Software licensing is a topic that a lot of people like to discuss, but seems to change very little. Customers would sorely love to have a more usage-based model for paying for software, versus paying a flat per-license fee. Some software vendors have endorsed the idea, but most find any number of reasons for keeping things just the way are.
It's not that they don't have confidence that their software is being used, it's just that usage-based licensing opens up a whole other can of software worms that most vendors don't want to deal with.
The first of those issues is the sheer size of the applications. If we all moved to a usage-based model, it would not be long before customers started asking about paying for only the portion of the software that was actually used. As we all know, most end users only invoke about one-tenth of any given application. If we change the model to one based on actual usage, it would not be long before customers were asking not to pay for the 90 percent of the functionality they never use.
This would then force vendors to repackage their software around a core component accompanied by a large number of modular components that would each have to be tracked separately, in terms of actual usage. That could be done, but it would lead to a lot of difficult conversations with investors about what happened to the revenue stream of the company.
Of course, one major upside of this approach is that systems would consume a lot less power. One of the reasons that systems consume so much power has to do with the amount of software that is loaded on the system. If we're running less software, less power is required. That too would substantially lower the cost of running a fleet of PCs.
It may be in the face of Green IT, combined with a push for more efficient licensing models from the customers, there might actually be some substantial changes in software licensing. With new capabilities from companies such as Flexera, which makes the technology that monitors and tracks software usage, many things are now more possible than before. The only real question is, do we have the force of will to implement them?