Information Playground

Invent and/or Deliver?

Most people that climb the software engineer promotional ladder get to the point where they ask:

"Wouldn't it be better just to generate the ideas and hand them off to other people to deliver?"

Often times this question gets asked during the end game (otherwise known as the grind).

Is it better to invent, or is better to deliver? Does the tradeoff of "innovating the fun stuff" trump the "satisfaction of delivery"?  Does it have to be "or"?  Or is it better to invent and deliver?

I spent the first fourteen years of my software career doing both, and then I left the "delivery" role behind to join a group known as "Advanced Development". This occured after EMC's purchase of Data General Corporation (I was working in DG's CLARiiON organization at the time of the acquisition). After the acquisition cleared there was a six-month "no-transfer" moratorium. When the six months expired I transferred down to Hopkinton to join a team with one rule: if you successfully come up with an idea that is funded for production you are expected to hand it off to some other team to deliver.

Of course, if anyone in the group had an idea that they wanted to pursue and deliver, they were welcome to leave the team!

I functioned in that role for about three years. The first idea I worked on ended up being known as StorageScope; the second idea I worked on ended up being known as PowerPath Migration Enabler.

In both cases I handed it off and let somebody else deliver it. And in both cases I felt like I was missing out. So I left that job and headed back to a true development organization. Why?

Because the camaraderie trumps the grind. I wasn't having lunch with anyone and discussing the latest Dilbert-esque moment that was obstructing our progress. I wasn't sitting in the team meanings having a laugh with the manager about his slick PowerPoint animation (how many hours did that take?). And I was certainly missing out on the part that comes after the grind: the release of the product and the results from the field.

Back to Development, Head Up

So I rejoined a development organization and have stayed in that role ever since.

Having said that, taking on the role of "Invent Not Deliver" had a couple of distinct values:

• I learned what I DIDN'T want to do with my career
• I learned how to keep my head up and continually study new trends and the latest thinking

It's the last item that is particularly important, especially at EMC. The roughly forty companies that EMC has acquired this decade has resulted in a plethora of combinatorial possibilities for innovators. Keeping track of the latest thinking (e.g. private clouds) also sparks new ideas and possibilities.

All of the above puts an innovator in a strong position to deliver new ideas, because someone who invents and delivers can best articulate the path to ultimate productization.

I'd be interested in hearing other people's thoughts on innovation from within or outside of development organizations.

Steve

World Innovation Forum

There were several innovation-related topics that caught my eye yesterday. Innovation has become a huge word this year, given that many believe that a down economy is a great time to innovate.

The first topic I saw was a reminder that the World Innovation Forum is occuring in NYC starting May 4th.

The second thing I saw was this tweet from Stu's coverage of the IDC conference in Boston yesterday: 

stu: #IDC Carr says Cloud follows the Innovator's Dillema: disruptive tech comes into low end and eventually good enough to kill sustaining tech

 

Both of these items have Clayton Christensen in common. He's the author of The Innovator's Dilemma, and he's one of the featured speakers at the World Innovation Forum (speaker lineup is here).

 

Is Cloud Low-End?

 

It's hard to get the full context from Stu's tweet but I have always pictured IT clouds as "big" as opposed to "low-end". Nicholas Carr asserts (based on my reading of Stu's tweet!) that cloud technology is disruptive because it enters the low-end and will eventually kill "sustaining" technology.

 

I dug around the internet and pulled up an article by Mr. Carr (November) which I believe provides more context. It defines cloud as a "utility" completely hosted outside of a corporation (a la Amazon). The IT outsourcing model describes employees performing computing and storage tasks for "pennies on the dollar", which I interpret to mean "low-end".

 

This of course is a different description of cloud than what I've been thinking (and blogging) about: private clouds. This is also not surprising, given the nebulous nature of the word cloud. For me, private clouds are not low-end products but rather large aggregations of virtualized servers, networks, and storage.

 

Is Private Cloud Innovative?

 

I think it's an interesting intellectual exercise to look at the agenda and workshops occuring at the upcoming World Innovation Forum and ask if Cisco's UCS and private cloud vision are innovative. What are the sort of things discussed at the Forum? Here's a list:

 

• Learn what an innovative culture entails and how to implement it in your organization
• Learn what successful leaders do to support innovation
• Find out why behavior triumphs over process
• Gain a deeper understanding of the nature of innovation
• Learn from real innovation practitioners

There you have it. Announcements aren't innovative, but the organizations that implement them can be. From the list above, I like "find out why behavior triumphs over process" and "learn from innovation practitioners". Executives can put an innovative process in place, and companies can make innovative announcements, but ultimately innovation occurs in the trenches.

Innovative behavior coming from the trenches is what matters.

Then again I may be a little biased. ;>)

The True Measure of Innovation == JUICE_FLOW

If I were to ask myself the question "Is Private Cloud Innovative?", I would respond with a resounding "oh yeah!", because the opportunity to work on new types of problems, especially this type of problem, gets the creative juices flowing.

Don't get me wrong, private clouds are being implemented in data centers already. The Forrester report is a current survey of deployed private clouds.

But the long term vision of a private cloud, in my mind, has huge numbers of identical beefy servers, large banks of JBOD-like tiers of homogenous storage, and a ton of cool software tying it together. Much of the storage software will be running as Virtual Storage Appliances (VSAs), a topic I've covered before (shipping storage files as opposed to storage hardware).

The "cool factor" for me is that designing a total solution goes beyond storage systems and intersects with a set of products EMC ships that are somewhat outside of my day-to-day responsibilities. Examples include SMARTS, Voyence, and Infra, for example. I know tangentially that SMARTS can analyze a highly-complex alert storm coming from cascading network errors (will be common in private cloud) and conclusively diagnose the "real" problem.

In other words, innovators have to bring different technologies into adjacent areas that are, though related, not the intended initial use case. I like it. Lots of cool work ahead.

Comments about private cloud and innovation welcome.

Steve

Innovation vs Creativity

Last November must have been innovation month in the blogosphere as a variety of the storage blogging crew weighed in on "innovation". I found the discussions to be thought-provoking and indeed have continued to think about the issue since that time. Discussions, for example, were found here, here, and here.

The reason this topic interests me so much is that my career has been guided by continual transfers to new teams that I find "innovative". Starting over has disadvantages (lack of visibility) and advantages (lack of visibility [as in I don't have to wear a suit and tie]).

After the blogo-banter about innovation, Stephen Foskett proposed his list of Top Ten Innovative Enterprise Storage Products, and Marc followed with his own.

CLARiiON made Stephen's list. I worked on the CLARiiON microcode for about seven years. I agree with the assessment that CLARiiON was (and is) innovative.

But was I?

Private Clouds and Storage Plumbing

I continue to muse about the private cloud vision and how it's altering the building of products for the information storage industry. At the same time I've been learning about heterogeneous storage layering by way of reading (and watching) the storage blogosphere.

One topic I learned about regarding heterogeneous storage layering was NetApp's "decrease primary storage guarantee", which involves putting a vSeries in front of third party storage. Chuck doesn't get it, SteveD likes the marketing. StorageBod also weighed in.

Another topic was Tony's multi-media series describing Hitachi virtualization, in which Hitachi storage can also front storage from third party vendors. It reminded me that The Storage Architect recently wrote a very good post describing the logistics of executing Hitachi virtualization.

Each approach, as pointed out by StorageBod and the Storage Architect, involves a significant degree of storage plumbing by IT staff. It's no small investment to connect heterogeneous devices.

The private cloud paradigm is meant to eliminate this investment.