On Thursday, I was at the Innotech CIO Summit in Portland, Oregon, speaking on cloud computing. I also participated in a cloud computing panel discussion later in the day with representatives of Salesforce and IBM. I would say that both audiences were aware of cloud computing, and even aware of its putative benefits, but were very skeptical about it, particularly with regard to data security and SLAs.

And certainly there was no sense of urgency on the part of the CIOs to do something about cloud computing. The questions I received were mostly about issues they perceived with cloud computing, and not practically-oriented questions regarding how to get started, or comparing the strengths and weaknesses of different cloud providers. And not question one about creating an internal cloud.

On Friday, I saw the other end of the spectrum at the standing room-only Under the Radar Cloud Computing event. This was the mother lode of cloud computing. The event had a general session in which several presenters gave their perspective on cloud computing. Besides the usual suspects (i.e., venture capitalists and a couple of cloud vendor representatives), there was a real live mainstream IT guy who is leveraging cloud computing today—Dave Powers of Lilly.

The general session continued with some company presentations. Interestingly, two of them, Eucalyptus and Abiquo, provide the ability to create cloud environments. These are aimed at internal clouds or at hosting providers that want to get into the cloud game. Both of them are open source, so provide a way for organizations wanting to create their own cloud to get started without licensing cost.

After the morning session, the conference broke into two tracks. I attended sessions on cloud computing system management and on cloud-based SaaS offerings.

The challenge of system management is something that you confront only after you begin using cloud computing. The most commonly-used cloud provider, Amazon, offers rudimentary tools to manage individual EC2 (aka virtual machine) and S3 (blog storage) instances. However, it offers nothing to manage the software components within the EC2 instances, nor does it offer any way to manage a collection of EC2 instances as an complete application topology. The most-commonly used enterprise management products like Tivoli and HP's Operations Manager have not been extended to fully support cloud computing yet (though I did see a demo of Tivoli managing an IBM-hosted cloud environment).

So, all in all, when you look to address this critical functionality, you need to look outside the cloud provider itself as well as the traditional systems management providers. So three system management companies presented: Cloudkick, enStratus , and Tap In Systems. Interestingly, enStratus focuses on security, specifically managing the keys and certificates used to authenticate in Amazon Web Services, which I can attest to is a difficult aspect of AWS to master. Cloudkick provides a general way to manage AWS EC2 or Rackspace Slicehost instances (Cloudkick won "people's choice" award at the event, BTW).

Turning to the Saas offerings, this was really interesting. One of the companies presenting was Symplified , an SSO cloud offering. Another was Zuora, an online billing company (Zuora was another winner at the event). Zuora was started by someone who came from Salesforce and saw how difficult it was to build an online billing engine and decided to create one as a standalone entity. Symplified was started by people who had already built and sold a packaged software SSO company. Their rationale for why a new cloud-based SSO company? While SSO is great, the effort to install and configure a packaged SSO product is too hard, especially because SSO is usually one of those "should" projects (in Silicon Valley parlance, a vitamin pill, not a pain pill.) The Symplified folks feel that offering SSO as a service will ease adoption. Moreover, they feel the pricing model, in which cost is tied to use, rather than cost being tied as a lump sum license fee to access to the product, will increase uptake of the service.

What I took away from this session: SaaS is moving into every software area, with a focus on ease of application use and better cost/benefit metrics. It's obvious that, for a mainstream IT organization, learning how to integrate SaaS offerings will be a key skill in the not-so-distant future. I would say the odds of startups offering traditional packaged software in the future are very, very low. Which means that IT organizations need to get over their discomfort with SaaS.

This was reinforced Tuesday evening. I co-chair the SDForum Cloud Services SIG and on Tuesday our meeting focused on investing in the cloud. Several VCs participated in a panel to discuss the impact cloud computing is having on their investment strategies. The key message: they are looking at investments very differently, with an eye toward companies that can provide a service that starts small within an organization and then grow as more people use it or additional functionality is used.

This typically means a much less expensive sales approach—no direct sales force knocking on doors, convincing CIOs to buy. SaaS is much more akin to open source, where little effort goes into major sales efforts; instead, the product is designed to be easy to obtain and use. This means another key skill for IT organizations will be to be comfortable with low-cost, low-touch trials. Around 140 people attended the event and the energy about cloud computing was palpable.

Overall, the atmosphere attending these events reminds me of the Internet as it began to emerge into commercial use. At a certain point in time, the technology vendor community, especially startups, just caught fire about the Internet. They were convinced that, once experienced, no one could avoid adopting their work lives to the Internet. At that same point in time, mainstream IT looked at the Internet with a skeptical eye, focusing on its shortcomings. At that time, I heard statements like "nobody is going to let their data cross insecure public networks" and "Nobody is going to put real business functionality out on the Web."

Of course, the indisputable benefits of the Internet overwhelmed the dubious responses. As we look back now, the chaos and cynicism is hard to remember, but believe me, it was there—and strong. But those attitudes didn't stand a chance against easy access to information, and I think it's unlikely that a jaundiced view of cloud computing is going to prevail, either.

Bernard Golden is CEO of consulting firm HyperStratus, which specializes in virtualization, cloud computing and related issues. He is also the author of "Virtualization for Dummies," the best-selling book on virtualization to date.