Amazon's S3 (Simple Storage Service) started out the year plagued with widely publicized outages and provoked a significant dither in the blogosphere over the future of SaaS and cloud-based services in general, but none of that has stopped it from undertaking impressive growth in the past quarter, and lowering storage rates in the bargain.

Starting in November, S3 will adopt a tiered pricing model which will retain the current $0.15/GB rate for under 50TB of storage, but then trend downward through a series of steps to a low-end price of $0.12/GB for anything over 500TB. If the storage rate was a good deal for you before, you've got to love it even more now.

Amazon's Web Services blog has details on both the growth and the new pricing structure.

Last week when Entellium's CEO, Paul Johnston, and Senior VP Parrish Jones unexpectedly stepped down and the company subsequently began to lay off staff, many of us wondered (despite the inevitable reassurances to the contrary) if we were going to see the first significant failure of a SaaS business and the subsequent ghastly fallout among customers who may lose both their software and their information in one fell swoop. Considering the state of the market and the thin margins in the business, it seemed like this was the first of many economically driven SaaS business failures. It turns out it was something more sinister. Johnston and Jones are looking at federal wire fraud charges from the wrong side of the prison bars today, facing up to 20 years and $250,000 fines for "overstating" Entellium's revenue to investors. In the economic climate we're currently in, with people baying for blood over various forms of corporate mismanagement, it's unlikely they'll find much sympathy when the case goes to court.

But the criminal sideshow is distracting us now from what was at the forefront of people's minds last week: how will the financial collapse of the service affect customers (although apparently there were fewer of those than we had been led to believe)? Was there a plan of some sort for returning the information? Several other SaaS vendors commit to delivering data in pre-arranged formats to customers in the event they halt operations... did Entellium provide such a guarantee? Apparently there has been some discussion between Avidian and Entellium's investors regarding a potential purchase of the company. If another company acquires the assets, will they continue operations? Will customers feel confident enough to keep their business there regardless?

As horrible as this downturn is, it may provide us with some of the first real data on what happens when your outsourced software services disappear, a mostly theoretical threat to this point which has been held out by more traditional software manufacturers as the boogeyman to keep customers away from SaaS offerings. It's obvious that things could go poorly in such a scenario; it's less clear that they actually will. I know of many businesses who have been bit by circumstances in which their traditional software vendor has closed doors, or simply discontinued the product, and it's a problem, but never an insurmountable one. What remains to be seen is whether SaaS failures will fall into that same vein (which is my suspicion) or whether they really are more risky for customers than traditional software company failures.

I don't mean literal train wrecks... sorry, I suppose that's a rather tasteless metaphor at the moment. But it's the one used in this Silicon.com article, which may or may not have been quoting Gartner analyst Scott Nelson. I haven't seen the Gartner report, nor does the article give it's name, only that it warns of the potential for "IT train wrecks" if certain warning signs aren't heeded.

I happen to think that's a rather obvious generality, but then the article goes on to give one (and only one... here, again, I'm unsure of whether to lay this at Gartner's feet or not) of these warning signs: spending big on hiring programmers to keep an old system running.

Of course, I would say it's equally intuitive to say that if you are spending big on any maintenance item it's probably a warning sign, but in this case, I don't think Gartner could have picked a worse example. Nelson's advice, according to the article, is that "Gartner said companies should focus on developing an application strategy that involves retiring older systems." This may be conventional wisdom, but as it happens, it's also pretty bad advice for avoiding train wrecks. Recent studies tend to support the point that it's more costly and more complex to develop new software than to maintain the old. More costly and more complex add up to more likely to run right off the rails, in my experience.

I would never advise against having an application strategy, but I would think long and hard about recommending that you center it around retiring older systems. I might say poorly performing systems, but the idea that you should evaluate a software system simply based on its age rather than its function is excessively simplistic and flatly ignorant of the costs of replacement for custom software in today's programming environment.

I don't know of many companies which don't work very hard to establish a sort of amicable brand identity with potential customers, and Google is no different. But there is a danger in taking these marketing messages at face value, and it's not simply that you might begin to think of these companies as your friends, but equally that you might start to see them as your enemies.

Both of these reactions are understandable, but entirely flawed. When you are a customer, or potential customer, dealing with a vendor, you are not in a personal relationship, but a business relationship. And words like friends and enemies do not apply. Your judgement should not be predicated on warm fuzzies or on fears of evil intent, but rather on a rational cost/benefit analysis, just like they tried to teach you in business school. It's not about whether the company is cool, or whether they are sneaky... it's all about the bottom line. If you don't realize that Google is in a business, and that you are entering into a business dealing with them when you accept their terms of service, then you might think of them as friendly; if you don't have that realization and they suspend your account or similar, then you might think of them as back-stabbing betrayers. But you would be wrong on both counts; they are a business, engaging in profitable activities. There is nothing personal about it. So long as you can align your own interests with theirs, there isn't anything wrong with that. But you do have to realize where your interests and theirs lie, and when they might differ.

Of course, that's why you read blogs like this, isn't it? We're reading the tea leaves of major vendors' intentions. But then, this is a business, too... where are our interests, and yours? ;)

Ever since Ballmer spilled the beans on the upcoming unveiling of Microsoft's "Cloud Operating System" scheduled to take place at this month's PDC (Professional Developer's Conference) in LA, I've been poking around for more hints on exactly what that is and how it will fit into the existing broad range of services offered under that label. There isn't a lot out there, which is unusual for even "skunk works" Microsoft projects, which tend to be long in the tooth and full of geeks eager to chat. We know that the company has been working on something internally dubbed "Red Dog" since early this year, at least, and that a host of other efforts have been underway in the company, from data center buildouts to revamping major enterprise applications into massive, hosted, services platforms, but it's unclear how, or even if, all of this will fit together into a unified offering of the sort that Ballmer apparently is referring to.

I find myself re-reading David Chappell's August whitepaper "A Short Introduction to Cloud Platforms" (misplaced the link, but I assume it's available on Microsoft's site somewhere) which was commissioned by the company as "An enterprise oriented view" of cloud computing. Chappell stresses the duality of on-premises and cloud applications, which echoes Microsoft's "Software+Services" mantra closely, and perhaps provides some insight into the company's direction. While most existing cloud platforms are intended to work primarily stand-alone (although they can certainly be integrated with on-premises or third-party services... such is their nature) it may be that Microsoft's platform will be designed to rely more heavily on existing on-premises applications and services. If that's the case, it could also explain the relatively short lead-time... if they didn't have to build the platform from scratch, or as an entirely stand-alone service, a great deal of leverage could be exacted from current product lines. Expect to see some integral components coming from places like BizTalk Server and, of course, Windows itself.

As I have said before, I think the Software+Services mantra is a good approach for Microsoft and potentially much more palatable to their existing customer base than the brave new world of SaaS. This approach to cloud computing may also find favor with current customers. I have to wonder, however, about the suitability of the platform as a true cloud service, at least as I define the term. I found it telling that the company finally (finally!) ripped the GUI out of its new dedicated hypervisor virtualization server, but it says something that a business attempting to offer performance-oriented servers insisted for so long on slapping an inefficient and resource-intensive GUI on their server operating system. They've worked hard to wedge the core Windows OS into almost everything they have done, including places it's not really best suited to be. If they have worked their cloud project up to speed on that basis, then it explains how it could be unveiled so quickly, but it may also mean that it's less well-suited to the large-scale platform services it is supposed to offer than some of its competitors.

Hopefully we'll find out more specifics in late October.

I posted a couple weeks ago about an effort at Hewlett-Packard to develop a lightweight, battery-conscious operating system to enable users to access some of their most commonly used applications without having to boot up into Vista, a time-consuming and resource-intensive process that seemed hardly worth the trouble.

Now it appears that Dell is moving the same direction, developing a Linux based subsystem for new laptops dubbed "Latitude ON" which will allow rapid access to common applications, outside the main operating system on the laptop, which promises multi-day battery life in the bargain.

At this point, few details are available regarding either the HP or the Dell effort, but if those "frequently used applications" include a web browser and an e-mail client... I imagine a lot of people will start weighing the pretty bells and whistles against the increased battery life, and spending a lot of time not using Vista at all. Enough of that, and they'll start to wonder why they are paying for it... not a good thing for Microsoft.