Tuesday, September 1, 2009

And now for something completely different...




I posted a number of tweets recently about ECM and Information Management (IM). Richard Veryard (@richardveryard) asked me on Twitter to assemble these tweets on my blog so that he could comment properly. So, here is an attempt describe and expand my train of thought in the format of a blog post (and hopefully soon get to read some good comments).

“The missing piece and root cause of most ECM woes is a lack of understanding of business context — how people and business processes use content. And therein lies the challenge: How can you determine the business context for all of the unstructured information within your organization?” - Forrester

When we perform a task, we usually do not access and use the information or tool(s) we need in isolation. We do it in a usage context; a certain place, time, and physical environment. This usage context probably contains other information items that we currently view or have recently accessed.

A lot of work, especially knowledge work, is also highly collaborative by nature. To perform a task or complete a project, several people are usually involved. The people we need to interact with when performing a task are also part of the usage context. Easy access and ways to interact with them is as important, if not more, than to access the information we need.

Unfortunately, many of the tools and solutions we use do not account for this context, which I prefer to call “usage context”. Instead, they designed as if there were no other tools and solutions surrounding them, as if their use is our primary purpose. This is disruptive for many workflows.

Thus, to help us perform tasks, but also to switch between tasks and resume interrupted tasks, we need to keep the information and applications we need in a bigger context than - the usage context. They need to be available in the context of their work. That is why we must learn to know the usage contexts and never loose sight of those when designing tools and solutions that are to provide us with the information and tools we need to do what we need to do.

That is the overall challenge that Information Management is to address. It needs to provide the information we need at our fingertips, available just when we need it. It needs to link our needs to the information items, people and tools we need.

The main problem today is that the information landscape is fragmented, with information distributed all around and in different formats. There is too much redundancy, too much information about the same thing. The information is often inconsistent and it is hard for us to tell which information is the most current, if it is valid, and so on.

The natural starting point when aiming to improve Information Management in an organization is to start with the usage contexts. When mapping information needs to information entities, we must never lose the usage context. This brings me to the problem of many ECM initiatives.

Simply speaking, ECM addresses the need to manage unstructured content in an enterprise. The need comes from the fact that unstructured content, in contrast to structured content (data), usually is not managed even though a lot of the information which is exchanged via unstructured content is often very important for the management and operations of an enterprise.

Although many today acknowledge the importance of information that is encoded in unstructured content such as documents, mail and wikis, the ECM discipline is still often seen as doing “gardening” in the environment around business applications to cut costs that drain IT budgets. It puts more emphasis on the storage, protection and preservation of content than on supporting access, findability and (re)use of content. The reason is that it sees content as a liability, not as an asset.

ECM often also attacks the “content monster” from many different frontiers in isolation; Web Content Management, Document Management, Records Management, Digital Asset Management, and so on. This can be benefitial, since different types of content poses different challenges and needs to be managed in different way. But it is very dangerous if the usage context is not considered and in focus. Even though the ECM initiatives help to solve some basic problems such as knowing which is the current version of a content item and preventing it from falling into the wrong hands, they can make the fragmentation of the information landscape even worse.

The thing is that we as users don’t care if the content that contains the information we need is structured or unstructured - we just want to have all the information we need to do whatever we need to do. What we don’t want to do is to go and look for it in different places just because it has a certain format, or to find the same thing here and there but without knowing which is the one we should use. We don’t want to spend most of their time getting ready to do their work (i.e. collecting and managing information).

From a business perspective, many ECM initiatives fail since they do not help enough to defragment the information landscape. They certainly do not bridge the gap between structured (data) and unstructured content, not even from a content/data management perspective. That is why we need Information Management to keep us from losing sight of the usage context where it makes no sense to talk about data and content as two separate things - it's just something that carries information.

1 comment:

  1. Interesting. I link what you are saying to what Christopher Alexander (in "A Pattern Language") says about positive and negative space. The proper task of Information Management is to create a positive space in which people can operate effectively with information and knowledge. What it often does instead is leave a negative and fragmented space - people are forced to occupy and mediate the awkward gaps between disconnected lumps of information system.

    Another key point is that the space is under-determining. It is not necessary for the designer of the information space to anticipate all possible contexts; what is necessary is for the user of the information space to be able to respond appropriately to a given context without being overly constrained by the way the space has been designed.

    ReplyDelete