Wednesday, August 31, 2005

Pros and Cons of Open Source

Suppose you need a breakthrough product, fast. What kind of team do you want working for you?

Readers of my blog probably consider such problems fairly often, but rarely in the context of cookie baking. For that perspective, I recommend you turn to this week's "Food Issue" of the New Yorker magazine, where Malcolm Gladwell chronicles Project Delta (aka, "The Bakeoff"), instigated by visionary foodie Steve Gundrum.
Gundrum, who runs Mattson, a leading food R&D firm, decides to commission three competing all-star teams, each aiming to produce the next super-cookie, both decadent and healthy. The three teams are explicitly cast in different molds:
  • Extreme programming--relying on two complementary experts in partnership
  • Hierarchical R&D--relying on a senior manager who directs a group
  • Open source--relying on virtual collaboration of a "dream team" of fifteen nationally renowned all-stars
Though the open source dream team is heavily favored, it does not win. In fact, it almost never gets a recipe assembled in the first place, spending most of its time debating what kind of cookie to create, rather than actually creating it.

When does open source make sense and when does it not? Gladwell argues that open source makes sense when the goal is clear (a new Unix), but open source produces too much creative friction when real innovation is called for (a breakthrough cookie recipe).

For a more formal treatment of the merits of open source, I refer you to the Open Source Initiative, and speficially this internal report by MITRE: "A Business Case Study of Open Source Software."

In the meantime, keep an eye out at the grocery store for strawberry cobbler cookies--the winner of the bakeoff. Yum!

Tuesday, August 30, 2005

Join the Flock

As I mentioned just a few days ago, there's a new blog you might want to check out. Mike Prescott's SocialSpace reports on social network analysis & software, knowledge management, and business.

Mike tipped me off to Flock and I am happy to share that bit of buzz with you. It's the latest hot demo piece of social software that the uber-cool get to preview while the rest of us drool over the screenshots. You can read the buzz yourself at the socialsoftwareweblog, which caught it from Roland Tanglao, who got it directly from the source at

Rather than report on the software (which I don't have anyway) I'd like to tip my cap to the viral marketers of the world and remind the rest of us to read the NY Times' unsettling account of why their methods are so successful.

Thursday, August 25, 2005

Goings on in the social capital network blog world

I have been relatively distracted from the blogosphere lately, but have been paying just enough attention to notice a couple developments worth passing along:

If you haven't already made a habit of looking at Centrality, I recommend you start now. Good things have been happening there lately. Visible Path has brought on four notable contributors: Steve Borgatti, Danah Boyd, Nosh Contractor, and Rob Cross. That's in addition to Stanley Wasserman and Stowe Boyd. The next notable addition I suggest to this site? Outgoing resource links. But then again, "It's not who you know, it's who knows you," so technically speaking it's hard to fault the blogging strategists at Visible Path.

Also, this just in: Check out socialspace, which stated purpose is "to discuss dynamic networks, their impact and the use of social network analysis to understand them better." It's far too early to tell where this blog is headed, but at least it's got Connectedness on its initial blogroll (which, thanks to SiteMeter, is how I found out about it).

Monday, August 22, 2005

Read my blog, stay employed

The great thing about fields like social network analysis is how they combine diverse disciplines like computer science and anthropology. Besides being inherently interesting, a diverse (but still integrated) skillset also provides anti-offshoring protection.

So says today's NY Times, in this article: "A Techie, Absolutely, and More." The article gives quite a bit of press to computer supported cooperative work (CSCW). For one expert's opinion on how SNA relates intimately with CSCW, see Danyel Fisher, who recently co-authored this nifty paper on using SNA to perform email triage more effectively. Judging from the co-authors on this paper, perhaps we can hope to see this feature in Outlook 2006?


PS: I think the word "triage" is really up and coming, closing fast on "dyads." They make a good ambiguous pair, "triage dyads" being either a sorting of my email relationships in Outlook 2006, or a best-practice-swapping exercise of emergency room interns.

Tuesday, August 16, 2005

Gossip is good--pass it on

Today's New York Times has a big article on gossip--a rather scholarly piece in the Science Times. It includes the latest findings, like this:
"There has been a tendency to denigrate gossip as sloppy and unreliable" and unworthy of serious study, said David Sloan Wilson, a professor of biology and anthropology at the State University of New York at Binghamton and the author of "Darwin's Cathedral," a book on evolution and group behavior. "But gossip appears to be a very sophisticated, multifunctional interaction which is important in policing behaviors in a group and defining group membership."
Perhaps my readers will agree that this news has been making the rounds for a while? In any case, I did also love this quote:
Given this protective group function, gossiping too little may be at least as risky as gossiping too much, some psychologists say. After all, scuttlebutt is the most highly valued social currency there is. While humor and story telling can warm any occasion, a good scoop spreads through a room like an illicit and irresistible drug, passed along in nods and crooked smiles, in discreet walks out to the balcony, the corridor, the powder room.
To delve deeper, see this APA article. Also, if you hear any scuttlebutt on "Gossip in Organizations" from Computational & Mathematical Organizational Theory, pass it on!

Social Network Analysis in an Imperfect World

One of the rules of thumb you hear in SNA circles is that you need 80% response rate to have a reasonably valid survey. Having used that rule for a while without asking why, I am now glad to see some explanation. A new paper in press by Borgatti, Carley, and Krackhardt shows how different measurements of network centrality degrade with imperfect data.

For example, suppose you want to know who are the top ten influential people in an organization. Naturally, you decided to measure influence with social network analysis. The problem is that you don't get all the data. What are the chances that any of the actual top ten influencers are included in your calculated list?

With a few wonderfully simplifying assumptions, BCK comes up with a rough and ready answer. If you are missing just 5% of the network data, chances are that your list of ten influencers has three rogues in it. Let 10% of your data slip through your fingers and your list of ten is probably just better than half right. If you're missing 25% of your data, then most of your "top ten" list is really just a selection of the relatively central influencers, and not a top ten at all.

For anyone fond of the veneer of precision SNA puts on fuzzy questions, these results are an equally precise grain of salt. For another dose of context, see here too.

Friday, August 12, 2005

Community blog at the Mass Bay OD Learning Group

Intrepidly following in the collectively blogged footsteps of UVScene and H2otown, I have banded together with Mal Watlington to launch a new community blog for the Mass Bay OD Learning Group. (And, no, that's not pronounced "odd learning group," as someone asked me yesterday. Rather, it's Oh-Dee learning group, for organizational development.)

Our new community blog is poised for take off. I encourage you to check it out and read about the fantastic ODLG workshop led last night by Barry Oshry. Never before have I been so swept into the drama of an organizational simulation. Perhaps that was the point--to make all the participants so crazy that our only hope for sanity to sign up for one of Barry's more official work$hops.

Wednesday, August 10, 2005

SNA used to capture Saddam Hussein

Many thanks to David Koelle at Charles River Analytics for sharing this headline with me: "Sociological skills used in the capture of Saddam Hussein." (Published just last month in Footnotes, the newsletter of the American Sociological Association.)

The article profiles the impressive work of of Major Brian J. Reed: "He reports using a layered social network analysis to locate Hussein prior to his capture. 'The intelligence background and link diagrams that we built were rooted in the concepts of network analysis. We constructed an elaborate product that traced the tribal and family linkages of Saddam Hussein thereby allowing us to focus on certain individuals who may have had (or presently had) close ties to [him],' said Reed."

So look out, all you evildoers. You can run into your structural holes, but you can't hide.

Thursday, August 04, 2005

Why we hate HR

This month's issue of Fast Company features a great cover story, "Why We Hate HR."
Whether you agree or not, chances are you'll find this article very thought-provoking. Briefly, it indicts HR on four counts: (1) HR people aren't the sharpest tools in the box, (2) HR pursues efficiency in lieu of value, (3) HR isn't working for you (the employee), and (4) the corner office doesn't get HR (and vice versa).

Personally, I love the article. And so just in case you're put off by its blatant hostility, I would like to take this opportunity to direct you to some related (and utterly civilized) items worth reading.

First, What Gets Measured Gets Done: Developing an HR Scorecard. This piece bemoans the disconnect between HR and business strategy (essentially points (2) and (4) above) and outlines a balanced-scorecard type of approach to mediate between the two.

Second, Watson Wyatt's Human Capital Index. Here we see intriguing evidence showing human capital as a leading indicator of value creation. In other words, good HR drives financial results more so than the other way around.

As interesting as that headline is, I am much more struck by the body of the Watson Wyatt report. They break down human capital into six sub-categories, each of which is further itemized, and then discuss the return on investment of HR effort put into each category (and each item). Most HR objectives correlate with increased market value (showing flexibility in work arrangements, for example). But some popular HR objectives correlate with decreased market value. Quoting from the report:

"We found that three practices in particular---360-degree review, longer-term developmental training, and implementing HR technologies with "softer" goals in mind---were often associated with a decrease in market value. Our hypothesis is that, while there is nothing inherently wrong with these practices, many organizations implement them in misguided ways. In these areas, companies must pay special attention to strategic alignment and appropriate execution."

They go on to say, "Multisource feedback continues to enjoy mass popularity, and many, if not most, businesses report that they feel it is successful." And yet evaluating superiors and evaluating peers are linked to a significant decrease in market value. "The truth is that it is a challenge to get multisource feedback right. It succeeds when an open culture is already in place. It succeeds when participants have been well trained to give and receive feedback. It succeeds when there is valid and reliable intrumentation and appropriate follow-up. When one or more of these elements is missing, multisource feedback can be a lengthy distraction that interferes with teamwork and reduces productivity and, ultimately, shareholder value."

Monday, August 01, 2005

Centrality: a social network view of leadership

Last Friday I visited with Charles River Analytics, a cheery bunch of PhDs who build DoD-sponsored software prototypes. After my talk, "Centrality: a social network view of leadership," we compared notes on leadership in the context of adversary modeling and leadership in the context of organizational consulting. Funny how such different desired ends can still lead to the pursuit of similar means.

If you want a copy of my slides, just email me and I'll send them your way.