It's axiomatic that in software of any complexity, the ecosystem of plug-in products, tools, compatible APIs, and developer community can become really important. In certain software product categories, the importance of the ecosystem can swamp any feature advantage that an upstart product may have, leading to what economists call a natural monopoly.
This doesn't apply to all product categories, though. And in very mature technologies, it is possible to unseat the established leader's ecosystem with a "good enough" substitute. Think "The Innovator's Dilemma," and then think Linux, Open Source, and Google docs.
But if you look at cloud computing, there's no part of it that could be called "very mature." The argument can be made that the cloud is all based on Web services, so the ecosystem doesn't matter as much as it did with traditional software. But that argument is naïve, at best: at the detail level, cloud vendors' APIs are proprietary, and learning curves can be quite steep. And that ain't gonna change.
So the number of tools, plug-ins, and developers can really matter when evaluating cloud vendors. Where in the clouds is the ecosystem critically important? The general characteristics of these segments are:
• Technical platforms, such as integration, logging, development, or deployment
• Social Networking platforms -- the usual suspects
• Advertising and gaming networks -- the usual suspects
• Commerce platforms, such as Amazon, eBay, or PayPal
• Services that are toolkits disguised as applications, such as Google or Salesforce.com
For cloud vendors that span two or more of these characteristics, there's a doubling of the ecosystem's impact. The valuation of Facebook and LinkedIn doesn't come just from the number of eyeballs.
At the other end of the spectrum, there are lots of cloud services where the ecosystem isn't critically important, such as e-mail, storage, news feeds, or photo sharing. And certain kinds of cloud apps such as invoicing, banking, or HR can get away with just a couple of connectors to popular accounting packages.
Measuring the Size and Quality of the Ecosystem
In any vendor competition, there's the stuff that counts and there's noise. The same goes with the ecosystem. For example, Microsoft can probably claim the most APIs of any vendor, the largest developer community, the most plug-in products, and the most consultants. But almost none of that applies to the cloud at this point, and nobody thinks they have the strongest cloud ecosystem.
So the first part of measuring the strength of the vendor's ecosystem is clearing away all the obvious marketing chaff. Next step is to look only at the part of the ecosystem that's relevant to the cloud service you will actually be using. What remains still needs to be vetted, as there will plenty of obsolete or uninteresting "information" about the ecosystem. For example, you may find 100 vendors who are nominally partners, but if only 20 of them are in your country and you've never heard of them before, some additional scrutiny is warranted. Likewise, you may find lots of partner products and services that may not have been updated in a long time. A strong ecosystem has vendors who have not only been there a while, but have released several versions of their products or services. If the ecosystem has a vendor ranking and rating system (Salesforce.com's AppExchange is a great example of this), look for the quality and recency of customer feedback.