Industry


Ads by TechWords

See your link here


Mark Everett Hall's picture
Mark Everett Hall

Sanity as a Service

Hate software. Give me SaaS.

Yes, my first two posts to this new Computerworld blog were critical of software as a service. After all, arguing that SaaS is hurting today's IT economy and going to cause IT integration problems in companies for years, maybe decades, are not ringing endorsements of cloud computing from a so-called advocate.

That's what I am, though, a SaaS advocate. But I am not a cheerleader.

SaaS is a fantastic option today for companies whose business units need IT tools in a hurry, but whose IT shops can't or won't do the work. Like outsourcers, SaaS providers can be good allies to CIOs.

But mostly, SaaS is what I crave as an end user. I am beyond bored with managing my own operating systems, network software and application portfolio for my Windows and Mac machines. (I gave up on my Linux box about a year ago. Too much trouble for too little ROI.) What I want, what I think most reasonably sophisticated computer users would prefer, is a robust, secure, reliable, diverse, customizable, high-performance and fairly-priced array of SaaS options.

I have more than a few demented and geeky friends who can't wait for a new distribution of Linux to arrive so they can test their favorite open source packages on the release. Nothing pleases them more than sharing hacks and shortcuts on Ars Technica and Slashdot or reporting kernel bugs to Bugzilla.

Thankfully, the vast majority of computer users are unlike those guys. (And all the ones I know are men; most women are too sensible.) Like me, working people think of their computers and the software they run as tools. Nothing more. So when it's time to upgrade an operating system or download a new application, we consider it a burden and a waste of our time. If only a skilled person would do it for us.

Then in walks SaaS, albeit tripping and stumbling, to save the day.

The SaaS model promises to eliminate the software maintenance aspect of our lives. But that cloud computing model and today's SaaS reality are just that, a model versus the real thing.

Still, the potential for SaaS is so great for both individual users and businesses, it needs critically-minded advocates to critique, cajole and commend SaaS providers as they slog their way down a long road to a new and ultimately better way to compute. I hope to become one of the voices of sanity along that path.

What People Are Saying

Hmm. I may be demented and geeky. Who knew?

Sounds like the "demented and geeky" guys are the ones who enjoy building and testing and tweaking the things your "working people" think of as tools. I might be ripe for hire for one of the SaaS providers.

Now, because my comments on one of your previous blogs kept getting marked as spam, I'm going to try to post them one more time here and see if I can get past the filter.

I see SaaS becoming more of an in house deployment model. Use the cloud computing paradigm, Host your own servers, and make the application available only internally, and over your VPN. No need to involve an outside provider at all. What is the advantage then? You eliminate all the hassles of deployment on the desktop, and can use lower spec'd or existing desktop hardware, saving money in desktop upgrades. Not only that, done right, you gain platform independence. Loser, Microsoft and, to a lesser extent, some hardware companies. Winner, everyone else.

The only other winner I see with the SaaS model is IBM, and companies that have an application that needs mainframe power, but only part-time, or can't afford the expenditure for the big iron. Using the SaaS model, IBM can provide the hardware and software companies couldn't get any other way. I think IBM has an earned reputation that makes them appear more trustworthy than other providers, and they have both the will and the lawyers to provide air tight non-disclosure agreements that work both ways, and provide an acceptable measure of protection for your data.

I take that back. There are two other companies that really are SaaS companies, and have been succeeding at it for some time now. It's just that that's not what people think of when they hear their names. Amazon, and eBay. The service that their software provides is a marketplace, and last I heard, it's working well.

That said, and looking beyond SaaS itself, can I coin a couple of new acronyms? SEaaS, and OSSEaaS ((Open Source) Software Engineering as a Service). I can imagine working in a company that provides that. Basically, it works like this. Joe Business has a need and doesn't have the IT manpower to do it all himself, but he knows roughly what the solution should look like. It's similar to what some other businesses are already doing, but he definitely needs some tweaking and refining to fit his business processes perfectly. Enter an OSSEaaS provider, let's call him Op, to save the day. Op sits down(likely via web conferencing) with Joe Business and takes stock of his needs. Op determines which open source cloud computing tools best fit his needs, and what tweaking needs to be done to make them fit perfectly into Joe's scheme. Op then determines the initial project cost based on the man hours he expects to spend custom fitting the tools, adds cost of the first year of software support, and determines the target date for delivery. He gives proposal to Joe, who accepts. Joe now has a nice custom SaaS style solution to his need with an agreed upon delivery date and a year's worth of tweaks as needed, for a nice fixed price at whatever terms of payment were agreed upon, up front, monthly, or whatever. Upon delivery, Joe hosts said solution on his own servers, accessible internally, and to his satellite facilities or home based workers via VPN, thus maintaining control of his critical data. Op gets residual income via software support renewals as long as Joe is pleased with the service. Hey, as I think of it, Red Hat, Novell, IBM, and SUN may be present day OSSEaaS providers. What do you think?

Agree and disagree

Sorry about the spam designation. As you know, that's not my bailiwick. (Maybe we use a "geekkeeper" as well as a gatekeeper...sorry, ;-))

I agree that internally companies will be adopting a more mainframe-like, SaaS-oriented architecture to deliver IT services to employee desktops, which are awfuly expensive for IT depts to maintain.

However, I strongly disagree that there will only be a few vendors who will be winners in the SaaS market. I think there will be many. Yes for Amazon, but more for providing cloud services than real software. Google, definitely. Microsoft, maybe. The jury is out on Azure at the moment. More importantly, there will be lots and lots of "Joes" out there offering niche services delivered via Amazon, Google and other cloud companies.

I put Sun, Red Hat et al in the cloud category and less in the SaaS side. (HP, too.) They are infrastructure not app-side providers.