A new report out from Forrester takes a chart-filled look at cloud computing, offering the analyst firm’s own definition of the cloud and attempting to dispel three myths they have noticed. Since we at GigaOM buy pretty heavily into two of these so-called myths — namely that a cloud is comprised of a scalable virtualized server environment and that it’s a low-margin business — I was eager to see where we had been led astray.

But I don’t think we have been. The report takes a big tent approach to clouds, applying the cloud moniker to both the end user market and to a class of goods it calls infrastructure-as-a-service. That’s far above the hardware level where Amazon, Mosso and GoGrid sit, and includes software-as-a-service and even consumer web applications like Zillow or Flickr. Such a broad definition doesn’t really help clear any of the fog for the industry, and would likely only serve to make the term “cloud” even more of a marketing tool than it already is.

As the chart shows, Forrester sees clouds where most people see web applications. The lower three tiers are in line with a post written by the CEO of RightScale, a startup adding a platform on top of the physical infrastructure of the cloud. We’ve tried hard to distinguish between the hardware layers and the layers resting on it, such as platforms-as-a-service and APIs offered for mashups, which Forrester calls application-components-as-a-service, but are we being too dogmatic? What we’ve called the cloud, Forrester calls hardware-as-a-service (HaaS). Personally, I could do with fewer aaSes in my life, but regardless, since our readers are pretty savvy on this topic, what do y’all think?

chart Copyright © 2008, Forrester Research, Inc.

comScore says that the United States has caught up with Western Europe in the adoption of 3G with 28.4 percent of American mobile subscribers having 3G devices versus 28.3% in the largest countries in Europe. That works out to about 64.2 million devices - up 80% from last year. When it comes to 3G penetration are Italy and Spain lead U.S. The growth in 3G penetration comes at a time when the data revenues are growing at a rapid clip for the mobile carriers.

If this story interests you, check out our upcoming conference:
Mobilize — The Next Generation Mobile Conference

With all the hype and excitement surrounding the release of Google Chrome yesterday, I, like so many, was eager to try the browser out for myself. What I didn’t expect was the overwhelming sense of déjà vu it would trigger in me.

I am a veteran in this industry, one whose first PC was a portable Hyperion I used when managing the sales of some graphics plotters back in the mid-80s. I went on to manage AST Computers’ Canadian operation, and from there went to Quarterdeck Corp., whose primary product was a DOS multitasking environment called DESQview, which was supported by QEMM, the most popular PC utility from about 1990 until the Windows 95 launch in August 1995. QEMM was a memory manager for PCs limited to 640K of primary DOS memory that effectively allowed them to create multiple 640K virtual machines. Simply put, DESQview allowed you to run Lotus 1-2-3, WordPerfect, cc:Mail and Harvard Graphics concurrently, taking advantage of the virtual memory architecture of the 386 and subsequent processors. Both AST memory boards, as well as their later line of computers, took advantage of DESQview and QEMM.

Reading the Google Chrome comic strip made clear the parallel to the emergence of QEMM and DESQview: All today’s browsers are effectively single tasking, in that only one tab can be actively processing, say, a JavaScript application at any given time (”inherently single threaded”), yet the tabs are interactive to the point where the misbehavior of an “application” in one tab can impact — and sometimes crash — the operation of the entire browser. Web 2.0 has brought about an array of browser-based applications and activities that require a more robust, stable, multiprocessing browser with each process assigned to its own memory space and associated data structures — which is basically how DESQview operated. Indeed, when I pointed this out during yesterday’s SquawkBox, someone labeled Google Chrome as “DESQview for the cloud.” Talk about “Back to the Future“!

But then Chrome goes beyond simply providing a true multiprocessing capability for web browsers. It eliminates memory creep/leak issues that I experience with Firefox; it has a “Task Manager” feature that allows you to view all running tasks and shut down any misbehaving tab without having to shut down the entire browser. Its JavaScript virtual machine architecture supposedly introduces speed, robustness and automatic memory management features. Its “Omnibox” feature combines the address bar, desktop/web search bar and browsing history to enhance, yet simplify, both the browsing and search experience. It addresses a range of security issues such as malware and phishing. But the real gem is that the entire development is based on freely accessible and reusable open-source code.

I installed Chrome and ran it on a quad-core desktop PC. Not only is it fast, but introduces an altogether different browsing experience than any I’ve ever had. For example today I had three windows open and when one crashed, sent a report to Microsoft and closed, the other two windows remained open and fully operative. The real test, of course, will come from using it over time, to see if it provides a smoother, more technologically transparent user experience as you add tabs and leave it running for a while.

In summary:

1. Chrome was developed from scratch as Web 2.0 has evolved to include a wide range of browser-based applications, thereby bringing new architectural and user interface demands to the browser experience.
2. But, as Alec Saunders points out, while it allows you to run web-based applications, it’s not an operating system that abstracts the hardware from the software. Leave that part to Microsoft, Apple, RIM, Nokia/Symbian and other platform vendors. Google wants the browser to act as an application platform, independent of operating system.
3. It brings out features that Google encourages other browser developers to incorporate into their respective browsers for a more stable, robust and secure browsing experience.
4. It’s beta, so there will be some bugs; and much like the Skype 4.0 beta, they’re looking for user feedback on the entire experience. The big question, as with some other Google applications, is whether it will ever come out of beta or will the key features first migrate to other browsers?
5. It’s not a threat to Microsoft, Mozilla, Apple or any other browser developer but rather a challenge to them to improve their own browsers such that all Google applications as well as any other browser-based applications can run smoothly and fully transparently to the underlying technology.
6. It has the potential to be an extension of the mobile Google applications I run on my Blackberry and Nokia N-Series phones, yet it can address more generically the issues of running any browser-based application on smartphones and other mobile devices.
7. As it matures, it has the potential to become a seed for developer innovation.

In addition to opportunities for application innovation, of course, fully transparent, smooth user experiences lead to significantly enhanced opportunities for Google ads. Nobody has ever made any significant revenue from a browser itself. Again I am reminded of my Quarterdeck days. We had a browser back in 1995 (including a feature equivalent to “Tabs”), but did not recognize it as simply a critical infrastructure component whose content and applications, not the underlying technology, would be the key to revenue generation. Been there; seen that.

Jim Courtney is an associate editor of Skype Journal.