Monthly Archives: October 2005

Technology and Human Resources

New technologies offer more and better possibilities and allow us to do things online that we have not been able to do before. Greater connection speeds, processor power, screen resolution, etc. benefit us and support a greater number and more complex activities. Skype and internet telephony, VoIP, IPTV, dynamic database driven web pages offer a vast range of new uses.

This is all brilliant and very welcome by the world-wide web community. However, from an institutional or organizational point of view there is one real downside, and it’s growing: support services that eat up human or system resources. In all this development frenzy one paradox stands out:

*old technology does not go away*.

Once a tool is introduced to an organization there is an inherent resilience that prevents or at least hinders replacement. There are several factors to this:

1) the human factor:
Introduce your staff to a new tool. They like it, adopt it and integrate it into their daily working process. Something better comes along, but replacing the old tool is near impossible without causing great upset and legacy maintenance. A good example of this is MS Internet Explorer. As a browser it has not developed since version 6. But would it be an easy thing for an IT department to decide to rid the institution of IE and use Firefox or Opera instead – no, not easy.

2) the system factor:
Integrated systems are the buzzword of the day. New tools are evaluated by their interoperability. Once integrated it is not a simple matter to dis-integrate and replace it with what maybe a better newer system and achieve the same level of integration. Try e.g. to replace your library system or e-mail system, will the calendar function still work in the same integrated way?

3) the proprietary factor:
Institutions buy into a number of commercial systems, e.g. Microsoft servers, Office, Novell Netware, Blackboard, and so forth. These are deliberate decisions at the time maybe following rigid procurement procedures. It is a well-known experience that replacing them is either a major upheaval for the organisation or simply impossible. A recent experience on my part was the decision to no longer support Active Server Pages (.asp). Suddenly we noticed how many MS Access databases connected to asp pages on our webservers. Some of them we were contractually bound to hosting them.

What this all boils down to is that newer technologies will normally be introduced into the institution as an additional application, not as a replacement for something outdated. This additionality comes with extra service provision (authentication, maintenance, back-up, etc.) that are tightly connected to human resources available.

Cut a long story short: We need to get much better in getting rid of old technologies and replacing them with new services. This would allow savings of service time provision storage space, server space, bandwidth, but most of all human maintenance.

Interoperability Risks and Choices

Gayle Calverley in a recent article for the JIME online journal made the observation that institutions might selectively and knowingly choose vendor lock-in as a trade off to certain benefits. She argues that interoperability might be a phantom of a potential undefined future change that may not even occur or negatively affect the institution, which is why many put up with proprietary systems.

An important note she makes is that there is a widespread misconception that open source (OS) is equivalent to non-proprietary and prevents lock-in. Far from being so she questions whether interoperability in reality has sufficient value to institutions to amortize costs going into implementing it for reasons of avoiding lock-in.

My observations stimulated by this debate go into several directions:
1) It seems that institutions don’t see their assets as being valuable enough to pro-actively prevent lock-in where this is possible (in too many areas it isn’t). Disposable instead of reusable is the design that is given preference despite all the lip service to the contrary. Proprietary de-facto standards are given more importance than the promoted specifications and standards that most decision makers don’t understand.

2) Changing institutional systems is difficult. Once a certain application is made available and fully implemented (including training and support systems) it is near impossible to change or take away. Even when better systems become available there is a surprising resilience of legacy applications (as portrayed in the Matrix!). Sadly even the so-called “standards” have to run on proprietary operating systems.

3) Market forces have shown in the past that lock-in is overcome by competition not by ad-hoc round-tables that produce arbitrary sets of standards. Format conversion appears when a competitor wants to increase their market share by eating into their competitors user base.

Business risks of lock-in for institutional content: little to loose = low risk of loosing out, or are they wrong?