Remembering Ed Iacobucci, thin client pioneer
In September 1998, I attended the first customer and partner conference of Citrix Systems, whose thin client technology was about to spawn an entire industry of application service providers (ASPs), the precursors of today’s cloud application vendors. That visit crystallized my own decision to found a website venture called ASP News Review to track and report on this new industry, a site that I built (using cloud services) and launched a month later.
While in Orlando (my travel for the event was funded by Citrix) I interviewed Iacobucci about his journey to become the father of thin client computing. Below is the copy that I filed for publication in the UK trade publication MicroScope that month. It’s a snapshot of how he saw his career at the point when everything he had planned for was coming together, and a reminder of how his vision lives on in the cloud we know today.
The short, stocky, bearded figure of Ed Iacobucci is not the sort of presence that turns heads on entering a room. He has a calm, thoughtful manner, one that might even strike the uninformed observer as dull.
The outward appearance belies the inner man. Iacobucci (pronounced Yakkaboocey) is no lightweight, and his influence over the computer industry is growing fast. Beneath the steady exterior there’s a rock solid lodestone of inner certainty. For nearly a decade, it has kept him obstinately committed to his chosen path. Now, it seems, destiny is ready to reward his patience.
As September kicked off, Iacobucci presided over Thinergy, the first conference organised to bring together the resellers, partners and users of the company he founded in 1989, Citrix Systems. Billed as the birthplace of a new industry dedicated to ‘thin-client/server computing’, it drew almost 2,500 attendees to a hotel complex in Disneyworld, Florida.
For three days, they discussed multiuser Windows in all its forms. Most topical was the recently-launched Terminal Server Edition (TSE) of Microsoft Windows NT 4.0, which incorporates the Multiwin kernel Citrix originally developed for its own version of NT, Winframe. Sharing the limelight were its Metaframe add-on for TSE and the ICA protocol that supports thin client terminals.
The event was a coming of age for Citrix and its works, which have become a popular means of giving terminals and low-spec PCs access to Windows applications running on a server. The $180m (£108m @ $1.67/£) turnover company has been achieving a year-on-year growth rate well over 100% this year, and already claims to be the eighteenth largest US software company. With a user base now estimated at eight million, ICA has surpassed its Unix rival, X, to become the most popular terminal protocol for GUI-based applications.
Speakers were calling Thinergy a seminal, landmark event. Yet Iacobucci, who is chairman and CTO of Citrix, began his opening keynote speech with a characteristically self-effacing statement. “Thin client-server computing is not anything new in terms of system architecture,” he told delegates. To him, it’s something that has always been obvious right from the beginning. That its time has finally come simply fulfils the vision that he’s doggedly pursued for almost a decade.
Enterprise to desktop
For eleven years before he founded Citrix Systems, Iacobucci’s career had taken him on a journey through the inner cores first of enterprise computing, and then desktop computing. In the early eighties, he spent five years at IBM’s software research laboratories in North Carolina, developing early network management products that would later form the core of IBM’s Netview system management software. In his spare time, he developed software on contract for the PC division during its early days. “That was an education in PC software,” he recalls.
He then moved down to the PC facility at Boca Raton in Florida. As one of the most experienced software developers on the site, Iacobucci was put in charge of the joint Microsoft-IBM team set up to develop a multi-tasking PC operating system. That project ultimately became OS/2, and also laid much of the ground for what we now know as Windows.
“I was very fortunate to have been able to spend my career on two sides of the fence,” he says. First of all it introduced him to all of the keenest issues IT managers face at the enterprise level. Then it gave him the opportunity to learn all of the issues that arise running applications in the PC environment.
He quickly saw that there was a serious flaw in the “information at your fingertips” model of client-server computing in which every application ran at the desktop. “I had seen how hard it was to do serious client server applications inside the middle of IBM in the 1980s with probably the best qualified team of system administrators I could ever think of,” he says. He felt it was not realistic to put application code on every user’s desktop and expect the average enterprise to be able to manage it.
There were other shortcomings too. “Having had the mainframe experience as part of my background, I didn’t even think twice about things like being able to move from one workstation to another and continuing everything I was doing,” he says. “Those ideas were part of me going into the PC world, so I felt that they were missing.”
Of course, many in IBM at the time disparaged the PC, but unlike them, Iacobucci also appreciated the benefits of desktop computing. A born conciliator, he sought a compromise that could accommodate both sides. “The key to computing in the enterprise I thought was balance. There was a need for some users to have a more centrally managed option,” he explains.
Having decided that such a compromise solution could exist, Iacobucci founded Citrix in order to create it. There followed several frustrating years. The first Citrix product was a multiuser system for OS/2. But disaster struck just as it was ready to ship. “IBM and Microsoft went to war right as we were releasing the first product. We had to write off about eighteen months there,” he says.
Many would have given up there and then, but Iacobucci persisted. “Personally I really believed in the vision of the problems that the IT managers would have [with client-server]. It was obvious to me that there was going to be a strong desire for this from a market standpoint. Some people say I’m too hard-headed but in this particular case I think that there was some merit to it.”
A remote access product for Windows began to show market promise, and then in 1995 Citrix released Winframe, its multiuser version of Windows NT. Coinciding with the beginning of Oracle- and Sun-inspired hype about network computing, it was an instant, if minor, hit.
It was still successful enough to arouse Microsoft’s interest. Early in 1997, the software giant announced it would produce future versions of multiuser Windows itself. Many believed that was the end for Citrix, but they had not reckoned with Iacobucci’s tenacity and experience.
“Microsoft was just learning about the marketplace, and we’d been in it for seven years,” says Iacobucci. “It’s a big difference. And the issues are different.” Citrix knew what was involved, and it knew what it wanted. Negotiations took almost three months, but the experience proved the company’s mettle. “We had our fastest growth, we signed our biggest customers, we had our lowest turnover rate in employees, all there in that same eleven weeks,” says Iacobucci. And the deal itself? “Actually it wasn’t a very bad deal at all – it was $175m for [Microsoft] to relicense its own product.”
Citrix is the first small developer to cross swords with Microsoft and fare so well. But Iacobucci benefitted from having worked alongside some of its top people a decade before on OS/2. “I don’t claim to be the world’s foremost expert on Microsoft. But I do know where they learnt about things like FUD, because I think in those days, we – IBM – taught them,” he says. “I think I understand from interactions and from watching them in the industry pretty well how they tend to react to problems.”
Some observers say that Microsoft still wants to undermine multiuser Windows, citing as evidence its insistence on charging per-seat licence fees. Typically, Iacobucci gives it the benefit of the doubt. “I do not think that it’s an intentional act on its part. I think it’s the ramifications of changing things like basic core licensing policies on the much bigger [traditional applications] business.”
While Microsoft’s licensing policy was a thorny issue among Thinergy delegates, the Citrix top brass were more concerned with setting its newly-formed industry on the right path. While the architecture may be nothing new, Iacobucci believes Citrix breaks new ground by making it possible to deploy applications beyond their native platform: “A freedom of choice architecture. And the notion that you really can do server-based computing for graphical applications.”
What is really winning the confidence of enterprise users, he adds, is the way the Citrix architecture eases the deployment and management of applications. “It broadens the appeal of the applications, it creates the opportunity for new markets.” The largest opportunity, he believes, is in the information appliance market. “We think that with the new thinking that this fosters – it’s going to create a lot of natural opportunities – we don’t even know what they all are yet.”
As those opportunities take shape, the company will have the role of setting and enhancing standards. But Iacobucci dislikes the suggestion Citrix could become the Microsoft of thin client computing. “Bad analogy,” he says. “We believe that we’re part of the industry, but not the industry itself.”
Citrix takes a consciously inclusive approach. “I think that people don’t like to be dictated to. They like a healthy level of participation,” he says. “But I think that there is an absolute need for leadership,” he goes on. And the quiet assurance that he and Citrix are ready to fulfil that role is unmistakeable.
Photo credits: Courtesy of VirtualWorks