I wonder will we see on-demand library systems emerge. I am thinking of services like salesforce.com (the poster child of the on-demand software phenomenon, it provides Customer Relationship Management services) or webex.com (a conferencing and online meeting provider). The idea here is that rather than installing local instances of an application (CRM or conferencing/meeting management in these two cases respectively) an organization can use a central, web-based application. Software functionality is not provided as a package to be installed locally, rather it is provided as functionality on the web. Software as a service.
Potential advantages are lower cost of ownership, less risk, and smoother and more frequent upgrade. Potential disadvantages include less local customization and flexibility. The model is in early days, but was given a boost recently by Microsoft's announcement that they were going develop network based versions of their applications (see Ray Ozzie's memo to Microsoft staff). Which prompts mention of Writely, a web-based word processing application.
In parallel, we are seeing other on-demand services. Thom writes about using an on-demand 'virtual server' from a remote provider.
Now, several library system vendors will 'host' systems. But this is somewhat different to what I am talking about here. Typically, they are hosting an instance of software which can be used remotely. An example of an on-demand service in the library arena is RefWorks, a web-based citation manager.
It may be that one reason we have not seen more on-demand solutions from existing library vendors is that running the two models together is difficult. Marketing, sales, development and support will all be done in different ways, increasing costs to supply. An on-demand solution has a different dynamic than multiple locally deployed instances. For example, it may take time for changes to fully ramify through locally deployed systems: users may be working with different versions, or have applied different patches, and so on. While it is possible to upgrade everybody simultaneously in the on-demand model. In addition to this tension, the on-demand product may cannibalise existing products.
An alternative would be for a new entrant to provide service, or maybe for an existing provider to try the new model in different markets (India, for example). I reckon that we will see some such moves in coming years.