Agree with both points from localzuk - on NIST and pricing of SIMS (and think I understand why pricing is as described). Believe that the NIST definition of cloud is fit for purpose and that it should be used as the benchmark to determine if services are 'true' or 'fake' cloud. SIMS as a thick client application (even with remote desktop deployments) isn't true cloud by NIST and isn't able to effectively share resources between customers. Effectively the footprint per client is much larger than with a true cloud service. This much larger footprint means much higher costs which translate into higher pricing to customers. To be able to drive down the cost to the schools the services need to be 'true' cloud and if they are would expect them to offer the service at a much more competitive price.
However would feel that localzuk is somewhat understating the costs/disadvantages for the on premise server option. As well as the capex there is the question of scalability/upgrades -assuming you need to add more users and remaining up to date with current technology. Does the on-premise option have the same level of resilence underpining the cloud SaaS service - so if that server breaks does that bring down the whole service, in cloud the architecture is likely to be designed to have no single point of failure.
As for the operational costs think it is rather more than just 'a couple of hours a month'. Firstly we need to assume that you have the right skills available to do the work and that they have those 'couple of hours free' each month. As well as deployments they will need to seutp monitoring of the environment and addressing alerts when they come up as well as the all too regular Windows O/S security patching.
Just want to make sure that when comparing prices of both options that the total cost of ownership is correctly calculated when determining which is the better value.
