Not quite 25 yr. ago was a very profitable outfit that provided time sharing (Tymshare - I believe it was a division of McDonnellDouglas)... there were others (IBM) ... and contrary to what you say, the model worked quite well then... and is beginning to work well again. At that time, I was working for a datacom co. that sold X.25 pads/switches that many co.'s were using to link to Tymshare apps. Today it should work even better because back then the "public data network" was not so public as it was usually an X.25 net that charged on a per usage basis. Now the net is free with the only charge coming from the ISP for net access.
Many softwware functions are moving to the web, I believe this transistion is more revolutionary. The consideration with INFM is whether they have the customer base and technology to be a survivor. There is little question that this is a valid approach, especially in the middle tier market. Any comments from the thread, this would be a good debate, as would comments regarding an in debth review of INFM's customer base, which is lacking on the thread other than a few comments on 'gaming', which is a limited nitch market.
ASP is a great idea for getting rid of your home answering machine. It's cheap, reliable and has benefits you won't get by owning your own machine.
What nobody seems to have figured out is that farming out ERP systems to an ASP doesn't confer any benefits. A number of "Industry GURU's" have expressed surprise that most industry users in the ERP space aren't dying to try ASP. DUH!
What do you get by going ASP? Share hardware costs, don't have to hire your own tech or ops people.... What else is there? BTW, these items, while marginally cheaper, are less than 5% of the cost of an ERP implementation.
How does going with an ASP cut down on training costs? It won't! How does it handle user specific modifications? With difficulty! How will going with an ASP make implementation go faster? It won't! The tech portions of an ERP installation are NEVER on the critical path. Defining needs, loading data, testing, training the users and doing modifications ARE the critical path and ASP won't help this a bit.
People who believe that ASP is a good strategy for ERP projects don't know what they are talking about. What customer would pay more for a less flexible solution that has no measurable advantages and puts them at the mercy of a third party? Double DUH guys. That's all there is to it.