Ruminations on tech, the digital media, and some golf thrown in for good measure.

Archive for October 26th, 2010

Why The Big Data Chicken Crossed The Road

leave a comment »

“The organizations that are going to be the most competitive are going to be the ones that make the most sense of what they learned as fast as they learn it.”

Does that line sound familiar?

That was the voice of Jeff Jonas, chief scientist with IBM’s Entity Analytics group.  Perhaps you saw him in this IBM TV commercial:

As Jonas asks in the spot, how would you know when to cross the road if you had to use a picture of the intersection taken five minutes before?

In a blogger briefing this afternoon centering on the topic of “Big Data” here at Information on Demand, Jonas explained that organizations must increasingly make sense of data as it’s happening, while it’s happening.

Jonas and several other scientists, researchers, IBM staffers, and even a customer (from Visa) elaborated on this phenomenon during the subsequent hour.

Jonas went on to explain that “new physics is happening with big data. Errors in data actually turn out to be positive. I’m seeing how systems get faster with more data.”

Companies are using Hadoop, as well as IBM’s BigSheets/BigInsights add-ons, to scale their analytics investigations into the real-time stream to solve new problems as they happen.

Which, of course, when you’re trying to make sense of real-time information (think air traffic control, or even street traffic control), could be the thing that allows you to cross the road based on that 5 minute ago snapshot.

Does your head hurt yet?

IBM emerging technology evangelist, David Barnes, explained that he’s partnering with his colleagues in the Big Data realm to bring Hadoop to the line of business professional, driving Big Data analytics down into the lap of the small fry front line businessperson (another common theme of IOD 2010, driving analytical capability deeper into the organization where decisions can be acted on information by those most inclined to do so due to their proximity to the customer).

As a social media marketer, I salivate at the mere prospect: You mean I can hone in what people are saying as they say it, or even after the fact, amidst a data set of 1.2 million Tweets?

Uh, yeah. Bring it on!

And Barnes wasn’t all talk, he had the demo to prove it:

That’s one example of the power of big data.

IBM’s Tom Deutch explained there are other powerful use cases that are emerging, but that more organizations need to take Hadoop out of the lab and into production in order that they be able to put theory into practice and take Big Data for a test drive to address very real business problems.

So if you have to ask why the Big Data chicken crossed the road…well, the answer will likely only come with a few revolutions of your Big Data analysis, but I’m confident that ultimately the answer is he wanted to get to the other side.

Now, and not five minutes ago.  : )

Written by turbotodd

October 26, 2010 at 10:36 pm

Steve Mills Keynote: Big Data, Big Picture

leave a comment »

In this morning’s first general session keynote at IBM Information on Demand, IBM Senior VP and Group Executive, Systems and Software, Steve Mills, got right to the bottom line on how organizations can go about implementing a smart information agenda and use business analytics to help them make better decisions.

IBM senior VP Steve Mills addresses the challenges of Big Data in his 2010 Information on Demand keynote.

As always, Mills painted in broad brushstrokes to help his audience see what has become something of a George Seurat “pointillist” painting, a sea of data, a mosaic of million and billions of bits and pixels of information that is piling up around us.

It’s increasingly daunting, both in terms of size and volume and velocity, and yet is an enormous business and knowledge insight opportunity as well.

So, you can either run for the hills, or you can buck up and dive into that sea, finding ways to organize it and make sense of it all…and maybe even learn something valuable for your organization.

The world is becoming more instrumed, interconnected, and more intelligent, but by leveraging this massive amount of new information, you can create a new kind of intelligence for your business, suggested Mills.

But it won’t come without some pain, trials, and tribulations.

Mills joked about the explosion in data and real world events, nodding his head to the ever-growing (but meaningless) Twitter and Facebook stream.

What Happens In Vegas…

“Remember,” he seemed to be warning the parents of teenagers in the audience, “what happens in Vegas…will stay on the Internet for a hundred or more years.”

Of course, with 44X as much data and content being generated over the coming decade, and with 80% of world’s data being unstructured (much of it that flow from the Internet, as my friend Ron ironically observed via Twitter), there’s a huge need for a structured approach to managing all this data.

Customers are clearly wrestling with this issue: 35% of customers will look to replace their current warehouse with a pre-integrated warehouse solution in the next 3 years, and only 14% have today.

And yet 83% of CIOs cited “Business intelligence and analytics” as part of their visionary plans to enhance competitiveness.

So, the IBM approach to mastering information for the purpose of optimizing business results is to build a flexible platform for managing, integrating,  analyzing, and governing information.

This is not a random path, but rather a structured, well thought through approach that takes an holistic look at information management.  Mills acknowledged we’re living in a federated world, one with a disparate set of information sources.

That’s why the Big Data challenge requires a Big Data approach, one that can help organizations deal with and benefit from massive and growing amounts of data, that can handle uncertainty around format variability and velocity of data, that can handle all that unstructured data, and one that can exploit big data in a timely and cost effective fashion.

IBM is offering a comprehensive set of solutions for Big Data, one in which interoperability will be key to addressing the unique challenges of the big data ecosystem.

Mills concluded with a big picture statement about all this Big Data: “We’re at an inflection point where IT is going to change the world in the next decade in ways even greater than that which we witnessed over the last 50 years.”

Written by turbotodd

October 26, 2010 at 6:25 pm

Dr. Atul Gawande: Use Your Knowledge

leave a comment »

I had very much looked forward to this morning’s keynote from Dr. Atul Gawande, and he certainly didn’t disappoint.

Dr. Gawande is a Harvard-trained surgeon and writer, having been a National Book Award finalist for his book Complications: A Surgeon’s Notes On An Imperfect Science, as well as climbing up the New York Times bestseller list for The Checklist Manifesto.

Dr. Atul Gawande speaks about effective knowledge management and use during his keynote session at Information on Demand 2010.


Gawande started his keynote on a solemn note, highlighting the percentage of war injuries to death from the Revolutionary War (42%) through the First Gulf War (24%).

Only in the most recent Iraq and Afghan Wars has that number begun to drop dramatically, to 10% in these latest conflicts.

What brought about this change?

Well, simply put, it’s not unlike the same relationship between data and progress that has been so pervasive in our conversations this year at Information on Demand 2010.

Gawande explained that as our technology for killing in war progressed over history, so, too, has our technology for healing, and even preventing, traumatic injuries.

It was a curious U.S. military colonel (and surgeon) who, after sifting through historical data from our more recent conflicts, discovered the antidote.

One, rank and file troops had finally taken to actually wearing their Kevlar in the field, which they had been resisting for years.

But also, medicine was moving closer to the troops, and we were able to, through these mobile surgical field units, able to reach soldiers inside the window of that critical “golden hour” which can often decide the fate of a patient after traumatic injury.

Furthermore, those surgical units had deconstructed surgery down to its most critical elements, whereby surgeons did only those most critical path procedures in the field, then “packaged” the patient up for further surgery or procedures in Baghdad or even at a base in Germany, where they had more equipment, personnel, etc.

The field surgeon would essentially put a sign on the patient explaining to the next surgeon: “Here’s what I did.  Please finish.”

Finally, the military introduced the use of checklists to ensure that patients were receiving ample supplies of blood, checking for medical allergies, etc. — common, mundane tasks but which could be lifesavers under specific circumstances.

Those few things were all it took to drive the rate of death from 24% to 10% — so much so, that in fact, the military now speaks to the survival rate, as opposed to the death count we often heard in earlier wars.

That is truly data that matters.

As Gawande more bluntly put it, “The metrics of war are now measured in the wounded, not the dead…The whole picture has changed.”

It’s that bigger picture which was the underlying moral lesson of Gawande’s talk.  What does such a change, he asked, tell us about what we’re doing and where we’re going when it comes to data?

As humans, we lived through milennia in a world of ignorance, where we didn’t understand our physical world, our environment, and the like. Though that has changed dramatically in the last half century, where, with reference to the Iraq and Afghan wars, we can save more and more lives in war, we still often don’t know how to execute on the knowledge (read: data) we already have.

Or, we often have the knowledge to solve major problems, but that knowledge doesn’t always reach the people who need it most and who can most act on it.

Having knowledge is not the same as using it, and using it effectively.

Gawande’s message was quite simple: Our challenge in the 21st century will be to learn how to cope with an increasing amount of complexity, data, and even knowledge, and then to find the right data and act on it and use it to human advantage.

To do anything less would be to leave untreated an open and scathing wound defined by ineptitude and ignorance, and we’ve come far too far for that.

Written by turbotodd

October 26, 2010 at 5:46 pm

%d bloggers like this: