This is very much an incomplete thought, or a thoughtlet as I’ve started to call them (read, I’m too lazy to sit down and write a full blog entry):
With all the Twitter failures, Google Mail failures and the most recent, Sidekick failure  getting people up in arms on how bad or risky the cloud model is, I thought I’d throw my $0.02 worth in. I don’t believe that the cloud model is any riskier than outsourcing your bookkeeping, something that society has been doing for centuries. It comes down to knowing what you are getting and understanding the value of your information.
Cloud FAILs are becoming more and more prevalent because of the greater awareness in both the media and social news buzz (I think there is a thesis waiting to be written, “The impact of social networking on society and shaping of perceptions“) and because there are now more options available today that there have ever been.
Why is “cloud” attractive?
For the cloud user there are a number of perceived benefits such as;
- Expenditure reduction/removal – When you are paying someone, supposedly, only for the hardware, software and supporting services only when you use them it can certainly sound great. However that is not necessarily the case or strictly correct 
- Lower barriers to entry – When the underlying components of the service are being shared between a number of customers, the costs for the system are spread across all the users. This in theory opens up the ability to access a wider range of services and solutions at a lower cost. A classic example is the online customer relationship management (CRM) application provided by salesforce.com where it not only provides the database infrastructure but gives you access to a suite of tools and reports.
- Or in the case of Twitter and Facebook, give you access to applications and information you wouldn’t normally think to access.
- Easily scalable/elastic – The expectation is that in a cloud model, services are extremely scalable and can grow, or shrink, as the business demands. Example is Verizon’s Computing as a Service (CaaS) offering.
- Flexibility – with recent announcements in the advances in Cloud Data Management Interface (CDMI), that is the ability to tag data with information on what type of data it is (backup, archive, encrypted, etc) and therefore what to do with it. You could, ideally, decide that production data goes to a certain provider, backup or archive goes to another and whether or not any of it needs to be encrypted.
What are the risks?
Well if you hadn’t guessed it already, one of the biggest risks is loss of data and inability to access your data/service when you want, or need, to. The next question is of course; what can you do to mitigate the risks? The answer to that is is as simple, or as complex, as; what is it worth to you?
In the instance of twitter, facebook, or your blog site you might not care. If it’s your customer records, your financial records or your online portal to your business, you may care a whole lot more. How you go about reducing or mitigating that risk is up to the worth of that data to you. I can tell you now that something like photo’s of my family are worth a whole lot to me, so much so that not only to I backup my laptop constantly I also do periodic copies to my NAS. In the case of you priceless corporate data you may look to have an additional copy locally or with another cloud storage provider. In the case of your website, or corporate portal, you may have strict SLAs in place with the provider and that is sufficient. Then again there is always insurance 😉
That’s great but is it really worth it?
Being a technologist I’m going to say it is. Having access to these tools opens you up to doing things you may never thought of doing before, either through offering you tools to look at your data in new ways or simply off loading that administrative overhead and allowing you to focus on other things that are more key/critical to what you do.
So when you move your data and services to the “cloud” remember you not only get only what you pay for, know what it is you are getting.
Again this is an incomplete thought at the moment. I’m sure that there is a cleaner and more succinct way of putting it.
 OK probably not really cloud, more a hosted service, but goes to highlight the point.
 The Cap Ex vs. Op Ex is something best left to the CFO to decide how he wants to handle it. After all, that’s why he gets the big $$$.