Joe McKendrick at Forbes dives into a pretty interesting angle on the whole phenom of "cloud computing" — what happens when you want to leave?
He maintains, with some justification, that a number of popular cloud platforms might actually end up being worse for customers, "erasing the gains we’ve made in terms of vendor dependence lock-in."
There is a whole ecosystem of vendors, analysts, press, and bloggers that have started to focus on this — and one theme that's emerging is that open source can be part of the answer. How companies approach this is, in some significant part, a function of size and technical sophistication. Generally speaking, the larger the company, the more internal IT resources they're going to have — and the more homework they're going to do on the technical details of the individual cloud platforms. These companies will probably find their way to cloud infrastructure vendors and projects; just Google "open source cloud" to get a sense of what the latest cast of characters looks like.
Those guys aren't really my focus here. I'm going to assume that they've got the discipline, the internal processes, and the resources to properly evaluate such offerings like they would any other strategic partner.
No, I'm more concerned here about the small to midsized companies that might be tempted by hosted ERP and CRM applications — which may or may not meet today's tightly parsed definition of "cloud" versus "SaaS." At xTuple, we hear all the time from companies who have fallen into exactly what McKendrick was describing: they sign up for NetSuite or Salesforce because it seems so darn easy, and then when the prices go up by 100% or more, and/or the company's needs change, it's next-to-impossible to get their data back.
There's a simple answer here that should be on any company's software evaluation checklist. If a cloud vendor won't commit to making a full export of your data available to you at any time, for a reasonable price, then run away and don't look back.
Ned Lilly
In October 2001, Ned co-founded xTuple, originally called OpenMFG, with the aim of bringing the worlds of open platform software and enterprise resource planning (ERP) together to solve the unmet needs of small- to mid-sized manufacturers. In 1999, he was a co-founder of Great Bridge, an early business built around the PostgreSQL database which is also the core technology for xTuple today. Great Bridge was incubated inside Landmark Communications, a mid-sized media company where Ned directed corporate venture investments, mergers and acquisitions (M&A) and startup activity. Prior to Landmark, Ned worked for a regional technology group in Washington D.C. and had a brief first career in political media — television, radio and a non-partisan news wire. He holds a B.A. from the University of Virginia and an M.A. from George Washington University.