Sunday, June 22, 2008

I've been thinking about entrenchment.

Entrenchment is part of the real world. Although most intelligent people are inclined to attempt to avoid or defeat it, entrenchment deserves better practical consideration.

Here is some background on that.

As a class of literature books on leadership give readers not only advice from someone claiming authority but also the vicarious experience of being an actual leader without the attendant responsibilities. Some even bear a close resemblance to practical philosophy. Finally, the information is useful to non-leaders seeking to better manage the potential effects of leadership initiatives.

A common leadership scenario played out in these works is the foisting onto an organization of change. Business slogans about accepting change are their own separate topic, but for here it is enough to say that resistance to change often should not be surprising.

Entrenchment in this context is seen to be the natural enemy of change. It is important to realize that both are equally mental abstractions, characters in the narrative. Where change in a business context can refer to layoffs, reorganization plans, productivity improvements, and such like, entrenchment refers to the culture and practices of people who represent "the old way of doing things." It goes without saying that entrenchment is regarded as bad for organizations, especially from the perspective of change-minded individuals.

Most of the time entrenched practices are the life's blood of organizations. When nobody is trying to institute changes, those people and practices are just fine for day-to-day operations. When painful "change" is instituted it is usually the work of someone seeking individual accomplishment at the expense of the stability and comfort others work years to achieve for themselves.

*** Somehow I imagine I have a thesis here ***

(Discuss the social contract and division of labor, two classical sociological themes, for want of a more hip gnosis.)

Leaders and Followers

Several points back up this thesis. First, looking to academic sources, the "social contract" associated with Rousseau and articulated by Levi-Strauss in Tristes Tropiques is at the basis of all leadership: that is, leadership is a dualistic construct requiring followers or else the concept is empty. The business context at times ignores this reality, attempting to convince everyone that he or she is a "leader" in order to trick them either into perpetuating the status quo, or embracing some new change, in reality turning them into willing if misguided followers. In either case the point is to encourage followers to identify more strongly with their work. For those individuals this may be leadership in some sense of the word, but the fact remains that the organization remains unequal in terms of the sacrifices it requires and the rewards it distributes. Leaders have more potentially to gain and to lose than the people they lead (although maximizing gains and minimizing potential losses is a part of the work of leaders, sometimes disproportionately so, as I hope to dicusss), and the social contract must be upheld, either well or poorly.

In the context of organizational change entrenchment is seen as an opposing force. In other contexts, entrenchment has additional defining characteristics. An entrenched person is someone with more seniority than authority who acts out of self-interest instead of "for the good of the organization." For example, as a visitor in an organization, if I ask for directions to a particular office, the organization-minded person will tell me correctly how to get there, and possibly provide other pertinent information, such as whether the office is closed when I might go there. The entrenched person will send me far afield, and will likely fail to provide additional help I might need. Entrenchment is generally considered to be the perpetuation of bad practices or habits in the face of better alternatives.

This implies much in view of the social contract and the division of labor. First, the entrenched person is by definition a follower in relation to the organization as a whole, and the organization-minded people are by definition leaders, whether or not this indicates their position in the organizational hierarchy. Second, whatever the power relationships are, the entrenched person's practices are intended to increase personal localized power at the expense of generalized integration with the system as a whole.

If a boss asks an employee to bring coffee to a number of meeting participants, the employee perpetuates the power relationship (see Giddens) by fulfilling the request quickly, accurately, and cheerfully. However, the entrenched employee takes a more nuanced approach to such an instruction, neither validating nor upending the power relationship completely. The coffee may arrive slightly late, or without enough cups, or be brewed poorly, or be provided at the expense of work on a high-profile project, depending on the agenda of the employee. An entrenched response always promotes the individual agenda of the employee at the expense of the efficient running of the organization, or, seen differently, at the expense of the power advantage of the boss. I will argue later that organizational efficiency is always in the eye of the beholder.

Within complex organizations, it is equally possible that the organization-minded person occupies a position subordinate to that of the entrenched person. The Peter Principle clearly articulates the tendency of people to be promoted to just beyond the highest position they are capable of holding well. In Dilbert, Scott Adams famously lampoons bosses who behave even more dysfunctionally than their employees. In fact, in The Dilbert Principle, Adams reworks Lawrence Peter's earlier work to claim that because it is easier to promote an incompetent person than to fire him, organizations are constantly disseminating incompetence by promoting incompetent people to management. Real bosses in real organizations, though seldom as thoroughly incompetent as Dilbert's boss, are equally and sometimes more capable of entrenched practices than their employees are. These include obstructing employees' work, withholding rewards due employees, giving instructions that contradict previous directives, assigning impossible workloads, and the list goes on.

To sum up, anyone in an organization can practice entrenchment. Entrenchment is a function of the social contract between leaders and subordinates. It is characterized by inefficiency, circuitousness, complication, by a nuanced subversion of power relationships to the benefit of the (entrenched) individual, and most importantly by an adverse relationship of the person perceiving the entrenchment to the person practicing or representing it.

Although it gets a bad name from those who wish to introduce changes to organizations, entrenched people and practices often are worthy of closer scrutiny by those who assume the mantle of leadership.

Tuesday, June 03, 2008

Holy Shit - What Have Programmers been Doing for the past 6 Months??

Today I was annoyed again by the recurring sign of my own decay: comma-delimited database fields.

Now I know that parsing lists was my favorite thing to do for a few years, which I've left behind in favor of building a lightweight and cheap library of code to do what to my limited view were the most fundamental of programming functions. All this was premised on a super-old-school VB-SQL paradigm, so when I see non-relational conventions like the comma-delimited data fields it annoys me greatly that I didn't take what I was doing more seriously and actually learn something.

Today I wrote a SQL-ish query full of conditional statements to create an ugly matrix of values using an ID field and one of these lists. It worked well on the first try. It was easy to write albeit repetitive, but all around so ugly I felt I must have descended into a place so bad I had no choice but to kick my brain's ass into some kind of action.

Out of laziness and ignorance I'd blamed the rise of shoving lists into database fields on xml, so my first thought was to add a couple functions to my dormant code library to handle tagged text, figuring it a logical first step toward xml. Now I realize I'm glossing over a lot of things, but I can admit that the friends of xml are ajax, ajax's friend javascript, and their old friend html. I've always claimed to know html a little bit, at least as far as being able to write a page that would run on every browser since 1992. But we never got around to exchanging friend requests, possibly because social networking has only been a software paradigm for a short while, and html and I have been out of touch for a longer while now.

When I got home to look at my e-mail one of the bulk feeds I routinely delete but sometimes read had a thing about opensocial, so I checked that out. Listening a while it seemed a step beyond even AJAX. I'd noticed that IT Conversations and like other places had made social networking a topic for a few months now, and I knew about social apps and waited out Web 2.0, but only today did I realize what was going on: why Facebook is so full of bullshit apps that all seem to be ways of realizing infantile interactions "I (insert verb) you," "I (insert verb) him/her," "So-and-so (insert verb)ed him-0r-her." Apps are today what photos were before, when the rise of flickr and countless other web 2.0 photo sharing apps revolutionized online communication.

The next big thing is clear: people will publish and exchange stupid little apps the way they had once published and exchanged stupid photos of their babies, dogs, vacation scenery, and predictable whatnot. And those apps will all do about the same thing: "I (insert verb) you."

"I (verb="poke") you." "I (verb="bite") you." "I (verb="tag") you." "I, through my virtual monkey avatar, throw monkey shit at you." This is the brave new world of the next several months. If I know about it, maybe it's already done. "You (verb="stick-a-fork-in") me."

v1=fDeathSpiral()

So anyway, the basics of OpenSocial are Identity (i.e. me), Friends, and Activities, which are generic interactions between Friends (including me). Simplifying everything this way will let OpenSocial be adopted by every social networking capital farm. And complex interactions can be deployed from such a framework as the need arises. "I (verb="castigate") you to (my friends + your friends)."

And the promise of it (if any) is that lightweight object-based apps will float around the web, serverless, linking friend to friend through botlike avatars, and allowing all interactions to proceed through a simple A X's B syntax. And thus the web will transform itself from mere infrastructure, easily mistaken for a bunch of tubes, to a digital analog of human reality.