Another interesting statement from the recent CITI forum, was UWC custodian of IT, Professor Derek Keats' observation regarding an often overlooked aspect worth considering when implementing a technology, the cost of replacing the technology with a competitor, or as he put it - the cost of exit.

...So, my question... to everybody out there, what should the business model then be? If we are not going to start taking Linux out of the... out of the little developer sitting at home doing some work, how do we get it into the corporate environment? The corporates are not going Linux because they dont have a single throat to choke, and that is the expereince that we have.in the market today.

Well, I'm not sure I'm the best person to answer this, I think Stafford would do a much better job seeing as how he's using... seeing as how, basically, Novell is trying to be the single throat to choke, and that is one of the ways in which Linux... GNU/Linux is penetrating the corporate environment.

I think that, unless we have big players like Novell in this space, you're always going to find that is the situation. It's...It's... Even in my own institution, as I've said - we are an enterprise customer as well as an R&D organization - and on the enterprise side, even my own techies, I have difficulty convincing them that they should be using Linux and one fo the reasons we have actually... we are in fact still a customer of Novell is because we do have that... capability - we know that if we go back to them and something is broken, there is going to be considerable pressure for it to be fixed, and if it isn't fixed that we've got someone we can hold accountable, so y'know I think that is what Novell is trying to do, they're not the only player in that space by any stretch of the imagination, and one of the beauties about Free and Open Source Software in the enterprise, and one that we seldom hear the proprietary vendors talk about, is the relatively low cost of exit.

So if, for example, tomorrow I wanted to pull out... say SUSE and replace it with another distribution of Linux, I could do that on 100 servers over a few days at no cost, that is the beauty of it and I think one of the things... one of the selling points that companies like yours need to get across to your customers as well, is that the total cost of exit is also a major factor when it comes to implementing technology.

Professor Keats also spoke about how UWC has moved their datacenter from almost 100% proprietary platforms to nearly 90% non-proprietary, and how heterogeneity allows UWC to choose the best solution for the task, including taking into account when "assurances" are necessary, and when "risk" is acceptable.

I think that one has to recognize that there is heterogeneity in the Free and Open Source Software space. and there's times when you'll want to take one approach, and there're times when you'll want to take another approach.

So, if I look at UWC, for example, we were, in our datacenter almost 100% proprietary platforms, I would say that we are now pretty close to 90% not proprietary, and we've been able to make that change because those people that hold us financially accountable, we've been able to give them assurances and so we need a partner like Microsoft, like... Microsoft?(a misstatement)... like Novell(intended to say), like MP, like Red Hat, someone who we can say to our funders, the... the government of South Africa and the students, and those who have fiduciary responsibilities to the... for the continued existence of the university, that we've got some assurances, that we're not putting the institution at risk.

I think its important to understand that in that respect, we are a typical enterprise customer, but there are spaces where we can live with more risk.

So, for example in our student labs; so, in our student labs we don't need the assurances that we can get from an enterprise partner; so we can deploy things like OpenSolaris, which we have deployed, another... another Free and Open SOurce Software platform that we haven't actually spoken about.

and, on the desktop, we also deploy Ubuntu, so our choices are based on practical realities of the situation in which we live, so I think within the provincial government I think the same thing can apply, there are places, there are spaces where you have some fiduciary responsibility where you have to be... you have to have another level of accountability... but there are other spaces where you can deploy other technologies, and I think its important for us to understand that we are dealing with a heterogeneous environment, we don't have to be just one size fits all.