I had a very interesting discussion with a colleague today about where IT is heading. She indicated that following a period of centralisation and the move to shared services, we are seeing the beginnings of a shift to some decentralisation. She suggested that the motives of IT in this shift were not always honourable (libraries are too hard, too demanding, too informed, too expensive to support , a damned nuisance etc), but that nevertheless there are opportunities for libraries in this move.
All the good reasons for centralisation in a corporate environment (development of standard environments, reduction in costs, sharing of scarce and expensive expertise, economies of scale, single point of contact etc) can actively work against the ability of the library to provide the services it has a mandate to provide to its customers. It can certainly limit the ability of the library to experiment and ‘play’ with the emerging technologies to determine if and how they can be of benefit.
The core premise behind the development of an SOE is to reduce variation so as to allow central management of a fleet of computers, either desktop or server. There are some very sound and sensible reasons for doing this
- cost – if everything is the same, deployments and support can be automated to a large degree and less expensive people hours are required.
- equity – in a standard environment, everyone has the same and so no group gets left behind because of insufficient priority given to replacement cycles etc.
- economy of scale – equipment and software can be purchased at discount when quantites are larger.
- staff resources can be deployed more effectively, and with less skill sets required, the available money doesn’t have to be spread as thinly across multiple skill sets and depth of skill can be purchased instead.
But if the adopted server platform is Windows, and the preferred platform for the tools the library wants to develop and use is *NIX (or any other conflict), what choices does the library have? To compromise on the developments and systems it works with, or go outside the IT environment – with or without IT’s blessing. Either way there is a risk. But it should be the library’s decision about the risk, as it is their business. IT is always concerned that they will get blamed if something goes wrong, after all they are the custodians of all things IT for the organisation. So how can we make this work so IT is comfortable, library can move forward, and customers get the services they need.
Libraries have to be careful not to throw away the benefits that centralisation has achieved, in the interests of freedom. Libraries need to realise the savings and other benefits too. So in moving away, we have to find ways to keep the benefits that can and have been realised, while we throw off the shackles and jump for joy at being able to finally put in place some of the systems and tools we think are essential to our business.
Maybe the optimum solution, if it is possible, (and I suspect in many cases it won’t) is partnership. Let IT provide us with the corporate tools we need (after all do we really want to be running email servers?) but take over the management of our unique systems. Will IT let us sit side-by-side with them and build a partnership of equals based on mutual respect? Probably not, at least in our environment, but it is worth talking about. My colleague says that IT wants to get rid of us and others, but at the same time feels that library, because we are such an informed group, could well set a model that the rest of the organisation could follow. That has promise.