Picture the scene. It is your first day in a brand new job. Your role is
that of information architect in a company with lots and lots of
information that needs managing. You have a sizeable budget, enough
people and (bliss!) board-level backing to take a green field approach
to implementing an EIM infrastructure.

At the start of the week you are understandably pumped, psyched and
ready to blaze a trail in your new role. You take a look at the
information assets that need to be managed. You build a conceptual
model of the information universe. You harden that model into a
collection of information structures. You further harden them by
selecting products/tools that support your chosen information
structures. You get all that done by Friday and head home for the
weekend, breezing through reception leaving an odor of supreme
confidence in your wake. You put you best sunglasses on and climb into
the company car. You tap the steering wheel to the beat of your favorite
music. Life is so good.

It's important to enjoy such experience when they come your way as they
are few and far between and, in my experience, they last one week on
average. From there, things can get real messy, real fast. Welcome to
the world of information management.

In the full flush of irrational exuberance that typically accompanies
the early stage of information management projects, effort centers
around picking the perfect structure for the information assets and the
workflows they traverse. Roughly speaking, there are three choices:

1) Put the information into a linked set of tabular structures - the
database approach.
2) Put the information into a linked set of hierarchical structure - the
XML approach.
3) Put the information into a linked set of freeform documents - the web
pages, word processor approach.

Unless you are very lucky, you will not be long into this exercise
before you find that none of the above are perfect for your needs. In
fact, if your experiences mirror mine, you will find that the closer you
look, the more exceptions you come across. Exceptions that add more
warts to your information model. Over time, the pristine edifice you had
in your head at the end of that first Friday, has metamorphosed into a
more carbuncular structure. A structure adorned with compromises and
"temporary" fixes. A structure held aloft by constant minding and the
collective telekinetic wills of the minds involved in its creation.

Here is a tip for calming your soul if this scenario rings a bell with
you:

There is no such thing as the perfect information model.

While I'm at it, here is another tip:

No information model survives the first encounter with real information
in the real world.

The real trick of information management is to recognize that all models
are compromises and that all information structures must constantly
change to reflect the constantly changing real world.

A tall order and not a problem I have a solution for (sorry). I do
however have some suggestions.

1. Do not knock yourself out looking for the perfect model. It doesn't
exist. Spend an appropriate amount of time, then stop, make a decision
and move on. The world will move on, whether you like it or not.

2. Do not castigate yourself when it turns out imperfect. That
perfection is unattainable.

3. Think of whatever structure you choose as being simply the physical
manifestation of the information from which you can derive any number of
logical views. At the end of the first Friday, when you think your model
is perfect, your logical view matches your physical view. As the weeks
go by, the physical view stays the same, but you can add and rework as
many logical views as you need to in order keep pace with what the real
world foists upon you.

4. If a particular logical view ends up predominating, you may want to
transform your information assets to reflect that. It then becomes your
new, improved physical view from which a new generation of logical views
will be derived.

There are a number of technologies/approaches that help in the practical
application of this idea of constantly evolving information structures.

There are various technologies that facilitate the creation of an
"information facade" - a way of treating a logical view of an
information asset as the physical view thus insulating applications from
the carbuncular structures that develop at the physical view level.
These are particularly powerful in the case of tabular structures from
databases as they allow you to move significantly beyond the concept of
"view" that is typically provided in database systems. Not surprisingly,
in the times we live in, these views are often exposed as XML[2].

Also of note is XFML[1] - a notation for faceted classification of
information assets. This meta-data driven approach allows you to create
a variety of tabular and hierarchical views from the meta-data attached
to your information assets.

The next time you are having one of those blissful first weeks in an
information management role, you might like to factor some faceting
technologies into your thinking on day one. It will repay itself in
weeks two, three, four . . .

[1] http://diveintomark.org/archives/2002/12/03.html
[2] http://www.infozone-group.org/prowlerDocs/html/proposal.html

http://seanmcgrath.blogspot.com

» posted by ITworld staff

ITworld

• Email
• Print
• Share
  • Slashdot
  • Digg
  • Stumble
  • Reddit
  • Newsvine