Feed aggregator
pernicious relativism and logical positivism
Just what is pragmatism? And why does this approach to work bring so many of us into conflict in the workplace?
It's been called 'pernicious relativism' by some and logical positivism by others. It seems to me that pragmatists are about getting on with things, but we operate in a world of people who seek 'one best way' to get things done. Maybe there is one, maybe there isn"t.
Maybe the important thing to do is try something and improve on it. And here we see the link between pragmatists and empiricists. As we start to deal with empriricism we start to migrate back to the idea of a theoretical "one best way"
Once again I find myself in need of the consultant's quadrants.... Pragmatism and philosophizing. Theory and pragmatism aren't opposites, they're orthogonal. And just where is the sweet spot between theory and pragmatism?
APSA 2010 Discussion
Unfortunately, I am not be able to make it to the American Political Science Association’s annual conference this year, starting today in Washington, DC. That said, there are many other political science bloggers hitting the panel circuit, of note: Dan Drezner, the Duck of Minerva crew, Seth Masket, and presumably several others.
While there are too many interesting panels to enumerate here, I would like to especially highlight the new Political Networks section, formed through the hard work of Scott McClurg. On that note, if there were one panel I would be sure not to miss, it would be “Political Networks: Structure, Interactions and Outcomes,” which is chaired by McClurg and includes papers from several of the top political network researchers in the world.
I would love to hear other’s impressions of the conference, what papers they really enjoyed, etc. If you are checking your RSS reader while in DC and would like to share, please do so.
Good luck to everyone presenting!
Definition of Construction Management
Developing definitions for context sensitive processes such as project management is difficult [see paper]. Over the last year I have had involvement with the efforts of the CIOB UK to develop a similar type of definition for Construction Management. The final definition is rather long, but is I would suggest both accurate and concise.
Construction Management is: The management of the development, conservation and improvement of the built environment; exercised at a variety of levels from the site and project, through the corporate organisations of the industry and its clients, to society as a whole; embracing the entire construction value stream from inception to recycling, and focusing upon a commitment to sustainable construction; incorporating a wide range of specialist services; guided by a system of values demonstrating responsibility to humanity and to the future of our planet; and informed, supported and challenged by an independent academic discipline.
The result of this work is now available for download and sets the benchmark for project mangers to aspire to achieve [download the CM definition].
Managing expectations
You aren’t really in control of a person’s expectations. Despite all the effort you put into setting realistic expectations about your team’s capability, or the difficulties of the problem you are tackling, people will still be influenced by other extraneous factors.
They’ve seen all their corporate projects run over budget and delivery disappointing results constantly. They’ve seen Facebook become an internet behemoth in less time that it took for their billing transformation project to get to it’s first release. Or they’ve simply gotten used to using Microsoft Powerpoint 2010 and expect all new systems to have similarly advanced features.
People’s expectations don’t match reality. No matter how hard you try to convince them.
What can you do? Function Points? Story Points? Planning Poker? Incremental releases? Rolling waves? Abandoning estimates altogether?
What’s your plan?
April 19 -21, 2010 – Conference: eComm 2010: Emerging Communications
I spoke at the eComm 2010 conference on April 20, 2010, talking about:
Mapping mobile social networks with NodeXL: finding key users, groups, and locations
The video is now available:
The video reviews the creation of maps like this:
That illustrate the connections among people who tweet the term “#ecomm2010″, scaled by the number of followers.
Abstract: Social network analysis (SNA) is a powerful method for gaining insight into the massive collections of connections created when many people connect to one another through mobile devices. SNA has been widely applied to desktop social media and is moving into the mobile world. Prominent studies of the “call graph” have been produced at national scales.
Mobile providers are applying SNA to identify key subscribers who can reduce churn and help gain adoption of new services and products. Network analysis has historically had a steep learning curve, but now new tools are making SNA easier for less technical users. This talk will describe social network concepts and their application to mobile data sets. A free and open add-in for the popular Excel 2007 spreadsheet called NodeXL (http://www.codeplex.com/nodexl) can perform many complex SNA tasks like data import, scrubbing, metrics calculation, clustering, and visualization. Applying this tool to call graph and subscriber data sets can reveal key positions in the network that can attract and hold other subscribers in the system.
Examples of network analysis of social media and mobile data sets can be found on the Connected Action blog (http://www.connectedaction.net).
Share and Enjoy:COTS Selection RESOLVED (You Make The Call)
Square One
The first suggestion on improving the selection of a COTS product is the same as any internally developed application, namely a better understanding of the goals of the project. While your stakeholders feel that they could see labor and customer satisfaction increases with a new system, it wasn't clear (intentionally) from the description that the stakeholders wanted a COTS package, only that the CIO preferred one. Was there a bias on the part of the CIO towards using a COTS package or did she think it just made more financial sense to go that direction? Asking additional questions during the initial discussion with the CIO would have been a good first step.
Next, why was a list of current system functionality given to the vendors? The CIO made no mention of wanting to try and replicate the existing system in a COTS package. If the users main complaint is speed and older technology, could the existing system have been enhanced to meet those needs quickly? The stakeholders did not say they were unhappy with the system as a whole, just those two aspects of it.
But a more fundamental question is what do the stakeholders need? If the existing functionality works for them, with the requested speed improvements, then it doesn't sound as if they were necessarily unhappy with what the current system. If speed and old technology were ways users were expressing their displeasure with a product that was in reality more deeply flawed, only a rigorous review of the business processes in comparison with the system processes will find the underlying problem.
At its core, not enough questions were asked. The price tag returned by the COTS vendors were set very high as recreating an unknown system on top of their product is not usually cheap. Such a high price tag is a method vendors use, and rightly so, to provide incentive for clients to reassess their request and frame the conversation in terms of actual needs and not just a functionality list.
What other thoughts do you have on this situation? Have you been in a similar place yourself? Let us know what you did in the comments.
Introduction to NetworkX at NYC Python Meetup
As many of you know, Aric Hagberg and I presented a half-day introductory workshop to NetworkX this summer at the Sunbelt conference in Italy. Given the cost and difficulty of actually getting to Riva del Garda for the conference (Aric and I can each attest to that), I understand that attending the workshop was impossible for most people.
Now, for those who live in the NYC area you have a second chance!
Next month I will be giving an abbreviated version of the workshop for the NYC Python users group, and I wanted to pass on the invitation to any readers in the NYC area that might be interested.
Registration doesn’t open until September 7th, but the talk is the 28th for those that want to mark their calendar. Should be a fairly small group, so there will be lots of opportunity to interrupt me and ask questions.
I apologize to everyone else who were unable to attend Sunbelt and do not live in the NYC area. At least you can still download all of the materials.
Project Phoenix, Something I Must Share
That project also had what was probably the most contentious vendor relationship, with the implementer not the software vendor, that I have ever seen. There were days that I, as a tester, refused to even enter the test lab as the environment was so toxic and cancerous. I still cringe to think about some of the acts in which I participated. We had a plant in the corner of the room that must have been fed a constant diet of Miracle Grow as it lengthened nearly before your eyes. The team joke was that we tossed it a consultant every other day in order to keep up such a pace. Definitely not the way to start out your life in projects.var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E")); try{var pageTracker = _gat._getTracker("UA-xxxxxx-x");pageTracker._trackPageview();} catch(err) {} var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));try {var pageTracker = _gat._getTracker("UA-4191072-4");pageTracker._trackPageview();} catch(err) {}
All that said, today's comic from 1.00 FTE brought much joy to my heart. If you've ever been involved in a Project Phoenix, then this one is definitely for you. Let us know more about your Project Phoenix in the comments.
Of Builders' Tea
My Readers are Awesome
The title says it all.
I am absolutely floored by your generosity in helping me build my bookshelf. To pay homage to your wonderful philanthropy, I have created the Bookshelf Hall of Fame page above, which lists those individuals who have been so generous in donating books to further my education.
Finally, I would like to individually thank Yuta Kashino and Stephen O’Connell, as they were the first contributors. I have actually been only slowly able to write this short post because I keep looking down and reading the books they donated!
Again, thank you.
Erasing your Project Identity
I find this concept novel, but think that as naive as we are as high school and college students, eventually the pervasiveness of technology in our world will make it so that this lesson is learned soon after we are taught that the stove is hot and the backyard pool is dangerous. It won't be something we generally worry about because we'll learn what to post and when to post it, just like we'll learn reading, writing and arithmetic.
But if that's true of our personal lives, what about our project identities? Once, and likely still today, each of us has a file in our manager's desk drawer that contains all of the past reviews, time off requests and possibly even official sanctions and reprimands. As all of this information recorded on dead trees transforms into data on a server in the cloud, what safeguards are in place to keep it confidential in today's world where top secret government documents can be removed on a CD labeled 'Lady Gaga'?
Personnel records are not the only items kept electronically; our project documents are almost always kept that way as well. Sure, you likely have a stack of signed capital request forms (CRFs), project charters or requirements documents moldering in a corner drawer or cabinet somewhere, but every revision of that document is likely also contained on a server or possibly your PCs local hard drive as well. What happens when that first requirements document you wrote, you know that one with a 5% misspelling rate because you forgot to spell check it, with grammatical errors in every other sentence and with completely non-standard project language used, is put into the hands of your new CIO as an example of your 'regular work'. Is there a way to have your project history reset?
Reset #1, Don't Make Mistakes
Probably the most unreasonable and unlikely option, I mention it first to remove it from discussion. Sure, we've all met that one project person who seems to have the Midas Touch and never is caught making a mistake, but its rare and unlikely to last for more than a short time.
Reset #2, Different Scenery
The first way to reset your project identity is to simply switch employers. Having worked for several different large corporations in the last few years, those original requirements documents I wrote as a junior BA are (thankfully) buried on some shared drive at my first employer. Sure, I have a copy of them on a CD in a desk drawer, but they don't live anywhere they can be accessed by anyone but me.
As I've progressed in my career, I've still had a few less than optimal requirements documents. Different reasons, such as lack of time to thoroughly vet the requirements or lack of review on the part of my stakeholders, is to blame for these unsightly blemishes, but they are part of my permanent record with each of my employers and the documents have me listed as the author, not my stakeholders.
Reset #3, Bury It
Given that we're all going to make mistakes and that we can only realistically move away from so many jobs before we earn a reputation as a short-term employee that isn't interested in sticking around, we can start to come up with ways to hide our mistakes.
The best way to bury your bad output is by producing a lot of quality work now and in the future. Producing excellent quality work will go a long way in removing any stigma associated to your past failures.
Moving Forward
Some of the methods I've mentioned will be effective in the short term while others will produce success for the remainder of your career. What are some other ways you can think of to reset your career mistakes (besides large electromagnets in the data center)?
The Spatial-Temporal Nature of Terrorism in Civil War
An excellent new paper was posted this morning to SRRN by two leading scholars in terrorism studies, Michael Findley and Joseph Young, that attempts to understand how terrorism—literally—exists inside civil war. In “Terrorism and Civil War: A Spatial and Temporal Approach to a Conceptual Problem,” the author examine geo-coded terrorist events from the Global Terrorism Database in an attempt t answer the question: Are terrorism and civil war distinctly different phenomena? As the authors note in the introduction:
Our strategy is to examine terrorism in the context of ongoing civil war, prior to the onset of war, and after the civil war ends. Our approach is both conceptual and empirical. We discuss the similarities, differences, and reasons for terrorism in different contexts, the most important of which is during a civil war. We then examine terrorism across these contexts and offer suggestions for integrating the study of terrorism and civil war.
I really like the innovative approach to data in this paper, and highly recommend it eveyone.
When do you need centralized – less centralized networks?
Social network analysis gives me a language to talk about and think about things I see people do. I rarely use the network measures quantitatively (does it help to know that actor A has a betweenness centrality of 0.2?) but more as structural descriptions of what I see and to explain what might happen. Centralization for example has come up a lot in my recent work with various development projects. Centralization describes the structure of the network as a whole, a highly centralized network is a hub and spokes network, one actor in the middle with a star shaped network linking him to others who are not linked to each other. In a network with low centralization on the other hand, everyone is linked to everyone.
When you start a project, you are often the mover and shaker and connector who develops links to all other actors and you move your project forward from this position being the hub, everything happens because you are there, you are in control. We all know projects which only got off the ground because of one tireless visionary who pushed and pushed and pushed for his or her vision. Great.
The only problem is that what is one of the most effective strategies for getting projects off the ground is not always the best for making them sustainable. So while it is only human to think that “more of the same” will make us successful in the future, it is important to understand that hub and spokes networks are
1. highly vulnerable to shocks: This could be a heart attack of the visionary or the hub being overworked or corrupted by power and stopping to be the mover he/she used to be,
2. Don’t harvest the wisdom of the crowd: Because all information is filtered through the hub, people don’t get together independently to think together, inspire each other and come up with new solutions. This is one of the reasons why hub and spokes networks are especially well suited to implement simple repetitive tasks where all answers are known and these networks are especially useless in solving complex messy problems where the answers are not known and might differ between locations or situations.
3. They don’t develop the bottom-up energy to maintain themselves and grow: Because everything is always a reaction to the push from the center, everyone starts to rely on that push and waits for the center to come up with ideas, activities and funding. This can lead to the typical situation in donor funded projects: As the funding expires the activities die down, even if they were extremely useful to the people involved and it would require only little own contribution to maintain them.
What does that mean? If you are really passionate about the change you are aiming for, remember: This is not about you. This is not about your organization’s signboard or your own name attached to the change. Go in, move and shake, develop your hub and spokes network as necessary but AS SOON AS YOU HAVE THAT, start implementing your exit strategy (which might take most of your project) by connecting actors that you work with and making sure that they will eventually be able to do their work without you. Get over the fear of loosing control because loosing it is a pre-requisite for having an impact beyond what you alone / your organization alone can do.
COTS Selection (You Make The Call)
COTS Implementation
Your CIO has recently received a great deal of negative feedback about one of the applications that nearly all of your internal customers use on a daily basis to complete their work. This application was developed internally and is highly customized for that business process and your organization. Most of this feedback focuses on the speed at which users can complete tasks and the age of the technology in question. Most stakeholders believe that a solution that uses newer technology that focuses on a touchscreen instead of keyboard entry would allow for faster throughput by users, resulting in a small labor savings and a large customer satisfaction increase.
Despite the dislike of the current system, its customizations and integrations with other corporate systems make it difficult to replace with a COTS (Commercial, Off The Shelf) product. Still, your CIO does not wish to develop another system internally when there are numerous COTS products that perform similar, if not exact, functions. You are instructed to initiate an RFQ (Request For Quote) process with the major vendors who have produce applications in this market.
Picking Up The Phone
You, being a person who keeps in touch with industry trends, know the major players in the market segment and start making phone calls with those vendors. Besides the standard information most vendors want to know, such as number of users, number of locations, operating systems used, etc, you also provide a list of functionality of your current system that you are looking to replace. You explain to the vendors that it is of critical importance that their systems replicate all functionality in your existing solution and that their product be able to interface to all your other internal systems.
As the vendor RFQs trickle in, you notice a disturbing trend... the cheapest of them is an order of magnitude greater in price than what it cost to develop and maintain the existing system since its implementation over a decade ago. While you know your legacy system is very specialized to your company, you are astounded as to the large number of development hours that will be needed for every one of the vendor systems to meet your needs.
Checking In
Now you are in a dilema... your CIO expects your findings to be on her desk by the end of the week and you know she won't be happy with the price tags on the RFQs. What do you say in your report? Is there anything you can suggest to help lower the prices? What could you have done differently in the RFQ process?
The Kings Arms
In Search of Power-laws: WikiLeaks Edition
Yesterday, a commenter reminded me of the very popular hobby among scientists of searching for power-law distributions in large event data. While the commonality of scale invariance in event data is quite well known—particularly with respect to conflict data—this has not prevented many researchers from seeking and finding these patterns in data.
As the commenter notes, it is likely that the WikiLeaks data will soon be annexed into this line of research. Before other researchers examine the distributional properties of these data more thoroughly, it is worth doing a quick exploration to show some of the issues in power-law fishing, and how to avoid them. First, we begin by plotting the WikiLeaks casualty data using the traditional log-log transformation and fitting a linear regressio.
The search for power-law distributions often focuses on the scaling parameter: . Scaling parameters where 2 < < 3 are generally accepted as fitting a power-law, thus the search is for values in that range.
When using linear regression to fit the distribution this parameter is calculated simply as the slope of the linear fit to the logged data. The above panels show this analysis for two different version of the data. On the left, the data are restricted to only observations with, in the words of Lewis Richardson, “deadly quarrels.” That is, WikiLeaks events where a death occurred, which accounts for friendly, enemy, host nation and civilian deathsg. Interestingly, we find that for the KIA data the scaling parameter falls just outside the necessary range to be classified as a power-law.
At this point we might conclude that the data does not fit our assumptions and move on to test other distributions. If we were particularly motivated to find a power-law in this data, however, one option would be to go back and loosen our restriction on the data to include not just KIA’s but all casualties, i.e., non-deadly quarrels. The assumption being that with more data points the “tail” of the distribution would be longer and thus more likely to fit a power-law. The right panel above illustrates this analysis, and as you can see in this case we find that the data do fit a power-law, with .
Unfortunately, even if suspending disbelief enough to accept the altogether dubious inclusion of more data points to force-fit a power-law, everything we have done up to this point is wrong. As was brilliantly detailed by Clauset, et al in “Power-law distributions in empirical data,” linear fits to log transformed data are extremely error-prone. As such, rather than rely on the above findings we will use the method detailed by these authors for properly fitting power-law on the WikiLeaks data.
In this case we need to do three things: 1) find the appropriate lower-bound for the value of for our data, which in this case are events with casualties; 2) fit the scaling parameter with ; 3) perform a goodness-of-fit test to test whether our empirical observations actually fit the parameterization of the distribution.
For the first step we are fortunate, as we know the appropriate minimum value , since these are discrete event data and we are counting the number of observed casualties in the data. Equally convenient, this allows for a straightforward maximum-likelihood estimation of the scaling parameter via a variant of the well-known Hill estimator. This functionality is built-into R’s igraph package so we can compute the new scaling parameters easily.
KIA Only
All Casualties
MLE
1.823
1.78
Using this more accurate methods for estimating the scaling parameter reveals that—in fact—neither set of data on the frequency and magnitude of violent events in Afghanistan fit a power-law. As a result, goodness-of-fit tests for power-law with this data are unnecessary, but as described in Clauset, et al. using a Kolmogorov–Smirnov test to measure the distance between theorized and observed distributions is a useful tool for checking fits to other distributions. There are several alternative distributions that may better fit these data, many of which are specified for simulation in the degreenet R package, but I leave that as an exercise to the reader.
There are two primary things to take away from this exercise: 1) power-laws are much less frequently observed than is commonly thought, and careful estimation of scaling parameters and goodness-of-fit should be performed to check; 2) it appears that the WikiLeaks data fall well short of proving, or even reinforcing, previous conclusions about the underlying dynamics of violent conflict.
As always, the code used to generate this analysis is available on Github.
Stakeholder Management Thesis
My original thesis has recently been published as a book by Lambert Academic Publishing AG & Co (www.lappublishing.com).
Details of the book are:
Project Relationship Management and the Stakeholder Circle [Paperback]
ISBN-13: 978-3838398167
Available from Amazon at: http://www.amazon.com/Project-Relationship-Management-Stakeholder-Circle/dp/3838398165/ref=sr_1_1?s=books&ie=UTF8&qid=1282809735&sr=1-1
The research described in my thesis underpins the Stakeholder Circle methodology and tools which led to the publication of Stakeholder Relationship Management: A Maturity Model for Organisational Implementation and the SRMM maturity model available from Gower Publishing at http://www.gowerpub.com/isbn/9780566088643
Call for Proposals: 11th Public Management Research Association Conference
Blog>> Origins Gets Press
The Straits Times today had a double page spread on the Origins Conference and its parent, the Singapore International Storytelling Festival. Some interesting quotes from our speakers that have me looking forward to the event even more. If you haven’t booked your place yet, time is running out!
Straw, reed and willow
Copyright © 2004 -2010 Knowledge Matters™ - all rights reserved
The Webpages of Durant-Law Consulting Pty Limited
and Occasional Blog of Graham Durant-Law
E-mail: graham@durantlaw.info
