PlanetHolt.com

The remainder of my time in Boston was as excellent as the first couple days. After Kelly and I spent Tuesday together, I had the rest of the week to spend at the MIT Information Quality Industry Symposium, and then spent some more time with Kelly on Saturday. The symposium was excellent—especially the keynote speakers—and I met a number of interesting IQ professionals.

The symposium, spread over two and a half days, gathered a couple hundred Information Quality professionals in Cambridge to discuss and learn about the issues facing practitioners. I attended a number of enlightening sessions, from a theoretical talk on entity / identity resolution (determining which records in a large dataset actually represent the same individual entity, and then associating them together going forward) to the TSA’s use of IQ to make the screening process more friendly for travelers.

Especially good were the keynotes. Mr. Mike Krieger, Deputy CIO( Despite his being Chief Information “Officer,” I understand that Mr. Krieger is a civilian.) of the United States Army, spoke about the unique challenges faced by an organization on the scale of the Army. He mentioned two distinct communities that he serves. The first are the strategic thinkers, mainly located at the Pentagon, who process and analyze vast amounts of data in a bandwidth-rich environment. This processing is intended for forensic analysis and data mining to find patterns and identify, predict, and prevent attacks (especially IEDs). These users don't require real-time access to data, and in some cases may be using data over longer periods of time.

On the other hand, the warfighters on the ground in places like Iraq and Afghanistan require real-time data for situational awareness, target identification, and performance monitoring for weapons systems. In contrast to the high bandwidth available to users at the Pentagon, soldiers in the field are often connected to Army networks over extremely limited links on the order of 16 kbps, and may not have significant amounts of processing power for even that limited data.

Mr. Krieger also mentioned the challenges involved in encouraging common data standards, free exchange of data (ignoring legitimate security concerns for the moment), overcoming ‘fiefdoms,’ and the use of common terminology and metadata. The common saying “Information is Power” holds true, and overcoming people’s natural inclination to hoard knowledge continues to be a challenge for both military and commercial enterprises. In the Army, reminding people of the common mission is helpful, but it’s often necessary to show how more openness can help everyone.

The second day’s keynote was given by John Bottega, Chief Data Officer for the Federal Reserve Bank of New York. Mr. Bottega is apparently well-known in IQ circles, judging by the reception he received and the comments I heard both before and after his discussion. He was careful to state that any opinions shared were his own, and did not reflect those of the Fed, which probably allowed him to speak a little more freely than he would have as an official representative.

Mr. Bottega has worked in finance for nearly thirty years, and gave an overview of how we got to where we are now, especially with regard to data quality. Traditionally, data had not been considered an important asset—when trading desks are bringing in (for example) $100M per year, optimizations to data standards or quality improvement efforts that may save the company $1M aren’t considered worthwhile when they require time and effort from the guys making the trades.

Unfortunately, those $1M savings probably ignored some side benefits. For example, there was no organized way to understand ownership hierarchies, meaning, for example, that when Lehman Brothers collapsed, its counterparties didn’t have an effective way to sort out which Lehman companies (and there were hundreds) were bankrupt and which were still operating entities.( As Mr. Bottega pointed out, from a counterparty’s perspective, it is just as bad to withhold money from a ongoing concern as it is to send money to a bankrupt one.) Companies did not have a real understanding of their true exposure to risk, which was probably a contributing factor to the recession. It’s not that better information quality systems and procedures would have necessarily led to better decisions, but one can argue that the absence of high-quality information prevented decision-makers from making what was actually the right decision in the long term. In other words, the decisions made may have been perfectly rational and “correct” given the quality and quantity of information available. Of course, the availability of perfect information doesn’t guarantee that decisions would have been made in line with systemic stability in mind.

I was updating my Twitter account throughout the conference with points and observations, as were a few others. It was kind of neat to follow some of their thoughts on Twitter and their individual blogs, and then actually get to meet them in person. Two people actually found me in sessions they were also attending, and introduced themselves (I assume based on my name badge and probably Twitter user photo).

I spent Friday afternoon after the conference wrapped up walking around the MIT campus and along the Charles River. After taking quite a few pictures, I made my way back to my hotel and waited for Kelly. We originally planned to take a Charles River cruise Friday evening, but since the forecast included rain, we decided to wait until Saturday and got dinner at Uno’s Pizza up on Harvard’s campus. After dinner I bought some beers and we hung out at my hotel, watching Zombieland and the second half of Shrek the Third.

Saturday we met for breakfast at my hotel, and then headed out to the Jamaica Plain neighborhood where the Samuel Adams brewery is located. It was very hot at the brewery, but luckily we only had to wait about 45 minutes for a tour and tasting session. The tour was much more technical than the Heineken and Guinness tours I’ve been on in the past, and actually showed us the equipment they use for brewing production beers. (The Boston location is mainly a research and development center; their output is only sold on drought in Boston. Bottled Sam Adams comes from breweries in Ohio and Pennsylvania.) After the tour, we were escorted into their tasting room, where we got to try a number of different varieties, including one that is not yet available for sale. Quite good.

After the tour, we headed back to my new hotel, the Hilton in the Financial District, and checked in. Once settled, we got dinner at McCormick & Schmick's Seafood Restaurant, where we shared an Ahi tuna appetizer. I then had a surf and turf main course, both of which were excellent.

Suitably fed, we made our way down to the pier where our Boston Harbor sunset cruise was to take off. The cruise was sold out, no doubt because of the beautiful weather. We secured a spot on the top deck and enjoyed the sights. It was a very pleasant way to spend the evening (although I think the New York sunset cruise was better). Near the midpoint of the cruise, a man proposed to his girlfriend, who accepted. Most of the women around teared up, although Kelly and I discussed all the ways that could have gone wrong. (Examples: Awkward place for him to be if she said no, since we had no way off the boat for at least an hour… what if he or she dropped the ring into the harbor? Also, while we were skeptical of the guy’s outfit, Kelly pointed out that if he was smart, he wore clothes given to him by her so there would be no way for her to complain about them later.) The cruise finished when the U.S.S. Constitution fired its traditional end-of-day cannon blast, and we made our way back to my hotel for the night.