, Leland Stanford, Charles Crocker and Mark Hopkins in mobilizing every civic connection they could (including of course Stanford's position as Governor). They would parlay a combined investment of less than $7000 between them into a fortune of hundreds of millions of dollars, largely generated through federal government funds. While larcenous in its looting of the federal treasury, it was marked by a level of ongoing cooperation among the partners and civic institutions quite in contrast with most of the cutthroat railroad operations across the country and that civic participation played a key part in their success.44

As noted last chapter, Stanford University itself was oriented by Leland Stanford towards engagement with the surrounding region's economic life. Frederick Terman turned Stanford into an institutional cornerstone through his leveraging of federal funds in support of new businesses in the region. The threat of the end of some of those federal contracts and the need to mobilize to keep them coming to the West in fact led to the creation of one of the first formal West Coast electronics business institutions, the West Coast Electronics Manufacturing Association (WEMA), later to become the American Electronics Association (AEA).45

The second phase of regional institutional organization around technology was ongoing support for building of the social institutions, formal and informal, that assist commercialization. That took the form of Terman's back-handed industrial policy via the Stanford Industrial Park, the support of local community colleges training talent in the region, the more informal networking facilitated by Fairchild's splintering and the Homebrew Computer Club's bringing together of hippie entrepreneurs with established research scientists in spawning the personal computer revolution. When the AEA's offices were moved to Silicon Valley in the 1960s, it ignored lobbying in favor of fostering the development of new companies and promoting networking among businesses. Other more specialized associations like the Semiconductor Equipment and Materials International (SEMI) and the Software Entrepreneur's Forum (SEF) would promote similar shared information in their industries.46 The Santa Clara County Manufacturing Group would work closely with the public sector in the 1970s in order to improve the region's transportation, housing and environmental problems. Anna Lee Saxenian emphasizes how the civic engagement of these Silicon Valley business associations contrasted sharply with the aloofness, even antagonism, of Massachusetts's Route 128 high-tech firms to working with the public sector (and often each other). Route 128 firms preferred to see themselves as global companies and rarely as local firms with interest in the region; all of this contributed to the slowing of ongoing commercialization of technology in the Route 128 region.

Smart Valley, CommerceNet, and BAMTA became new iterations on this tradition of cooperation that would inevitably benefit the region. Despite their fierce competition in the microcomputer market, Intel and Advanced Micro Devices (AMD) along with Motorola would team up in 1997 with three Bay Area Department of Energy research labs (Lawrence Livermore, Sandia and E.O. Lawrence National Laboratories) in a partnership to work on dramatically shrinking microprocessor circuits- a $250 million investment by the companies that highlighted the public-private cooperation in the region.47

LESSON 3: REGIONAL AND GLOBAL NETWORKING IS MORE IMPORTANT THAN NATIONAL ORGANIZATION

Having gone from organizing to attract federal funds to consolidating production support in the region, the late 70s and 80s would see much more emphasis by Silicon Valley firms on organizing to defend their market against the Japanese. In 1975, the CEO's of National Semiconductor, Intel and AMD met over a meal and founded the Semiconductor Industry Association (SIA) that would spend the next twenty years promoting intervention by the federal government against the Japanese. The SIA would, along with an AEA newly oriented to national lobbying, lead lobbying efforts to demand that the United State government develop tough trade agreements with the Japanese government to protect market share for US companies and open up the Japanese market more to US firms. Most of the electronics consortia of the 1980s, from North Carolina's Semiconductor Research Corporation (SRD) to MCC to SEMATECH would be justified based on the Japanese threat. It was in this period that most of the original Silicon Valley semiconductor firms started defining their home "region" as the United States rather than Northern California, and local business institutions suffered from the neglect.

Ironically, it was the lessons of Austin that would reinforce the advantages of regional location for innovation and show the limits of national efforts such as MCC and SEMATECH. While both of the latter did assist US firms in regaining competitiveness in real ways, the limits of that help to national firms located far from Austin, along with the much more dramatic gains for those locating in Austin, delivered a solid message to the Silicon Valley firms of the forgotten advantages of regional proximity. Whatever the national participation in these consortium, it was those firms who strengthened their regional presence in the Austin area who reaped the greatest benefits from the consortium's technology spin-offs.

At the same time, the nationalist thrust of the 80s was being tempered by the reality of an increasingly global production system in electronics. Many companies based in Silicon Valley would get as many as 60% of their components from Japanese firms, often using electronic communication to wire designs to manufacturing facilities in the Far East.48 Firms like Sun Microsystems were forming tight relationships with suppliers like Fujitsu. Even the decidedly nationalist SEMATECH, with an explicit buy-only-American policy, began seeing itself less as an institution to build US firm strength versus Japan firms in a head-to-head struggle. Rather, it aimed to "help U.S. companies gain sufficient technical muscle to join international partnerships...from a position of strength" in the words of MCC/SEMATECH chroniclers David Gibson and Everett Rogers.49

With the establishment of CommerceNet and BAMTA, there seems to be have been barely a thought of stopping at national borders once non-regional members were accepted. BAMTA Project Coordinator Karen Greenwood notes the "tension between regional and global [and] I say global rather than national" since she sees that as the opposition driving the economic focus of firms in the electronics industry in the 90s. Like most involved, Greenwood saw the international participation merely strengthening the lessons learned for all involved, since their participation would inevitably do more to help the overwhelming number of Bay Area companies that the other way around.50

LESSON 4: STANDARDS, NOT RESEARCH OR PRODUCTION RELATIONSHIPS, ARE THE LARGEST FRUIT OF CONSORTIA

Despite the best intentions of their founders, most private sector initiatives can't deliver the basic research they need for break-through innovation or the just-to-market techniques that companies want for immediate production. After Craig Fields, head of DARPA, was fired from his job in 1989 for promoting ties between government and business as the key to moving the innovation of basic research to market production, he was hired as head of MCC in 1990. After talking with consortia members and examining the tensions in the project, he noted that the four goals of the organization included (1) giving competitive advantage to member companies; (2) helping the United States; (3) advancing science; and (4) advancing industry. Field concluded that these goals were mutually exclusive:

Twenty years ago...these four criterion were reasonably aligned. The way the world has gone, that correlation is much, much lower. Almost all large companies that are part of consortia are multinational, so helping the company doesn't necessarily create high-value jobs in the United States. For many companies, business success is only modestly related to technology advancement; it's more closely related to issues of good marketing, closeness to the customer, and continuous small incremental gains and improvements rather than some breakthrough technology.51

His conclusion was that basic research was useless in private consortia and should be left to the public sector. Private consortia should concentrate on the public goods it could deliver for its members and the public, namely the standards and agreed-upon protocols that would make collaborative production easier in commercializing the basic research supplied from the public sector.

Partly, this was based on a revised view of what had allowed the Japanese to supplant US semiconductor firms in the late 1970s through their VLSI consortium. While the research from that collaboration of Japanese firms was seen as fruitful, including creating the then-breakthrough 64K random-access memory chip, what became most critical for the commercial success of Japanese firms was a new shared standard among Japanese firms for production. This have a boost to local Japanese contractors and equipment markers, especially compared to US equipment suppliers which had a fragmented set of standards for production. With those new standards of production, the dependence of the Japanese companies on foreign equipment suppliers dropped in three years (from 1977 to 1980) from 75 percent foreign-dependence to less than 50 percent.52 In turn, one of the most successful initiatives of MCC has been the creation of the ATLAS Standards Laboratory as an independent subsidiary to speed the creation and adoption of electronic information standards in a variety of MCC related research areas.

CommerceNet and other Smart Valley initiatives took these lessons and worked to create consortium relationships that would create these less tangible public goods. Venture capitalist Bill Davidow used the lessons he learned as an executive at Intel to help launch Smart Valley and says:

I see these things, whether SEMATECH or Smart Valley, or Joint Venture, as attempts to address the issues of an economy and a government in transition...These institutions spring up in order to drive some of these transitions or put band-aids on situations that need repair. So what's going on is that we are going to an information-based economy where what is important in society are the intangibles and not the tangibles. It's not that food isn't going to be important, or a roof isn't going to be important, but a lot of wealth will come from an intangible base. What I see Smart Valley doing is building up the intangible asset base in Silicon Valley so it can be competitive in the world economy.53

No basic research was even attempted at the Smart Valley initiatives and, since they could draw on the participation of engineers in companies already located in the region, the consortia could maximize use of the money that they had (orders of magnitude less than SEMATECH for example). They concentrated on the networks and standards promotion that would build that asset base and expand the electronic marketplace of public goods. Whether developing the best system of on-line cataloguing or brokering agreements on standards for secure financial transfers on the Internet, CommerceNet drew on those already involved in the industry to push forward

Government funding's role both facilitated cooperation towards the new electronic commerce standards and gave the results of that cooperation more legitimacy. "There is a movement towards these processes," notes Smart Valley's Michael McRay. "By admitting government funds, it makes it happen faster and gives presence to the consortia. Market forces will eventually come to bear and produce similar results, but many years down the road. The funding is a catalyst." Many in the consortia have expressed worry that recent cuts at the federal level may endanger the sort of cooperative endeavors like Smart Valley and allow competition over proprietary standards to impede the development of the technology.54

While consortia members hype the advantages to the public of more stable standards and a quicker development of an electronic marketplace, they are also clear on how the consortia members benefit as well. First and foremost, by helping companies get in early and define the shape of the electronic marketplace, companies involved get a strong leg-up on later competition. "It's pretty clear from how this is going that early entrants to this area have a significant competitive advantage, and if you wait a year, you may be out of business," argued CommerceNet Vice-Chair Bruce Lowenthal. By working together, consortia members structure a marketplace that reinforces the technical and social advantages they already have in their firms. On the broadest front, the goal of the firms involved was to make sure the Internet became the dominant networking standard: "To the extent that you accelerate things, alternatives wither It makes non-IP networks extremely unlikely." This helped a range of firms, mostly centered in the Bay Area, whose existence had rapidly become tied to the Internet versus companies promoting proprietary private networks that were based mostly outside the region. Eran Gross, who oversaw the BAMTA efforts on Health Care and Aerospace, summarized BAMTA's goals as "helping to encourage an open platform environment, not just so it will work on all machines, but by doing things together help create a model...to be duplicated."55

From a more specific company position, Lowenthal noted that his company, Tandem computers, was in a better position to influence not only the technological standards adopted for electronic exchange but also what corporate allies benefitted as well. "If you're your in the pilot projects, you help develop the standards that are being used, so it is oriented to the strengths you have, technical or personnel." Tandem has had large banks as key customers for years, so Tandem happily worked through CommerceNet with the bank company members to make sure that any financial security transfer systems favor banks over upstart financial competitors. Since some industry will have to issue digital "keys" to authenticate financial transactions, Tandem wants technology that favors their banking allies. Tandem computers supply the mainframe computers that process over two-thirds of all credit card purchases and 80 percent of all ATM transactions, so Tandem had an interest in making sure any standards would be easily integrated with those banking assets. "It's more in our interest to have banks do the certificate-issuing. Banks know us, they like us, they get to meals together with us, so it helps us if they create the standards. [By participating in CommerceNet], we not only know the technology but our friends are doing this" so it cements corporate alliances around the technology.56

Similarly, BAMTA was created so that companies spanning the multimedia production process from content creation to distribution to software could create technologies that work smoothly together, thereby giving them an advantage over those attempting to create technologies in isolation. The creation of the BAMTA Collaboratory, literally a physical lab where different hardware and software could be tested together before commercialization, was designed to give companies a legup before full market testing. Kodak committed to provide digital-processing equipment to the venture specifically so that other companies would design their software with Kodak's equipment in mind. For smaller companies in the region, it allowed them access to an array of equipment that might be prohibitively expensive to test in different configurations. All of this facilitated the creation of standards that can integrate an array of different equipment, but focuses on the specific integration that favors those who learned how to design for that standard in the first place and whose equipment was the base around which the standard was shaped.57

The Role of Standards in the New Economy

So what ultimately are these standards, these "intangible assets" in the role in the economy?

In the production of already established goods, standards push the economy to regularize production and decrease the dependence of smaller suppliers on proprietary standards that leave them hostage to the strategic decisions of large manufacturers. These open standards allows fixed hierarchical relationships between firms (and within firms) to be replaced by a much broader array of market relations in goods. Smaller firms can have less fixed investments in specific relationships (i.e. goods and training associated with proprietary standards) and are therefore freer to deal with a multiplicity of firms in the market. By reducing what economists like Oliver Williamson label the "transaction costs" of dealing with different firms, market relationships can supplant hierarchical ties between firms.58 In this limited sense, electronic commerce around standardized Internet protocols promises a more marketized set of relationships between buyers and suppliers in a range of industries. In a related manner, electronic commerce through vehicles like smart catalogs address another traditional reason for tight buyer-supplier hierarchies, namely limited information about the availability of alternatives.

However, as the evolution of CommerceNet and Smart Valley showed, there is a deeper effect of standards in the new information economy, namely the creation of a whole new marketplace for a range of information goods that had never existed before. Instead of just changing buyer-seller relationships in an established market, the creation of the Internet has brought into being a marketplace for a whole new range of information goods. And in every step of the creation of that Internet-based marketplace, trust has had to be built around standards so that new buyers would participate in this new marketplace.

CommerceNet's Randy Whiting posed the distinction as between improving the functioning of individual markets, driven by buyers who might want more efficiency from their contractors, versus creating a marketplace, where a collection of sellers want to create whole new markets to sell their goods:

There was an impetus for a lot of other companies to participate, to think of this not just as a market but as being a marketplace where multiple markets could exist...A marketplace is more of a platform and an environment where I can have multiple series of products with a certain locality, in this case a virtual modality...When you're on the Internet, there is a huge cost of infrastructure that will not come from the [individual customer's] side, so there is a large push for the seller to create the infrastructure. "Buyers make a market, sellers make a marketplace."...The founders [of CommerceNet] said they had to get major corporations in, and the corporations, being very entrepreneurial, saw an opportunity to sell stuff; they wanted a marketplace to pull their buyers in. That's why the banks and others came in later, rather than small contractors. I think the initial idea of pilots was to try something, see what happens and what works, then let's test these things and see how to make it more efficient to create markets. If we build these structures that enable the marketplaces, it allows these markets to spring up.59

In the information age, much of the infrastructure are the "intangible assets" of standards which convince customers that goods they purchase will be compatible with other information-goods-- the holy grail of interoperability--and that they are not buying into a dead-end technology. Standards also build confidence by developers of supporting technology or "content" that they are not wasting their time on a here-today, gone-tomorrow proprietary system that can be changed at the whim of the system owner (as happened to the developers of Microsoft's original proprietary Microsoft Network.) The whole process of creating an open standard is tied to building trust between a whole range of companies and the buying public. In the case of the Internet and its related standards, this encompasses an incredible range of industries from computer hardware makers to telecommunications companies to banks to software developers, making the trust issues all the more difficult.

The regional nature of the standards process comes from the nature of trust itself. There is nothing that bars organizations based in different physical locations from building the trust needed to make the standards agreements that could facilitate such new marketplaces, but organizations are run by people and it usually takes day to day interactions and embeddedness in a range of social spaces that transcend specific business relationship to create that trust. And in pathbreaking new markets, it usually takes the assistance of government, which overwhelmingly operates within regional spaces, to help referee the relationships that allow trust relationships to develop and help legitimize the standards that come out of such processes. The original Internet developed under the supervision of the government operating overwhelmingly in three or four geographic regions and the federal government's drawing back from supervision of the process undermined the ease of building trust in far flung geographic spaces even further.

The role of Smart Valley plus the range of regional companies deals around building the Internet-based electronic commerce standards reflected the need for some regional focus for even a global networking project like the Internet. Regional participants in the Bay Area welcomed national and international companies to join, since their ideas could only strengthen the acceptance and trust in whatever standards were supported, but the day-to-day work of the network would depend on the personal ties deriving from the Silicon Valley region. Over and over again, CommerceNet and BAMTA members would emphasize how personalistic the organizations was in operation and how dependent success was on the individual ties people could build through the process, far beyond the formal business-to-business relationships they brought to the organization.

And with 50% of the organizational members located within the Bay Area and over 75% of day-to-day participation coming from within the region (along with the highest concentration of Internet domains in the world), the region was bound to have a disproportionate influence and derive disproportionate benefits from the evolution of an Internet moving forward in the direction desired by them.

All technology-based economic changes have had similar regional concentrations that helped facilitate the creation of new markets, either through monopolistic proprietary agreements or in the more diffuse models of production districts. Whiting, a Nebraska native, compares CommerceNet's role to the agricultural stockyards of his hometown of Omaha where mercantiles, slaughterhouses, banks, grain markets and the physical transportation infrastructure for cattle created whole new markets for pigs, cattle and secondary products that could be create out of this physical concentration of producers. (In fact, Whiting's graduate thesis was on "Alternative markets for pig manure waste disposal systems", a worthy preparation for dealing with a lot of software code leftovers). "What we're doing is facilitating the Internet creating a marketplace, just like the Omaha marketplace. Just as you had the most significant locations for cattle--Kansas City, Omaha, and Chicago in early part of the century all on rivers--the Bay Area had the right things in place to add the things to the Internet to make things easier, add security, metrics" and the whole range of standards to pull companies into this new marketplace.60

If one wants an example of the hazards of standards forged outside such regional negotiations, a perfect negative example has been the battles over the Digital Video Disc (DVD), a critical storage medium as computers converge with television. DVD's are a giant advance on compact discs which would allow up to seven hours of music, a full-length movie or massive amounts of computer data to be stored and accessed from a single disc. A consortium of ten global technology and media companies, including Time Warner, Sony, JVC, Phillips Electronics and Toshiba initially agreed on standards for the devices, but the result was anything but an open or stable standard. The interests of Hollywood and the recording industry led to tight controls on how easily information on a disc could be shared with computers or other machines - the antithesis of the ease of information exchange on the Internet. Worse, as new machines were introduced in 1997, the consortium partners rapidly broke down into warring standards camps as each sought proprietary advantage over its global rivals; by 1998, there were no less than six mutually incompatible DVD standards in development or on the market.

The result: instead of the 3 million machines expected to be sold in 1997, only 500,000 DVD players were sold that year as consumers sat out the standards war. An expected 22 million sales in 1998 would quickly fall to an estimated 3 million DVD-devices. The failure to agree on standards had driven a stake in a potentially explosive market. In the absence of governmental involvement or some degree of regional coherence, global competitors had found no way to maintain agreement on standards, thereby hurting not only their rivals but their own share of a much smaller pie.61 Without organic connections and the iterated interactions of regions that favor concentrating on growing the pie over slicing a bigger share, these global rivals starkly highlighted how a failure on standards leads to broader economic failure.

Silicon Valley's Regional Model versus its Rivals

The problem for Silicon Valley firms is that even geographic proximity does not automatically create the standards that propel economic growth, especially in the absence of a firm alliance with government. The drive to create CommerceNet and the whole host of assorted consortia was an acknowledgment of the need to structure those regional advantages in a way that built confidence that all participants would gain from the expanded growth of the Internet.

The reality is that despite the Internet's success, many of the firms in the region continually struggle with the danger of proprietary technologies upsetting the trust needed to sustain the collaborative model that has fueled the growth of firms in the region. At the top of the list of dangers is of, course, Microsoft. Microsoft used a combination of its early alliance with IBM and hardball tactics to build its proprietary operating system monopoly on the desktop. From that base, Microsoft would extend its proprietary standards in the 1990s increasingly into the market for large-scale business computing, formerly the province of mainframes or UNIX-base network servers. While the Internet at first appeared as a danger to Microsoft, even a dagger at its throat, Microsoft also saw that success in molding those standards in a proprietary direction could extend the company's control throughout the whole world of corporate computing.

Through a combination of in-house software applications and developer tools optimized for its proprietary standards, Microsoft created an all-pervasive computing environment that promised any corporation that its needs would be met. The Microsoft solution might be less innovative than any particular competitor but Microsoft's very completeness and pervasiveness across all sectors of computing would make up for its rigidity.

In fact, Microsoft's rigidity could be an advantage when compared to the weak standards that pervaded the UNIX corporate environment in the 1990s. After the heyday of the 80s when government purchasing requirements had enforced a broad UNIX standard on the industry, the industry had divided into warring UNIX camps and left customers uncertain that their needs would be met in the fragmented UNIX environment. By 1997 Microsoft NT computer servers were outselling UNIX servers.62 Despite the odes by writers like Saxenian to the flexible design relationships in Silicon Valley, it was clear that in the absence of strong standards and government support for such standards, proprietary models had a decided advantage in yielding the market stability and monopoly rents that a company like Microsoft could reap.

The geography of this showdown was clear. With Microsoft dictating standards from its Redmond headquarters based on chip designs from its ally Intel, allied companies or even grudging competitors writing for Microsoft's operating system could locate anywhere they could attract a good workforce, from Utah to Texas to South Dakota. When Bill Gates was called before the Senate Judiciary Committee in March 1998, Microsoft's defenders on the panel included Michael Dell of Austin-based Dell Computers and Douglas Burgman of Great Plains Software (based in North Dakota), while Gates critics were notably all from Silicon Valley.

While national and global alliances would be important in Silicon Valley, the continual innovation and interaction between hardware and software advancements would require much more localized development in the region (and to a lesser extent in Route 128). In the 1990s, one response was the rise of massive manufacturing subcontractors like Solectron whose Milipitas CA factory would be doing manufacturing simultaneously for companies including IBM, Hewlett Packard, Cisco, Motorola, Sun, and Apple. To a certain extent, single contractors like Solectron were replacing the more diverse networks that had been key to manufacturing in earlier years. "The teams here are oriented around customers," Solectron California's President Walt Wilson would say, "and they're fiercely loyal to them. You can go out in any of [our] buildings and ask anyone what team they're on, and they'll tell you. They'll say they work for Intel or IBM or HP and that Solectron just signs the checks. That's customer focus at its best."63

But even as Silicon Valley firms sought to finesse manufacturing innovation from the economic pressure of Wintel competition, the UNIX wars made clear that strong open standards were the key to the region competing economically against proprietary steamrollers like Microsoft. The consortia like CommerceNet built around Internet standards were the first step in the process, but companies like Sun and Netscape saw the need for broader solutions that would expand open standards from the operating system to the tools used by programmers. The Java language, with its promise that any program would be able to run on any computer, no matter its hardware or operating system, became a key part of that strategy. As described in Chapter 3, Sun worked assiduously to build the consortia links that would turn Java into a broad standard. And the explosion of Java-based firms in the Silicon Valley reinforced the idea that regional firms would reap the greatest advantage from collaboration on such open standards.

Still, Microsoft responded by seeking to coopt Java and other Internet standards into its proprietary approach by creating versions that would run only under Windows environments, thereby violating the whole point of these standards. And its continued success in linking contracts for its Windows operating system to installation of its browser enhanced worries in Silicon Valley that Microsoft could engineer a proprietary approach to the Internet.

With the need to generate stronger global support for its standards, Netscape took the unprecedented step in March 1998 of publicly revealing its browser source code - the usually top-secret guts of any program. Netscape invited anybody who wished to modify the code and even resell their own version as long as any modifications to the code were republished publicly on a special web site. Faced with the onslaught of Microsoft's proprietary approach, Netscape decided that the regional commercial commitment to developing standards was insufficient. It needed to marshal the resources of the global programming community and it needed to open its code to gain the kind of trust needed to ensure their support.

The idea was, harking back to the original ARPANET vision, to invite the participation of the whole Net community in developing the tools and standards embedded in the browser software. "It's no longer Netscape alone, pushing the client software forward, but now it's really the whole Net," said Bob Lisbonne, Netscape's senior vice president for client products at the time. "For Netscape, this gives us a way to engage the creative, innovative abilities of literally orders of magnitude more people than we could ever--really any commercial software company could ever--afford to just put on their payroll."64 With hacker enthusiasts lauding the decision, thousands of developers would download the source code within the first day and major modifications of Navigator were released onto the Net by independent developers from all over the country within weeks. The idea was that Netscape could release exiting modifications in its continual upgrades of both browser and server software. What it lost by giving up control of its code, it would make up through selling customized business versions and server software and by preventing Microsoft's control of standards which would be Netscape's deathnell.

Netscape's action highlighted the continued importance of public-interested software development, often called freeware, which had survived much of the privatization of the Internet. Most dramatically, despite the focus on the Microsoft-Netscape rivalry, the most popular Web server on the Internet was neither company's but rather a free system called Apache. After the NCSA developed its Mosaic browser software and its original server software, the NSCA as part of the government privatization had ceased aggressively updating its software. Instead, a geographically dispersed group of software programmers, some at universities and some in private business, began collaborating in 1995 to update the NCSA server to increase its power and manageability. Most of the programmers participated out of altruism and the long-term hacker ethic of promoting free software. The result was a Web server that in 1997 was used in 44 percent of Internet sites, compared to just 16 percent that used Microsoft and 12 percent using Netscape's software. And that list of sites included McDonalds, UUNET Technologies, HotWired, Yahoo Inc., CSB, the FBI and IBM, which passed over its own Lotus Domino server in favor of Apache when it put its "Big Blue-Gary Kasparov" chess match on the Internet.65 Similarly, one of the favorite Web programming languages is a free language named Perl which has similarly been modified and improved through a global network of collaborators.

Most dramatically, Netscape announced that it would begin making all its software applications available in the Linux operating system, a freeware version of UNIX that has become the fastest expanding operating system in the world with three to nine million copies on computers around the world. Linux was described by Wired magazine in 1997 as "[Window] NT's most serious competitor, the only viable alternative to the Microsoft monoculture."66

Remarkably, Linux was started in 1991 by a student at the University of Helsinki in Finland whose first name, Linus, led to the naming of the language. At that point, a whole series of free UNIX tools had been developed by programmers connected to GNU (a self-referencing title standing for GNU is Not Unix) foundation, itself founded by an original MIT hacker named Richard Stallman who objected to the increasing commercialization of university research. What this network of free software hackers lacked was the core of the operating system, called a "kernel", which would tie all these tools together into an alternative to the commercial UNIX. Linus Torvalds wrote that kernel and from his university post would use the Internet to coordinate improvements in this new operating system with help from hundreds of enthusiasts around the globe.

Freely distributed based on what GNU called "copyleft" principles - the license Netscape adopted when it released its source code - the Linux operating system could be distributed freely or packaged with documentation and sold for modest amounts backed by technical support by companies like Caldera and Cygnus Solutions. Extremely popular in developing nations like South Africa, Cuba, India and the Philippines, Linux also began to eclipse other forms of UNIX in the US partly because of its price but also because many people considered it technically the best operating system in existence. Linux was the first operating system to include Java capability, so every increase in Java programs adds to its functionality.67

Netscape's source code unveiling and its announced support for Linux throws into relief the different geographies and economic of trust that separate proprietary standards, commercial open standards, and freeware standards. With proprietary standards like Microsoft's, everyone trusts (or fears) that Microsoft will enforce whatever standards it dictates from its company-specific development, so hardware and software partners on such proprietary products have a stable enough environment to locate almost anywhere they think appropriate. Alternatively, freeware collaborators on free software systems have increasingly use the Internet to build trust based on altruism and the hacker ethic of achievement without needing to share any geography - the extreme example being Linus Torvalds direction of the evolution of Linux from his lonely station in Finland. Without expectation of financially capturing the social benefits of their creation, they are free to innovate without restriction. On the other hand, with commercial open standards, collaborators need the repeated interactions and day-to-day commercial interactions of shared geography like Silicon Valley to generate financial gain while assuring that multiple collaborators all profit from innovation. With Microsoft's proprietary approaches gaining ground, Netscape and other Silicon Valley actors reluctantly saw an alliance with the global freeware model as necessary for survival - abandoning some profits in order to maintain the priority on innovation that gives them a regional advantage in the remaining commercial aspects of technology development.

Silicon Valley as Corporate Center for High Technology

However, even as the leaders of Silicon Valley are making alliances with the free software radicals of cyberspace, it is in its other international alliances that the most radical changes in the cooperative shape of the region are taking place. With the scale of the Internet, there are vast implications for the changing place of the Bay Area in the global economy and the role of less skilled workers within the region's politics. Researchers Richard Gordon and L.(?) Kimball argue that physical location of design and production centers has become more and more disengaged from any objective measures of infrastructure (traditional or research assets) or inherent production links, but instead reflects the placement of firms within an international distribution system. Multinational firms increasingly see regional locations as a place to strategically distribute complementary products for maximum corporate coordination. The key to understanding Silicon Valley's dominance "no longer concerns discovery of those objective factors likely to promote independent generation of indigenous regional high-tech agglomerations but comprehension of a region's social and economic position within the corporately organized integrated circuit of global production."68

Agreement on Internet standards shadows the integration of new players into dominance of the coordination of key global industries, from media to manufacturing to banking. Just as New York City became the central coordinator of international capital and corporate governance in the 19th century, the Bay Area is making a bid through its integration of diverse industries to assume large portions of coordination functions in the coming 21st century. Venture capitalist and key Smart Valley founder Bill Davidow noted in 1995, "If you look at what's happening in the economy in the last few months, Silicon Valley has become the headquarters of the network-based industry." He argues that the East Coast is five years behind the Bay Area in implementing and integrating their economy around the new information economy and in a rapidly changing economy, that is giving a lead to regional firms that is giving them leadership over a spectrum of industries. The Bay Area has a combination of finance, defense, semiconductors, biotech, medical devices, software, fashion, media and tourism that, combined with the new technology, is creating an integrated corporate coordination center.69

The nature of this fusion of industries tied to the Internet is very much one of elite cooperation, based on organizations like Joint Venture and Smart Valley which built the trust that fosters the corporate coordination needed for regional economic ventures. Older corporate centers like New York are furiously mobilizing new cooperative ties to hold onto their global status. Even as Silicon Valley has organized ties with San Francisco's "multimedia gulch" to leverage technology into "content" dominance in software and new forms of publishing, New York is seeking to use its traditional publishing and corporate ties to leverage its publishing houses and traditional media into dominance of the new media. Already in New York City, the 18,000 people in over 1,100 new media firms outnumber those working in either television or newspapers in the city and a total of 71,500 workers in the whole multimedia industry there generate $3,8 billion of revenue for New York-based firms. New institutions like the New York New Media Association are cultivating a "Silicon Alley" where firms cluster next door and can even cut deals in the elevator.70 Whether New York or the Bay Area win out or what portions of global coordination they end up sharing with other global cities around the world is all part of the economic conflict of the developing technology. However, comparisons to New York City also bring comparisons to the way that the elite of New York, cultivating international ties around the globe, eventually severed much of the relationship between their prosperity and the standard of living of those at the bottom of the economic scale.

The company leaders of Silicon Valley admit as much in interviews and the founding report of Joint Venture: Silicon Valley sketched out that viewpoint in rather stark terms. In looking to the future of the region, the report proposed three future regional scenarios: a "High-Tech Manhattan" of corporate headquarters and little else, a "Virtual Valley" of niche companies allied in stable industrial clusters, and an innovative "American Technopolis" of thriving startups and new technology creation. The latter would deliver the most for the region, while the other scenarios would be either a draw or a net loss for others in the communities of the Bay Area. However, the assumption of the scenarios was that in any scenario, the companies would do well. The worst scenario of High-Tech Manhattan still "results in a win for companies" even as it is "a lose for the regional economy and community."

While there was a preference, both personal and technological, by industry leaders in keeping a strong Silicon Valley economy, the assumption is clear that companies will win out in whatever scenario. In a thoroughly global economy, companies would seek out other regions that fulfill their needs if the Bay Area fails to do so. With this reality, any offers of collaboration with local officials take on more the character of an "offer they can't refuse" than a long-term mutually dependent relationship. The reality is that the economic action of technology innovation is inescapably local but the power of corporations to pick and choose their venues is global and outside the control of the local actors who desperately try to negotiate with these global partners. Those who have more power globally therefore have a stronger hand for negotiating locally within a region, while those unable to muster such power are out of luck. The end result is more inequality within regions.

Inequality and the Contingent Future for Non-elite workers

The immediate benefits of the Internet boom for non-elite members of the Bay Area community have been substantial in recent years with large job growth, low unemployment and wages rising at rates as much as five times the national average. By 1996, the average wage in the main Silicon Valley areas was $43,510 compared with $28,040 nationally.71 But what is clear is that few lasting institutions are being created that lock-in such shared benefits over the long term.

The mid-century wave of manufacturing growth in the Bay Area region and around the country was the creation not only of corporate cooperation with government around shared economic growth but was also the result of new institutions that built-in a share of that growth for most workers and community members. While Henry Kaiser and his fellow industrialists were enriched by the industrial take-off of World War II and its aftermath, labor unions also grew exponentially with the expansion of jobs in what was called the region's "Second Gold Rush." New health care institutions like the pioneering Kaiser Permanente health program would outlast the factories that funded their initial creation.72 In what researcher Harvey Molotch and David Logan called the age of "growth coalitions", cross-class cooperation would fuel growth in urban areas throughout the country. These growth coalitions would use the proceeds of growth to invest in a range of public goods that would benefit the broad public and lift the wages of all workers (even if the greatest benefits might remain in the hands of the elite).

While many community members shared in the initial burst of employment and higher wages following the Internet boom, it is unclear how real the shared wealth is for many workers and even less clear whether these benefits will be transitory for all but the corporate elite. From 1992 to 1996, median family incomes increased 14% in the main Silicon Valley area, yet average apartment-rental rates increased over 23%. Partly this is the function of cramped geography and rapid growth as Silicon Valley added 125,000 job in the period but only 26,000 new housing units.73 But the low unemployment rate in the region reflects as well people being forced to leave the area when they can't find the job. ``With the high cost of living, if you don't have a job, you can't afford to live here,'' publicly argued Lois Koenig, assistant to the director of the San Mateo County Human Services Agency, in noting the pressure on low-income families in the region.74

Even for those remaining in the region, the reality of the Internet "boom" was one of rising inequality and greater hardship for those at the bottom of the economic ladder. Even as engineers and professionals saw significant wage gains, wages for the lowest-paid 25% of workers actually fell 13% between 1989 and 1997. 20,000 residents of Santa Clara County found themselves homeless at some point during 1997.75 In the 1990s, poverty among children in the Silicon Valley region increased from 10% of children in 1989 up to 13% by 1995, with three-quarters of those children concentrated in only 30% of the census tracts.76 When President Clinton's advisory board on race came to San Jose for a hearing, they walked into a firestorm of anger as many of the 700 attendees, including San Jose school board member Jorge Gonzales, denounced the 400 percent rise in CEO saleries during the decade, even as the average wages of Latinos in Santa Clara County had dropped 6 percent.77

Beyond the increased costs of living and persistent pockets of poverty, there is a sense of the ephemerality of the boom for many workers in the region. If the World War II manufacturing boom pioneered new forms of job guarantees and widened benefits for all workers, skilled and unskilled, the recent Internet boom seems to be pioneering precisely the reverse. Including all categories of contingent workers--temporary, part-time, self-employed and contract workers--the best estimate of the size of the contingent workforce is 27% to 40% of workers in Santa Clara County, the heart of Silicon Valley. This sector of the workforce is growing two to four times as fast as overall employment and nearly all net job growth in the county in the last ten years was accounted for by the growth of contingent employment. Contingent employment has accelerated along with the Internet boom; in the first nine months of 1995 alone, employment in temporary agencies in the immediate Santa Clara area grew by 41%. Since the early 1980s, employment in temporary agencies has grown by 150%, a rate more than 15 times the overall employment growth in the region. There are more than 250 offices of temporary agencies operating in Silicon Valley and Manpower Temporary Services, the largest national temporary agency, operates 15 offices in Silicon Valley alone, placing over 5,000 people a week by late 1995. Much of this contingent work is facilitated and accelerated by the Internet where jobs can be posted, reviewed, accepted or rejected with greater ease. New IRS rules, enforced initially against Microsoft, which cracked down on use of independent contractors has pushed many high-tech firms to deal more with employment agencies, further regularizing the position of even the more professional class of temporary workers who had been able to operate as independent contractors.78

A report by Working Partnerships, a non-profit backed by the South Bay Central Labor Council which covers most of Silicon Valley, points out that what all these contingent workers share is that "their terms of employment stand outside the standard employment relationship on which the framework of employment and labor law was built. The fact that all contingent employees are outside the standard employment relationship means that they are vulnerable to rapid economic change and have difficulty being represented."79 To put it in other terms, if the Internet boom stalls, regional employers can instantly drop back to 1984 employment levels without laying off almost any full-time workers.

Already, large areas of Internet support and routine software programming are being outsourced from the region. CommerceNet leaders emphasize the number who have already moved out of the region. "Safeway [supermarkets] just announced that they are moving their data center to Arizona," noted Bank of America's Mack Hicks in 1995, "taking their best-paying white-collar jobs there, leaving the CEOs here to enjoy the Bay Area."80 In a period of high innovation around the Internet, technology companies have rapidly expanded local employment in the first burst of standards creation as immersion in regional dialogue has been at a premium. But if development of technology around the Internet becomes more routine, more of that employment will follow companies like Safeway out of the region. Overseas software programming is becoming a real threat to the more routinized programming jobs. India's 1600 engineering colleges and technical schools now graduate 55,000 students a year, contributing to a 45% annual growth rate in India's software industry. The south Indian city of Bangalore already does $280 million a year of software development work for Silicon Valley firms, including Hewlett Packard, Oracle, and Apple.81

This work largely focuses not on new products but on support for more standardized software projects, but the very success of Internet standards will create more such projects as the initial burst of innovation ebbs and Internet communication continues to facilitate global coordination of such routine projects. The CommerceNet Internet marketing working group concentrated on the ways to use the Internet to market from global corporations. The Internet is a natural way to direct customers, suppliers and internal company communication to various components of a corporation. With the Web, one click of the button can connect customers to on-line service, sales or technical support--all of whom can be in different cities and even different continents. All of this bodes ill for the more contingent members of the Bay Area technological workforce.

And beyond the likely geographic redistribution of work, there is the reality that new technology is being used not to just relocate but to replace workers. At the lower end of skill, on-line ordering forms will replace many data entry clerks as customers essentially do their own data entry work as they electronically register their order. New software increasingly replaces mid-range technical skills, creating an endless upgrading of skills that lasts only as long as new software explodes in the workplace and employers demand that constant influx of new support. "I worry about a potential financial bust on the Internet," argued Stephen Roach, chief economist for Morgan Stanley and a former Federal Reserve economist, "or a culmination of a software product cycle." Roach has been highlighting the way technology has been a key tool in corporate downsizing and increased profits and has worried that the current technology-induced cycle of business expansion will falter with the peaking of the current wave of technology innovation. 82

Even if the boom sustains itself, it is clear that the constant skill upgrading demanded by the technology is itself an instrument that divides the educated elites of the Silicon Valley region from other workers. Smart Valley's Bill Davidow acknowledges the inequality being driven by the new technological cycles:

We have relatively short electronic cycles, but if you look at what has happened culturally to people, it is ridiculous to talk about a profession going away in five years, but that is what you are facing. Two hundred years ago, if you learned how to make shoes at twenty, you could expect to be doing it at sixty. But you have people who learned what they thought were valuable professions, people who thought nuclear energy was the future or who got caught in the mechanical engineering profession when people found no one needed them for years...it's nice to talk about constantly retraining people and making them more adaptable, but the rate at which you can do that and the rate at which people will accept that is limited.83

Jobs are increasingly dividing into one set of jobs where a desperate upgrading of skills is required and another set of increasingly low-wage service jobs that are growing even faster.

Aside from global production, the other key element of the global economy, immigration, accentuate this divide. Rather than training less skilled US-born workers for new technical jobs, high technology companies, especially in Silicon Valley, increasingly hire foreign-born immigrants from places like India, Taiwan and other overseas Chinese enclaves. One in four people with science degrees in the United States were born in another country. Leaders of Silicon Valley firms became strong public advocates for the rights of (skilled) immigrants to the US and helped to defeat most provisions in the 1996 immigration law that would have adversely effected them. Conversely, Silicon Valley engineers increasingly are served by the influx of unskilled Latino immigrants (one ethnicity with markedly little representation in the professional strata of the region) in a range of menial tasks from janitorial work to fast food service to nannies.

In this way, the global economy--through global outsourcing, immigration, and the breakneck pace of skill obscelesence in the high-tech field (which only a minority can keep pace with)--are working to tear apart any career ladder or even mutual work dependence between these two classes of workers in the region. The public policy promoted by industry in the region increasingly serves only the upper core of professionals who are part of their regional organization of skills and standards. The rest are left behind as even the physical geography of the region is reshaped in a new politics of abandonment.

Regional Polarization and the Politics of Abandonment

If elite collaboration in the region increasingly concentrates on building the "intangible assets" of standards and trust, there has been a corresponding deemphasis on the more tangible public works assets that once tied regions together and, as well, benefitted the broader community. What electronic communication threatens to do is reconfigure the economic geography of regions and allow elite information workers to partially bypass the social problems of urban areas. Telecommuting, a prime project of Smart Valley, has emerged as one key component in that transformation. Telecommuting over the Internet opens the possibility of some of the largest reconfigurations of work and home relationships since the rise of suburbia and the possibilities of even larger changes in the future.

For decades, geology had locked Silicon Valley into the rather narrow corridor of the San Francisco peninsula and only slowly had it grown shoots over the mountains into Santa Cruz County and into the southern parts of Alameda County. However, under the pressure of limited housing, a whole new explosion of development had been built along Route 280, an interstate highway inland over the hills to the east of the urban centers of Oakland and Richmond (although assiduously avoiding those largely minority cities themselves). The sprawl has even reached out to more distant areas of the San Joaquin Valley. The number of people commuting into Santa Clara County from other areas of the region increased from 68,000 in 1980 to 113,000 in 1990, according to the U.S. census, and continues to explode.84 Connecting Silicon Valley to both the residents and the new satellite offices of many technology firms along the 280 corridor has become a greater and greater priority for Smart Valley firms.

Kathy Blankenship oversaw the telecommuting project for Smart Valley and emphasized the way telecommuting can expand the radius of effective collaboration in the region. "With traffic congested so badly in the region and housing so expensive, the region is threatened with losing a lot of skilled workers. The fact that 280 is completely blocked says a lot of people in the San Ramon valley are coming to the Santa Clara region. You could have a greater number of people with job satisfaction in the Bay Area with telecommuting. You'll have less people saying maybe I should get out of here and go raise ostriches in Montana."85

Smart Valley launched its "Telecommuting pilot project" in February 1995 and has continued to publicize its results to companies seeking the best way to implement it in their workplaces. "We decided to focus on telecommuting in its own right. There is a confluence of service, productivity, economic, and environmental benefits," said Eric Benhamou, a director of Smart Valley and president and CEO of 3Com. "Telecommuting doesn't depend on massive technology investments." The telecommuting pilot's goal was to involve a dozen Bay Area companies from different industries in telecommuting programs. Early participants in the beta test included 3Com, Hewlett-Packard, PacBell, Silicon Graphics, and Stanford University, 86

Blankenship noted that people are much more willing to commute larger distances if they can telecommute two or three days per week and come to the office the other two or three days for the face-to-face interaction needed to do their jobs most effectively. Smart Valley itself is an example of using telecommuting to enlarge its potential employee base: "Smart Valley has people from Danville, three people in San Francisco, we have a couple people in Portillo Valley/Palo Alto. We wouldn't have those people if people didn't know they could telecommute different days. That was a real benefit for me as a prospective employee coming from Danville. You still have to learn how to use that flexibility. Its a benefit they can give their good employees." 3Com's Eric Benhamou estimates that with 20% of employees as telecommuting part of the time, businesses can get 80% of potential benefits of productivity gains and expanding the regional space. The key is that by eliminating two or three days of commuting per week, telecommuting can make working in Silicon Valley bearable for these professional workers living in the new suburbs of the 280 corridor or San Joaquin valley.

This latter point raises the issue of what this new telecommuting geography means for workers who have to use the regular concrete highways and transit each day. It is a faint hope that the telecommuters will ease the traffic snarls and gridlock that are day-to-day realities in the Bay. Transit agencies like CalTrans have looked at telecommuting as an alternative to roads, but the result of taking some people off congested roads is that others see that the roads are clearer and decide to drive. With congestion at the levels around the Bay Area, as you take people off the road, new people will get on the roads.87

Overall urban concentrations of business are not disappearing in the age of cyberspace, as academic Manuel Castells has emphasized in noting that the costs of telecommunications infrastructure often reinforces top metropolitan areas where the volume of information traffic makes technical investments most profitable. Throughout the last few decades, larger metropolitan areas have ended up with a disproportionate share of information processing jobs. This regional reinforcement of central cities defies predications of spatial decentralization, but Castells does note the regional polarization between central cities and their suburban "urban villages" which combine the advantages of access to cosmopolitan centers without the costs of the central city.88

In the context of Joint Venture's push to relax government regulation and keep taxes down, its push on telecommuting begins to emerge less as a source of public interested policy than as developing the technology for elite engineers "coping" with the hellish driving situation in the Bay Area. With a crumbling road infrastructure and a declining tax base, the political support for infrastructure improvements that benefit the average commuter may completely disappear as higher-income telecommuters find their own strategies to cope with gridlock that are independent of the needs of many non-professionals. In fact, one of the main accomplishments Joint Venture claims for itself in its first few years is helping to pass an exemption from sales tax for most manufacturing equipment89--a billion-dollar hole in the state budget that came as savage budgetary battles were being fought trying to fund earthquake retrofitting for existing roads and bridges.

In the context of public policy for local transit, high technology executives have followed the same principles of supporting resources for their core of elite engineers living in or commuting to Silicon Valley while working to deliberately undermine revenue sources and transit that serves non-elite workers. In the core of Santa Clara Valley, millions of dollars have been poured into a new light rail transit system carrying just 21,000 passengers a day to stops at IBM, Adobe Systems, Intel and Cisco Systems, and to major destinations such as the San Jose Convention Center and San Jose Airport with connections to a new $244 million Tasman Rail Project that will bring commuters back and forth to Mountain View with stops at major employers like Hewlett-Packard, Lockheed-Martin and Netscape.90 With the rail system receiving an 85% tax subsidy for operating costs from local, state and federal funds, residents of poorer minority neighborhoods like East San Jose have complained at public forums that the bus system serving their areas have continued to be underfunded with few connections to the rail system, leaving them trapped with few transportation options.91

The technology firms of the Bay Area have been as targeted in pulling in funds for longer-range commuter trains, while undermining the public funds in other areas that would serve the transit needs of poorer residents. Along with helping push for expansions of the Bay Area Rapid Transit (BART) system that serves San Francisco and much of the Eastern part of the Bay to reach out to the new high tech areas of the Route 280 corridor, the Santa Clara Manufacturing Group have pushed for a new "Altamont Pass" passenger rail line from downtown Santa Clara through the suburbs of southern Alameda County and out to the San Joaquin Valley where over 31,000 residents were commuting to the rest of the Bay Area by the mid-1990s. With connections to BART and the light rail system, this $40 million line would help connect the emerging grid of suburban commuting lines in the region.

To fund the Altamont line, the high tech executives of Santa Clara had to convince Alameda County to contribute an initial $2.9 million to the system in 1996,92 a contribution that county made in the same period as it implemented an 11% reduction in services (including the elimination of all night service after 10pm.) in the AC Transit bus system serving 225,000 mostly poorer, minority residents of the East Bay. Bus riders in the system have formed the AC Transit Bus Riders Union to protest the billions of dollars going to suburban rail lines even as services are cut for poorer residents riding the bus every day. They have patterned themselves after the Los Angeles Bus Riders Union which in 1995, in conjunction with the NAACP, won a lawsuit forcing that county to allocate transit funds in a less discriminatory way.93 The East Bay group is small now but their formation reflects the economic polarization being fueled by transformation of the region due to the high tech industry's leaders.

The Economist, hardly a critic of wealth, has described the emerging Silicon Valley as "a grander version of one of California's less attractive creations, the gated community: rich, elitist and insular."94 The civic mindedness of its business organizations ends up translating into an extremely targeted use of public funds for the benefit of its core of innovative workers, while deliberately undermining the social infrastructure that does not serve its direct regional needs.

Nowhere was this reflected more than in the 1996 statewide elections, largely hailed as the debut of Silicon Valley firms as an active political force in the state. Locally in Santa Clara, the high tech business organizations strongly backed a measure to raise sales taxes for a nine-year $1.1 billion fund to pay for local rail and highway improvements to decrease local road congestion and support the light rail construction (with almost no funds earmarked for buses, notably). At the state level, the main event for technology executives was a $38 million campaign to defeat an initiative backed by senior groups and consumer lawyers, Prop 211, which would have made it easier to sue companies for defrauding investors, especially pension funds. However, as it became clear that Prop 211 would go down to defeat, Silicon Valley executives leading the campaign backed moving $1.75 million to defeat a statewide tax initiative, Proposition 217, that would have helped local governments across the state invest in schools and other local funding needs. If passed, the measure would have blocked reductions in income taxes passed by the state government for those making more than $110,000 per year, thereby raising $700 million for local government, funds that the state had taken from those local authorities during the state budget crisis in the early 1990s. The $1.75 million spent by the high technology PAC against Prop 217 was in addition to large individual contributions from Silicon Valley executives like Tom Proulx, the co-founder of financial software company Intuit who gave $410,000 to defeat 217. Prop 217 had been leading before this last minute infusion of Silicon Valley spending and was only defeated by a margin of 49%-51%. The end result of this fall political action was a commitment of new sales taxes for improvements in local Santa Clara transit, while the executives were able to block funding for less wealthy counties desperate for funds to serve their needs.95

Many Silicon Valley labor unions such as the building trades and local Democratic officials had, in defiance of statewide labor unions and the Democratic Party, supported the technology companies in their anti-Prop 111 campaign, but they were outraged when that campaign morphed into an anti-217 campaign vehicle in the last weeks of the campaign. But this merely reflects how disposable technology executives feel cross-class alliances have become, a marked contrast to the mid-century days when business and labor had created strong alliances in supporting funding for education and infrastructure statewide. In many ways, the new politics of elite business collaboration harks back more to the agricultural era of elite grower cooperation in California where substandard fruit was dumped on poorer California communities and food price hikes in urban areas led to food riots during World War I in that land of agricultural plenty.96 As will be described in later chapters, many local communities are experienced a similar technological famine while surrounded by the center of global export of the technology.

As the next chapter will document, even traditional industries like banking and utilities, which had traditionally been natural allies in support of public-interested investments in broad-based regional growth, have been impacted by the new technology in ways that have increasingly cut their self-interested links with poorer communities. All of this has contributed to the paradox of vibrant regional politics that matter most to multinational companies, while local communities are increasingly disempowered by the technology at the local level.

----------------------------------------------

Endnotes for Chapter 4

1 The Joint Venture Way: Lessons for Regional Rejuvenation. Report produced by Joint Venture: Silicon Valley Network. San Jose. 1995.

2 See Bright, Julian. "The smart city: Communications utopia or future reality?" Telecommunications (International Edition) v29, n9 (Sep 1995):175-181. A NAME="3">3 Leong, Kathy Chin. "Siliconnected valley. (Silicon Valley companies form Smart Valley Inc. non-profit to build giant information network)" InformationWeek, n438 (August 16, 1993):22.

4 Joint Venture: Silicon Valley. An Economy at Risk: The Phase I Diagnostic Report. Report prepared by the Center for Economic Competitiveness, SRI International. 1992.

5 Schnaidt, Patricia. "Cruising along the super i-way: cyber-pioneers look to the information superhighway to carry education, health care, and commerce applications. " (Interoperability: Practical Solutions to Complex Networks special... LAN Magazine v9, n5 (May, 1994):S8 (7 pages).

6 "CommerceNet moves east." Telecommunications (Americas Edition) v29, n2 (Feb 1995):12 and Brown, Jim. "More public/private sharing in Mass." Network World v12, n20 (May 15, 1995):45.

7 Greco, Monica. "AMTEX bears first fruit." Apparel Industry Magazine v56, n4 (Apr 1995):32-38.

8 Messmer, Ellen. "Ford testing a 'private Internet' to link suppliers." Network World v12, n11 (Mar 13, 1995):14 and Messmer, Ellen. "Automakers Drive Private IP Network." Network World v13, n40 (Sep 30, 1996):33.

9 Interview with Michael McRay, June 5, 1995.

10 Bank, David. "Smart Valley to get $8 million; goal of U.S. grant is to help turn Internet into electronic marketplace." San Jose Mercury News (Nov 24, 1993):8D.

11 Interview with Tom Masotto, June 13, 1995.

12 McRay Interview, Ibid.

13 Randy Whiting interview, July 21, 1995.

14 Karen Greenwood interview, September 28, 1995.

15 Paul Kutler interview, November 15, 1995.

16 Bill Davidow interview, November 11, 1995.

17 Mack Hicks interview, June 22, 1995.

18 Whiting, Ibid.

19 Mason Myers interview, December 9, 1995.

20 Joint Venture's Index of Silicon Valley 1997: Measuring Progress Towards a 21st Century Community. Prepared by Collaborative Economics for Joint Venture: Silicon Valley. 1997.

21 Kathy Blankenship interview, September 28, 1995.

22 Eric Benhamou interview, July 21st, 1995.

23 Joint Venture's Index of Silicon Valley (1995, 1996, 1997).

24 Bruce Lowenthal interview, July 21, 1995.

25 Sinton, Peter. "Businesses Wired on Intranets." San Francisco Chronicle. September 26, 1996.

26 Bar, Francois. Configuring the Telecommunications Infrastructure for the Computer Age: The Economics fo Network Control. Dissertation. City and Regional Planning at University of California. 1990.

27 Garten, Jeffrey. "Why the Global Economy is Here to Stay." Business Week. March 12, 1998.

28 Mark Masotto interview, June 13, 1995.

29 Whiting interview, Ibid.

30 Lowenthal interview, Ibid.

31 Brandel, Mary. "On-line catalogs are booting up." Computerworld, Electronic Commerce Journal Supplement. Apr 29, 1996:5. and "AMP unit targets online commerce." Electronic Engineering Times, n919. Sep 16, 1996:24.

32 "Boring, boring, boring: Business-to-business e-commerce is a revolution in a ball valve." In The Economist's "In Search of the Perfect Market: A Survey of Electronic Commerce." May 10, 1997.

33 Hicks interview, Ibid.

34 McRay interview, Ibid.

35 Krantz, Michael. "Doing Business on the Net is Hard Because the Underlying Software is so Dumb. XML Will Fix That." Time. November 10, 1997.

36 Karen Greenwood, Smart Valley Project Manager for BAMTA, September 28, 1995.

37 Harrison, p. 23.

38 Lowenthal interview, Ibid.

39 Myers interview, Ibid.

40 Kutler interview, Ibid.

41 Much of the history of MCC in this chapter comes from Gibson, David and Everett Rogers. R&D Collaboration on Trial- The Microelectronics and Computer Technology Corporation. Harvard Business School Press. Boston. 1994. They also give a good overview of the role of consortia in modern electronics. See also Burrows, Peter. "Craig Fields's not-so-excellent adventure." Business Week, n3362, Industrial/Technology Edition (Mar 14, 1994):32; Port, Otis; Burrows, Peter. "R&D, with a reality check." Business Week, n3355 (Jan 24, 1994):62-65; "MCC Digs Out of the 'Celestial Sandbox of Computer Science'". Electronic Business v18, n8 (May 18, 1992):67-68; Carey, John; Bartimo, Jim. "If You Control . . . Computers, You Control the World". Business Week, n3170, Industrial/Technology Edition (Jul 23, 1990):31. Verhovek, Sam Howe. "Austin Rides a Winner: Technology." New York Times. January 31, 1998.

42 An Economy at Risk, p. 51-56.

43 The Joint Venture Way: Lessons for Regional Rejuvenation. Report produced by Joint Venture: Silicon Valley Network. San Jose. 1995, p. i18.

44 Birmingham, Stephen. California Rich: The lives, the times, the scandals and the fortunes of the men & women who made & kept California's wealth. New York. Simon and Schuster. 1980.

45 Saxenian, Annalee, 1994, Ibid.

46 Saxenian, AnnaLee. "Contrasting Patterns of Business Organization in Silicon Valley." Working Paper of the Institute of Urban and Regional Development, University of California at Berkeley, #535. April 1991

47 "Chip Makers, U.S. Announce Venture." The Associated Press. September 11, 1997.

48 Florida & Kenney, p. 105.

49 Gibson and Rogers, p. 520.

50 Greenwood interview, Ibid.

51 Craig fields in Gibson and Rogers, p. 215.

52 Gibson, David and Everett Rogers, Ibid.

53 Davidow interview, Ibid.

54 McRay interview, Ibi.

55 Eran Gross interview, November 15, 1995.

56 Lowenthal interview, Ibi.

57 Greenwood interview, Ibi.

58 See Oliver Williamson. The Economic Institutions of Capitalism: Firms, Markets Relational Contracting for a full view of how "transaction" costs of relationships create hierarchies in the economy.

59 Whiting interview, Ibid.

60 Whiting, Ibid.

61 Noer, Michael. "Stillborn." Forbes. March 19th, 1998.

62 McGarvey, Joe. "Intranets, NT Shape Server Market." SoftBase. Jun 1, 1997.

63 Vasilash, Gary S; Bergstrom, Robin Yale. "The five hottest manufacturers in Silicon Valley: Customer obsession at Solectron." Production v107, n5 (May 1995):56-58.

64 Kornblum, Janet. "Netscape sets source code free." News.Com. March 31, 1998.

65 Moeller, Michael. "Fort Apache: freeware's spirit outshines commercial products." PC Week v14, n23. June 9, 1997.

66 Glyn Moody. "The Greatest OS That (N)ever Was." Wired. August 1997.

67 Sullivan, Eamonn. "Freedom is priceless, even when it's free." PC Week v13, n47. Nov 25, 1996.

68 Gordon, R. and L. Kimball, 1987, p. 168.

69 Davidow interview, Ibid.

70 "Netting New York: http://manhattan." The Economist. May 25, 1996. and "Silicon, Silicon Everywhere." From Future Perfect? A Survey of Silicon Vally in The Economist. March 29, 1997.

71 Markoff, John . "Gold Rush From Software Animates Silicon Valley." New York Times. January 13, 1997.

72 Johnson, Marilynn S. The Second Gold Rush: Oakland and the East Bay in World War II. Berkeley. University of California Press. 1993.

73 Joint Venture's Index of Silicon Valley 1997: Measuring Progress Towards a 21st Century Community. Prepared by Collaborative Economics for Joint Venture: Silicon Valley. 1997.

74 Simon, Mark. "Down Side Of Peninsula Job Boom." San Francisco Chronicle. March 24, 1997.

75 Hof, Robert. "Too Much of a Good Thing?" Business Week. August 25, 1997.

76 Joint Venture's Index of Silicon Valley 1996: Measuring Progress Towards a 21st Century Community. Prepared by Collaborative Economics for Joint Venture: Silicon Valley. 1996.

77 Freedberg, Louis. "Race Panel Gets an Earful In San Jose; Talks dominated by hecklers, complaints about wage gap." San Francisco Chronicle. February 11, 1998.

78 Ward, Leah Beth. "Tax Rules Squeezing Independent Contractors." New York Times. November 10, 1996.

79 Shock Absorbers in the Flexible Economy: The Rise of Contingent Labor in Silicon Valley. Report prepared by Working Partnership USA. 1995.

80 Hicks interview, Ibid.

81 Barlett, Donald L. and James B. Steele. "Say goodbye to high-tech jobs... many are moving offshore." The Philadelhia Inquirer series. 1996. and "Silicon, Silicon Everywhere." The Economist. 1997, Ibid.

82 Markoff, Jan 1997 and Uchitelle, Louis. "Service Sector Guru Now Having Second Thoughts." New York TImes. May 8, 1996.

83 Davidow interview, Ibid.

84 Fimrite, Peter. "Altamont Rail Plan on Track: A Central Valley-South Bay link." San Francisco Chronicle. December 2, 1996.

85 Interview with Kathy Blankenship, 9-28-95.

86 Schnaidt, Patricia. " Cruising along the super i-way: cyber-pioneers look to the information superhighway to carry education, health care, and commerce applications." (Interoperability: Practical Solutions to Complex Networks special... LAN Magazine v9, n5 (May, 1994):S8 (7 pages).

87 Blankenship interview.

88 Castells, Manuel. The Informational City: Information Technology, Economoc Restructuring and the Urban-Regional Process. (Oxford: Basil Blackwell. 1989).

89 Joint Venture Way: Lessons for Regional Rejuvenation, 1995. i.13.

90 Pimentel, Benjamin. "Congested Silicon Valley Slow to Board Light Rail." San Francisco Chronicle. February 24, 1997.

91 Gaura, Maria Alicia ad Marshall Wilson. "Silicon Valley Commuters Ask for More Transit: `Unlock the Gridlock' forum in San Jose " San Francisco Chronicle. March 21, 1997.

92 Fimrite, Peter. "Altamont Rail Route Gets Boost. Alameda County panel OKs funds for commuter line." San Francisco Chronicle. December 21, 1996.

93 Bowman, Catherine. "Bus Riders Union Won't Stay Seated. AC Transit passengers want a say in service." San Francisco Chronicle. March 31, 1997.

94 "The changing dream." From Future Perfect? A Survey of Silicon Vally in The Economist. March 29, 1997.

95 Lucas, Greg. "Prop. 211 Loses By Wide Margin." San Francisco Chronicle . November 6, 1996 and Simon, Mark."Prop. 211 Foes Share Riches With Other Campaigns: $1.75 million in past week given to anti-217 forces." San Francisco Chronicle. November 1, 1996.

96 Blackford, Ibid. 295