This article can also be found in the Premium Editorial Download "Storage magazine: Exploring new options for disk-based backup."
Download it now to read this article plus other related content.
In technology, things both cool and stupid are most often architected, designed, developed and built in a vacuum. Sometimes we build things for the wrong reason or simply because we can. Sometimes we build things for the right reason--the world needs it and the business case is compelling. And sometimes we do the wrong thing, but for the right reasons.
Let me give IBM some credit. It was arguably the worst-ever N.I.H company in storage, but it seems to have finally seen the light. IBM spent 54 gazillion dollars over the last seven years trying unsuccessfully to build a Network Appliance (NetApp) NAS-killing product line only to wake up and smell the cheese recently. IBM finally recognized that while this might have made sense if it was ever able to actually do it--and time was static--and this really was strategic to IBM, neither scenario was true. So it did the smart thing--it did an OEM deal with NetApp.
So why should IBM get a piece of the action, unlike so many others who continue to get nothing when it comes to NAS? Management woke up to the fact that it makes a heck of a lot more money on all the services and ancillary products that go with keeping the customer happy
Could this be an epiphany? Perhaps, but there's a precedent. IBM's DS4000 is made by Engenio, and it just so happens to be the most successful storage array in the company's long history. IBM finally figured out that the RAID code so core to Engenio wasn't so core to IBM, and customer satisfaction was the issue.
Now, if IBM can figure this out, why can't everybody else? EMC is another big N.I.H shop, and one that doesn't like OEM deals. If it finds something that's cooler than what it has, it tries to buy it. I guess I can't blame EMC, but not everyone is for sale. NetApp is just as bad--and so on and so on. I find it funny how fast the "open up" culture can get slammed shut. Startups complain about the N.I.H dilemma in the big guys, but as soon as they're bought by one, they act worse. For example, Pirus Networks got Sun Microsystems to agree that it couldn't do what Pirus did without taking a thousand years, so Sun bought them. Now the Pirus guys have the reputation as being the worst N.I.H guys in the business. Doth thou forget from whence thy came?
To a degree, N.I.H is good. It means engineers with egos believe they can do anything. I like that attitude. But it takes a logical business person to sift through the bravado and make the determination of what is truly strategic, intellectual property and what is best for the customer and, as such, the company.
If every product requirement discussion started with "What does the customer need and truly benefit from having?" and then migrated to "What's the best way to deliver that value to the customer in the shortest time frame?" many of these issues would go away, and we'd have the benefit of seeing more technology faster. The world likes buying from big companies. The way to lose a customer is to overpromise and underdeliver. The best way to profit from a customer is to partner with them and provide the tools and services they need--regardless of the "inventor."
This was first published in June 2005