Open Clouds? Yeh, sure…

I woke up this morning to see the fulminations of an entire industry focused on an Open Cloud Manifesto that wins, on face value at least, the oxymoron award of the year. It’s not open, it’s as much a manifesto as this blog post is a ham sandwich, and its impact on the future of cloud computing will, in my mind, be absolutely minimal.

With 20-plus years of covering so-called standards and so-called open movements under my belt, let me just state it as simply as possible: there will never be an “open cloud” that can provide that elusive “fire your vendor” or “switch to on-premise” capability that open proponents all dream of. Ain’t gonna happen. Our industry simply cannot, and will not, organize itself in a way that is as mutually altruistic as “open” seems to imply. What will emerge is multiple APIs, multiple standards, and a crowded field of providers that will be maneuvering to be better, cheaper, and more hegemonistic, all at the same time.

And that will be a beautiful thing. Until the market definition and discussion becomes mature, whereupon a brutal de facto standards war will emerge, allowing users to clean up in a highly competitive market with lots of vendors vying for the right to be the low-cost, high-value commodity provider.

We are way too early to the cloud game to start defining these standards, and to have manifestos that declare in absolute terms (which are then retracted in time for the conclusion) how the market needs to organize itself. What is needed well before we argue about the details is some hard thinking — and real functionality — that moves the market forward towards what cloud computing is really supposed to be. There’s lot of theories, and some practitioners in the market are showing us their first baby steps. But the real long term value — business value to users, not business value of vendors — is still to be defined in ways that justify the kind of statements that the Open Cloud Manifesto is claiming to make on behalf of the industry.

I say this fully cognizant of all the first generation cloud — aka SaaS, aka on-demand — products and services that are in the market today. Many of them are useful, many of them provide value, and very very few of them have begun to unleash the real value of the cloud. Which is NOT, to be as emphatic as possible, about merely running on-premise software and business processes in a cloud. The real value of the cloud — as I have fulminated before — is about doing things in a SaaS-model that cannot be done for love nor money in an on-premise model. The fact that too many of the signatories of the Open Cloud Manifesto have yet to demonstrate in a practical way their understanding of this fundamental principal convinces me that we have a lot of thinking and doing to do before we can start manifestoing, much less talking about open clouds.

This is truly a case of getting so far out in front of a nascent market that it starts looking a little silly. Can we please organize our efforts around defining value for customers, instead of moving quickly to create a vendor war over a market that still needs a lot more careful thinking and definition?

Didn’t we learn anything from the SOA market’s resounding thud?

2 thoughts on “Open Clouds? Yeh, sure…

  1. Very elegantly put, Josh. We are currently at the stage (re cloud) of analysts writing to analysts. The sanity check is that, in competitive environments, dogma doesn’t matter and no vendor is going to follow a manifesto of any kind, ever.

Leave a Reply

Your email address will not be published. Required fields are marked *