Earlier this morning, the editorial staff at eWEEK fell into an interesting discussion about the latest Amazon Web Services S3 (Simple Storage Service) outage and how big a deal it is. Opinions varied from big deal to not so big a deal.

I'm in the not-so-big-a-deal camp. Look, even the massively parallel, all-powerful scaling Google had a few hours worth of outages in its nascent App Engine service last month when a bug in its datastore servers messed up the data returns for developers. This lasted a few hours.

I know, I know. You're going to say apples to oranges. Amazon's EC2 (Elastic Compute Cloud) and S3 have been around for like five years while App Engine can't hold a candle to it in terms of breadth and depth. Amazon has hundreds of Web services customers, while App Engine is still in preview period.

Just as the earth rotates, day turns to night and the seasons change, outages happen. Google's search has been occasionally wonky, albeit only for a few minutes or so at a time. By the time angry users blast the company in blogs, the search is back online.

Om Malik is correct in suggesting that: The S3 outage points to a bigger (and a larger) issue: the cloud has many points of failure - routers crashing, cable getting accidentally cut, load balancers getting misconfigured or simply bad code.

Read Write Web's post on S3 today captures this essence well, too, and my colleague Michael Hickins ties these together and points to the resulting opportunity potential in this post.

Here's my take: Look, chaos reigns on the Web. As long as you have humans programming for it, you're going to have some outages. It's happened to Google, eBay, Yahoo, Amazon.com and every other major Internet service provider at some point.

In the last 10 years, the amount of transactions executed online has skyrocketed. The more transactions we conduct, the greater the margin for error. Sure, Twitter's scalability issues are annoying as all get out, but we live with them because we love the service.

Wait a minute, you say: Amazon S3 users are running businesses and can't afford to go down. Well, neither can Amazon or eBay, but they go down drinking their own Kool-Aid. How many billions of dollars have those e-commerce vendors lost from executions that didn't get completed during those outages?

After our editorial call, my Microsoft Watch colleague Joe Wilcox argued in this post that Microsoft could make some moves to improve on what Amazon Web Services has done.

OK, but let's assume everyone is working toward that 99.999 score, that holy grail percentage of scalability.

Bearing in mind the numerous paths of unpredictability and chaos outlined by Malik in his post, you could argue Google, whose search pretty much is available 99.999 percent, is best positioned to lead in infrastructure scalability.

BigTable, MapReduce and the Google File System have been crafted exceptionally well. But suppose Google finds a way to duplicate the success of its search architecture to its App Engine and proceeds to offer a storage and database service as infrastructure the way Amazon Web Services does.

Google may be best positioned for reliability and success as a platform infrastructure provider for scaling Web applications. But Google is still merely dipping a toe in this potentially deep pool with App Engine.

If Google corrects the failures the other platform providers endure, it may ultimately become the eminent Web-hosting provider in the world.

We say Google Universal Search now, but will we one day say Google Universal Platform as a Service? What do you think?