December 28th, 2009
(With respects to Aarawak for messing with their tag line)
I’ve been thinking recently about the role of documents in an organisations process, and the slowly dawning realisation that they are increasingly irrelevant, barriers to information sharing, and possibly damaging to all that come near them. Ok that may be a little strong, but let me ellaborate.
When the ”document’ becomes the focus of your efforts, and not the information in them then they become a problem, not a solution. You’ll need document management, audit and access control. when all you really want is the latest info, in an easy to find and use format, also know as ‘a Wiki’.
Worse, documents, when created to satisfy the needs of a process or procedure are often resented by the creator, written after the fact and as quickly as possible, and as a result can include inaccuracies, disinformation and missing details that are needed, in short document suck and act as a barrier to information sharing.
Develop a wiki as part of your project, allow all the actors in your team to contribute as needed, and not only do you have a record of the activity, discussions and decisions made on the project, you have living documentation to support the product once its released.
Now in order to make sure that the information being added to the site is findable, you still need to have some control over the the structure, language and fair use of the wiki. The idea is to make sure that on any product wiki, any one type of information (IA diagram for a web site for instance) will always be in the same relative location. (mysite/ia/diagrams). How you manage this is up to you, either build the structure and contributors follow it, or you do a little post editing, the end results are the same.
In any event ditch the documents, focus on the information instead.
July 15th, 2009
Information is indeed power
When a child is young, she happily wanders around her world, unaware of the dangers she faces or how to avoid them. She relies on her parents, her instincts and as she gets older, her experiences to guide her and help her learn what’s right and what’s wrong, what’s safe and what’s unsafe. She develops a behaviour and knowledge to stay safe and grow safely into an adult.
She relies on her ability to recall past events, the lessons learnt from those event and associated activities or incidents that allow her to make decisions based upon passed learning.
The mind creates association between actual and potential situations, and develops a behavioural response that allows us to be safer, cleverer and more aware of our surroundings. Added to the direction of our parents, and instinctive knowledge passed on through generations then we have a learning and information system in our head that allows us to access an ever growing pool of sometimes only loosely connected information and use it efficiently in our day to day lives.
Knowledge, information and data is constantly being collated updated and created as we go through each day, just as you learn new things and store them, your organisation creates document, emails data in databases, spreadsheets, web content, paper documents in an ever expanding store of information. It’s saved in a myriad of applications, file locations and web solutions.
But how do you relate one piece of information (knowledge) to another, and how do you recall it.
Recall and retrival are increasingly important when you consider that new information worldwide increase on average by 30% every year, you and your colleagues face a 100% increase in the document you need to search through every 3 years…
Think you have a lot now? Just wait…
And yet we store this data in silos, separate applications that don’t talk to each other, they separated data from data from users from all our other ‘stuff’.
We create barriers to knowledge share and expect our colleagues to find these disparate knowledge chunks, how many organisations create new documents in and file them on file structure, and save them as:
Some random file name.doc
In
file://my location/another folder/yet another random name/misc/
And how much of that data is audited…ever?
Now add to the mix, just for fun, those additional applications an organisation will have in their infrastructure that are used to store, create and, if your lucky, share information on the same subjects or related to information in that other application or the documeng saved in randomfolder_name/ that guy over there just created.
Result?
Your staff members unable to find relevant documentation on a particualr subject, repeat the mistakes of others, they duplicate documents and knowledge already created and worse create updates it in alternative locations so there is no single authorative version of any document in your organisation…
And this isn’t rocket science.
Structuring your content (knowledge ) so that you can find, use and manage it across the entire organisation does not take long and will take considerably less time than the hours spent looking for it in any number of locations.
Implicit metadata (taken from the document you’re creating) makes structuring and categorising much easier for the user, and can be used to suggest tags.
Training and demonstrating the process of ‘tagging’ alongside the retrieval process and benefits can help convince detractors that the process is indeed worthwhile, and allows VAS.
Search
Ok so what about search…it works, we can search our repositories and document stores, so we ca do this instead of categorising content right?
True but, if you search your content for a string “document_type my_customer” you get all documents with those terms.
So you get clever and use the AND statement to join your search parameters, now you get just documents that have both terms, but not the ones that use the customername or the project_title. And you will still get all documents all versions even if they just mention the document you need. In short it’s a simple fact your search returns are only as good as your content, and don’t forget all those knowledge silos that you can’t index….
Search plus is a retrieval technique that lacks relevance for this reason you can only ever rely on return quantitative as opposed to qualitative response, by adding some aboutness information to your content you can vastly improve the quality of the returns given and thus cut down on the number if searches make and time spent making them.
Conclusion
1. Don’t limit you options
2. Think about your content and how it’s organised so that your users (Staff) don’t have to.
3. Time spent preparing content is time saved (magnified) retrieving it
4. Use control vocabularies alongside folksomic tagging (see point one)
5. There is no ‘single solution’