At the end of each evolution in enterprise software, one hears rumors of the death of enterprise software. One hears that slow growth, low valuations, and consolidation will kill innovation and venture capital investment in enterprise software. Accompanying the rumors are truckloads of data about venture capital fleeing the sector. Enterprise software has gone through long periods of fantastic growth followed by shakeouts and consolidation as new computer platforms and their attendant new software paradigms emerge and gradually replace the old regime. Venture capital usually slows to a trickle during these periods of shakeouts and consolidation when weak enterprise software companies sell to stronger companies at valuations that embarrass even the most stoic venture capitalist.

Gordon Bell, an early Digital Equipment Corporation engineer and industry pioneer in 1972 described -- in what has become known as Bell's Law -- the computer industry and the enterprise software industry that accompanies it. He said that technology advances in semiconductors, storage, interfaces and networks bring about a new computer platform approximately every ten years. The new platform offers not just better functionality in new areas, but is also usually lower priced than the incumbent platform. Platforms (Bell called them classes) of the past include: mainframes (1960s), minicomputers (1970s), networked workstations and personal computers (1980s), browser-web-server structure (1990s), web services (2000s), and mobile and cell convergence with computers (2003). Bell predicts home and body area networks will form by 2010.

What does this mean for enterprise software? Today, a new wave of enterprise software industry expansion is pulsating through the economy, and the trickle of venture capital flowing into the industry today will soon turn into a spring freshet. Those who subscribe to Bell's Law would describe the industry change as a new platform, marked principally by Saas and cloud computing. The industry visionaries are referring to this new era of enterprise software as Enterprise 3.0, and new opportunities in enterprise software abound.

A brief review the evolution of enterprise software will put today's optimism about the enterprise software industry into perspective. Enterprise 1.0 occurred during the great mainframe expansion that began in the early 1950's and ran until the minicomputer revolution in the early to mid 1970's. Enterprise 1.0 was characterized by "Big Iron" mainframe computers with a few thousand dedicated connections to the machine, and once-a-day batch processing. IBM was dominant in this field with several other players that together were referred to as "IBM and the Seven Dwarfs." The dwarfs were Burroughs, UNIVAC, NCR, Control Data, Honeywell, GE and RCA, later after mergers referred to as "IBM and the BUNCH" (Burroughs, UNIVAC, NCR, Control Data, and Honeywell). Operating and application software were initially written in-house by programmers dedicated to their mainframe systems, until mainframe adoption spread into most large enterprises in the late 1950's and early 1960's.

At that time, independent software companies emerged to write specific applications. The Computer Usage Corporation (CUC) was founded by two former IBM employees in 1955, and by 1967, CUC had 700 employees in 12 cities. The Systems Development Corporation (SDC), a division of the RAND Corporation, was formed in 1956 to develop a large air defense system. SDC employed hundreds of programmers and was referred to as "programmer university". The Computer Sciences Corporation was formed with five founders in 1969, and had 68,000 employees by 1990. Most of these early entrants into independent software development were formed by programmers writing custom programs for individual customers. By the mid-1960's, however, independent software companies began developing and marketing software packages that could be used by many different types of customers. One innovator in this era was Informatics, which wrote and sold the hugely popular Mark IV database in the mid-1960's.

Although minicomputers were developed in the 1960's, their widespread adoption in the 1970's marked the flourishing of Enterprise 1.0. Digital Equipment Company, formed in 1964, was the first successful minicomputer maker, but other companies along Massachusetts route 128 joined in the growth of the minicomputer market during the 1970's: Data General, Wang Laboratories, Apollo Computer, and Prime Computer. Tracy Kidder won a Pulitzer prize for his non-fiction book The Soul of a New Machine, detailing the development of Data General's minicomputer. In 1984, there were 91 minicomputer companies in the United States. By 1990, there were less than 10.

The last hurrah of Enterprise 1.0 was the flourishing of software companies developing products for the minicomputer market. Some names were American Software (1970), Tesseract Systems (1970), Walker Interactive Software (1971), ASK Computer and Ross Systems (1972), Compuware (1973), Cyborg Systems (1974), Computer Associates and SAS Institute (both 1976), and Candle Corporation, J.D. Edwards, Oracle Corporation, Softool (all formed in 1977). The most successful enterprise software at the time was Computer Associates, which acquired dozens of software product companies. A well-known pioneer during this period was John Cullinane who in 1968 founded Cullinane Software, which was the first software product company to go public, in 1978.

The PC platform was the death knell for minicomputers as client-server architecture took over the enterprise in the early to mid-1980's, heralding Enterprise 2.0. Enterprise 2.0 was marked by data continuously available and updated, millions of connections to the network rather than mere thousands, and data available from the network almost anywhere, rather than just at a terminal connected to a mainframe or minicomputer. The client-server architecture required entirely new software at the system level, management level, and at the client level. With decentralization and distribution, the advent of networks, and Marc Andreesen's Mosaic user interface to the Internet (later commercialized at Netscape), Enterprise 2.0 was at its height, and another flourishing of enterprise software companies took place. There were not only Netscape, Microsoft, Oracle, Peoplesoft, Sybase, Informix, Platinum Technology, BMC, BEA, and Red Brick, but also Arbor, Aurum, Broadvision, Scopus, Simware, Sun's Java platform, and hundreds of other companies offering platforms, management software (e.g. Remedy's helpdesk software), security software, enterprise applications, and of course even client-level applications. The industry consolidated again in fits are starts over the next decade, accelerated by the recessions in 1990-91, the mild slowdown in 1994-95, and the tech bust of 2000-2002.

Today, the software industry is at the threshold of Enterprise 3.0, where data is continuously updated and available all the time from multiple devices anywhere in the world, with billions of connections to systems and users through online networks that are not tethered to a specific enterprise's system. Saas and Cloud computing are part of Enterprise 3.0, and cloud vendors are capitalizing on the infrastructure needs of the new paradigm. Enterprise 3.0 is characterized by vendors solving highly specific problems and providing highly customized solutions for customers by bringing together just the resources needed for that solution, and doing so on a model where almost all the infrastructure and development are outsourced in one form or another. The hosting of the data may be outsourced to a hosting company, the software development might be outsourced to a development team, other software might be purchased on a Saas model, and storage might be purchased on a terabyte basis from a cloud vendor. The ability to collaborate and affiliate easily are central to effecting these solutions for customers and making money in Enterprise 3.0.

There is a flourishing of SaaS, cloud, and infrastructure vendors filling market spaces in new Enterprise 3.0 sectors. Companies like Cast Iron Systems, Cloudera, Corticon Technologies, Gridapp Systems, Instantis, Kace, Marketo, Mobclix, Nirvanix, SOA Software, SmartVault, Vkernel, Wize, and Zetta are a few of the companies penetrating the multitude of new segments opened up by the Enterprise 3.0 paradigm. Many of these companies have received venture funding, and the venture capital community is once again very interested in enterprise software, albeit looking for specific niches that each venture group perceives as potentially high growth. Fortunately, Enterprise 3.0 offers a superabundance of these niches for investment capital. Accordingly, venture capital is alive and well in the new enterprise software market, and that is reason for optimism in the enterprise software industry.

Thomas C. Klein is a Shareholder of Greenberg Traurig, LLP. Email Thomas at kleint@gtlaw.com.