On top of the problems earlier this week at Salesforce and Salesboom, the competing SaaS CRM providers, today Amazon's S3 storage service was down for around three hours this morning, dealing an additional blow to confidence in web-based services.

As with Salesboom, Amazon's response to these issues appears to have been entirely inadequate; unlike Salesforce's trust.salesforce.com monitoring and communication site, it appears that Amazon has no centralized or dedicated communication and response mechanism, and the response to customers, at least initially, was an underwhelming "We're investigating."

As I have said before, such issues hardly erase the utility or advantages of SaaS... if you have never worked at a business which had internal problems of this magnitude disrupting service, you've been exceptionally lucky. If you think that an SLA is impervious armor, then you have a poor grasp of math (99.9% uptime means 40 minutes allowable downtime per month). But it does seem true that SaaS is being held to a higher standard. Considering the numbers affected, perhaps it should be. On the other hand, the question is ultimately an economic one, as it is in every organization: can you charge enough for the service to make it bullet-proof without going broke? Considering the aggressive pricing on most of these services, the answer may very well be "no." If that's the case, then SaaS may have to wait for customer expectations or budgets to catch up with reality.