Peter

From a community standpoint; do you think the service fees charged from force.com will limit or accelerate the growth potential of PAAS? Will we see more developers pay for force.com or not? Should we expect them too? Something I have been thinking about.

Brian

Brian,

I sometimes think that on my Form 1040, I should list my profession as "economist" rather than my usual fill-in of "consultant" or "engineer." I see the key issues surrounding most technologies in terms of the balance of value versus cost, and I believe that in the long run prices reach equilibrium with value. Overpriced goods and services attract competition; underpriced goods and services fail to offer a competitive return on investor equity. (I've previously commented on the relationship between these ideas and "net neutrality" at http://www.eweek.com/c/a/Infrastructure/Will-Web-Apps-Need-COPEing-Mechanisms/ )

Pardon the prologue, but that's what seems to me the key to the question of "What's PaaS worth?" I believe that an enterprise-grade platform, with peak-load capacity and uptime percentages and security that meet corporate and customer expectations, is worth more than a purely social or recreational platform that's supported by advertising revenues. As of now, that's the competitive posture of Force.com. Personally, I'm comfortable with that positioning.

This should come as no surprise, given that I spent 18 years writing for PC Week / eWEEK -- whose enterprise audience is accustomed to paying what things are worth. I didn't understand as well the consumer and enthusiast audiences of publications like PC Magazine, and thus I rarely felt that I had any useful thoughts to share with that audience. I know what Force.com is worth, and I believe that it offers good value -- and that, in the long run, excellent value combined with excellent execution is what wins. Hey, it's working for Toyota...

Thanks for being part of the conversation.

- Peter

Meanwhile, The Register reports (http://www.theregister.co.uk/2008/02/13/sun_datacenter_2015/) that rumors of the impending demise of Sun's data centers are premature. I don't think this weakens my argument, but it does remind us that organizations sometimes take time to acknowledge and adopt fundamental change.

Peter.

Slightly off topic but I'd welcome your thoughts on how to address the question "What happens if my PaaS provider goes under?" In an on-premise world my applications will continue to run even if my application provided is no longer trading or is acquired or de-supports their product. That gives me some time to implement another solution. In a PaaS environment if the platform goes away then my application goes with it. I've seen other Saas providers reply to this question with "it's no different then any other vendor going away" but I think we need a better answer than that. For sure both my legal folks and my prospect's legal folks need a better answer. Should we be demanding that a PaaS provider has a minimum notice period that they must give before a service can be turned off? Thanks. Kevin.

In any application delivery decision, there's a tradeoff between speed of development and independence from any given platform.

Imagine that you decide to bring your idea to market as a platform-independent application. You need to begin, most likely, by evaluating and choosing a cross-platform GUI library. You then need to write your own code to perform high-level functions that a platform’s API might otherwise handle for you. The Force.com API, being especially high-level, makes this differential cost especially high.

You then need to decide which customer platform stacks (OS, database, middleware) to support; you have to write code and installation scripts for all of them. You need to do all of this before you even start selling code that does anything original, or even useful.

Alternatively, you could make your debut in the marketplace as a Force.com application. You spend zero time evaluating, choosing, and supporting customer infrastructure, because there’s no footprint from your application on customer infrastructure. You develop your key business logic in Apex, if you wish, or you implement some of that logic in Java or .Net or anything else that can be exposed to Force.com as a Web service. You go to market quickly with a product that’s inexpensive for the customer to evaluate and adopt. How much is that worth in speeding the process of becoming a viable company?

Hypothetically, you might find a year later that there are customers who want things you could only deliver as an on-premise application -- although that list gets shorter all the time -- or you decide to offer an open-source version of your product. Perhaps you do this to insulate yourself from service outages. You re-host the business logic quickly, you implement the things that Force.com was doing for you – at considerable expense, but you can do that, and you only implement things that work and that you want to keep – and you crank the input of your version 1.0 users into your version 2.0. But you do this as a going concern, with an existing customer base and revenue stream, and without the risk that someone else will get first-mover advantage while you’re laboring through a traditional development cycle just to get version 1.0 out the door.

Thanks Peter. Certainly understand your arguments from an application developer's perspective. I think for potential clients looking for comfort that the platform isn't going to disappear is to look at the number of subscribers and assess whether a million plus users means the platform has reached critical mass.