By Andre Griffith
Some years ago I attempted to ‘sell’ a senior executive the idea of adopting “thin client computing,” He was not keen on the idea. “It never really caught on in North America,” was the justification for his skepticism………not, perhaps, exactly the degree of analytical effort that you would want to see in evidence in a decision-making process. The fact is, however, that the lack of take-up of “thin computing” in the mid to late nineties and in the early years of the twentieth century was due to several factors. One very significant factor was the actual culture of desktop computing which placed a great deal of power in the hands of users. ‘Thin computing’ took almost all of that power away and encountered strong resistance. By comparison, we in Guyana were still largely in the era of “green screens” for most big operations, and for a lot of others, the actual ratio of computers to workers was low, such that the hurdle of user resistance would have been much lower.
Another factor was the “irrational exuberance” – to borrow a term from the former Chairman of the United States Federal Reserve Bank – that existed in the late nineties into the turn of the century, during the dotcom era, when there was a frenzy of spending on Information Technology and a lot of “creative” (to be charitable) concepts about valuing information technology and information technology companies.
At the more extreme level, there was the infamous “number of clicks” metric which some dotcom companies used in their prospectuses, having no revenue – much less profits -to show to potential investors.
The euphoria did not escape traditional companies which were exhorted to invest in IT or perish. One particular myth that was being pedaled was that companies needed to be spending 6-7 percent of their revenues on IT.
Driven by this empty benchmark spending argument, companies splurged on IT mega projects, many with spectacular failure. When, inevitably, gravitational forces regained control, most of the dotcoms were history and many mainstream companies were left wiser but lighter in the pocket and a few sobering trends started to take over.
One trend was to start to subject potential IT to the same investment appraisal process that applied to other ventures in the company. As a natural consequence of this approach was an emphasis on so-called “total cost of Ownership” or (TCO).
The TCO concept looks at the cost of a total IT installation over its expected life. It not only looks at the capital cost in terms of software implementations, but also considered the associated costs of hardware and human and other resources needed to keep a system going over its total lifetime. It is with the emphasis on TCO that ‘thin computing’ began to rise, as companies started to critically examine and cut back on IT budgets.
The ‘thin computing’ concept applies mainly to two characteristics of the system. The first characteristic is that ‘thin’ client computer on the user’s desk is nothing more than a device to receive user input and to display system output. No processing is done on the users workstation; rather, all information is sent to a central server (or bank of servers) for processing.
The second aspect of the “thin” system was the network traffic. The amount of data sent to and from the users work station is minimized by various means so that the network requirements are reduced. Thin computing is not a new concept. The first generation of multi-user computer systems were in fact thin client systems.
To see that the characteristics of the ‘thin’ client computing were tailor-made for Guyana, we only have to look at some of the defining features in our environment. Bandwidth is limited and expensive, skilled human resources are in scarce supply (in fact in much shorter supply than IT people think), capital is limited and cost of capital is high. On each of these counts thin computing comes out ahead.
With respect to bandwidth, ‘thin computing’ allows you to function over low-speed networks which obviously cost much less than higher speed networks that would be required to run applications in their traditional “fat“ mode. In fact the current generation of “thin’ systems” were developed primarily as a response to bandwidth limitations.
With respect to financial resources, while the higher end ‘thin’ client hardware typically costs as much as a traditional desktop computer, this is probably the hardware makers’ attempt to squeeze as much margin as possible out- -of – sales since they know that ‘thin’ client computers beat traditional desktops on TCO, hands down.
In any case, there really is no need to buy the higher end systems since many of them, in my opinion, defeat the purpose of ‘thin computing,’ by putting back a number of features in the hands of the user which is undesirable.
Again, with respect to hardware, the useful life of a thin client computer is at least twice that of a traditional desktop which apart from inherent savings on hardware, goes a long way to get you off of the treadmill of desktop operating system upgrades. Desktop operating systems upgrades are not only expensive in themselves but could be far more so if they mean that you have to also invest to upgrade your business applications to work on the latest gem put out by Gates and Co.
With respect to the lack of skilled human resources, this is where ‘thin’ client computing really comes up trumps. A user’s machine can be a significant source of problems on your network since despite improvements in operating system security, one of the fundamental philosophies of desktop computing was putting power that previously existed on the mainframe in the hands of users.
Like everything else in life that power can be used for better or for worse and many IT managers would probably agree that it is, most times, for the worse. Many problems in your networked environment are caused by users (or their sub-skilled administrator friends who have privileged access) ‘messing’ with configurations. Additionally, the large number of devices and features available on a typical desktop simply provides more opportunity for getting something ‘screwed up.’
The ‘stripped down’ nature of the ‘thin’ client work station automatically reduces the number of problems that occur on a users machine and support personnel consequently need to make fewer site visits thus you can probably reduce the number of administrators that you need to cover all of your sites.
To be clear, the real advantage in our local context, is that you can now afford to hire one person who really knows what he or she is doing, rather than three or four that are feeling their way.
Apart from these primary benefits, thin client systems also offer other advantages that are valuable in their own right among them being the increased security that you are able to achieve by controlling what goes where in your systems. From a security standpoint, some companies have used these systems to guard against theft of intellectual property.
‘Thin computing’ is rightfully on the rise in Guyana. At least five state and para-statal organisations have invested in them in the recent past. The real kudos however go to the local private sector companies that invested in these systems close to a decade ago – long before they ‘caught on’ in North America”.
