I wanted to get a “Jump the Shark” type reference in the title of this post – but it didn’t make sense, and frankly is a pretty played out term. Anyway…Sage Software has just announced that they are going to make their applications available in the cloud on Amazon Web Services (AWS). Hmmmm…

To me, what Sage has done here is essentially skip a major phase in the evolutionary cycle of application design, development, deployment, management etc. needed to fully excel in the cloud. That stage is SaaS.

Look, it may seem nit-picky. After all, spinning up applications on a one-off manner on Amazon is in a way very similar to what Sage does now – in terms of its partners selling and managing app stacks on a 1:1 basis. (In short, in an out-dated and increasingly inefficient model.) But, there is a major benefit one learns in managing SaaS applications – that of scale, how to share resources effectively, manage remote upgrades, etc.

(Here is where the political history buff in me would love to go on a diatribe about how important it is to fully realize evolutionary cycles in order to completely reap the benefits of mature development – and cite the fall of totalitarian communism as a clear end result of not fully realising Marx’s dialectic. But I won’t go there…)

Sage has talked about SaaS a lot, but as an analyst I never really saw much come out of it. It’s SaaS products acquired in the Accpac acquisition years ago were OK – but that was early SaaS. A lot of vendors, Sage included – have yet to impress me with any sort of scalable, highly available web architecture.

I am curious as to how much had to be altered to some of Sage’s older CRM products (specifically SalesLogix) in order to port them to the cloud. I gather not much – part of the beauty of AWS is that you can pretty much support any environment up there – as long as you build that stack and maintain it yourself or instruct partners how to maintain that very specialised stack. Compare this to a standards-based model where your app can run in any cloud environment that support – say Linux, Apache and PHP (in short – every cloud environment you can think of).

I wonder how this will play out… Will the absence of SaaS really matter in the long run, if we are truly moving to the cloud? After all, isn’t the cloud where everything is happening? It is if you ask a company like Salesforce.com. But wait, their core app is SaaS, and not cloud…so… maybe there is something important about hitting your SaaS stride before bringing a strong Web architecture into the clouds. Again, a hearty “hmmmmm…”