Thanks, Travis. Corrected.

It's impossible to call a winner at this point, but I would follow closely major vendor announcements in the coming weeks and months.

Thanks, Amnon.

Agreed, Peter. In my CloudConnect talk I mentioned that this phenomenon has occurred even before open source software with the adoption of PCs and Unix as two examples.

Yes, Jelastic seems like a good fit. Updating the post.

Actually, after looking at your site again, I remembered that the reason I didn't include you is because yours is a private PaaS solution. There are others in that category that I did not include such as GigaSpaces Cloudify (http://www.gigaspaces.com/cloudify) and Cumulogic (http://www.cumulogic.com/). I'll add all of these solutions and yours in a separate category.

Diane -- I did look you guys up but it looked like you were a .Net solution.

Thanks, Brian. I think that the objections you are raising to the dominance of PaaS and Public are temporary issues. These predictions are for the long term (think 5+ years out). Today, we have a confluence of reasons that are keeping them from coming true, namely: Public IaaS and PaaS offerings are still immatureThe market is in flux, with new offerings and new features coming out almost literally every dayCustomers are still learning what works and what doesn't All that will eventually be resolved.

Makes sense but that's why I wrote "IaaS doesn't make sense, except for a limited set of scenarios". You're talking about that limited set, in my opinion. I completely agree with your analogy between PaaS and software frameworks -- and I think it proves my point. Just as the vast majority of apps being built today use a framework, unless they have very unique needs (and even then there might be specialized frameworks), the same will be with PaaS.

Thanks, James. Perhaps complex is more of an issue than opaque. In one sense, it will become more complicated to decide from whom, when and how to purchase compute resources. And because of this complexity, it will become less apparent what the "right" price is. That said, I believe that in parallel we will get new technologies and people that will help in making that decision, and thus create a more efficient market.

Thanks, Ross, and great to hear from you. It's been a while. I am not very familiar with Gartner-speak on this issue, but indeed, under their terminology, I focused on Java aPaas here. I have written about Integration-as-a-Service almost two years ago: http://gevaperry.typepad.com/main/2010/01/integrationasaservice-heats-up.html. I guess it's time to give an update on the space.

Thank you, Anders. Link corrected.

Kent -- Thanks for the comment. What do you mean by "filtered"?