Data Classification and the MER Conference

Data Classification seems to mean different things to different people.  In our File Investigator products, we classify each file by

1. Platform it is typically found on (ex: Macintosh, MS Windows, Linux, …)
2. Storage method(s) used (ex: Archive, Digital Audio, Vector, …)
3. Types of Content inside (ex: Database, Personal/User Data, Spreadsheet, …)

That fits the meaning for some, but not all.  The government is most interested in classifying data into separate clearance levels so they can decide who has access to the data.  Document Management products use classifications to decide what area of a server each file is saved to and what data should be backed up often, seldom or archived instead.  Electronic Discovery products use classifications to help narrow down the search for evidence.  Which use is right for you, and would it be advantageous to merge these uses into a single product or service?
A friend recently attended the MER (Managing Electronic Records) Conference in Chicago, and provided me with some information on Orchestria.  I was familiar with most of the other vendors, but I had not heard of this one.  Orchestria recommends the following list of best practices:

1. Deploy a common platform to classify and control converging information obligations
2. Identify, classify, and control information across multiple channels
3. Classify and control data in motion and at rest
4. Support both user-directed and automated classification models
5. Analyze and classify information across multiple dimensions
6. Deploy at multiple points within a corporate infrastructure

I interpret this as a single solution that encompasses all corporate data from all departments and provides the necessary security and customized actions using automated and manual classification methods.  Wow, that was a mouthful, and probably a run-on sentence.  They’re obviously working to be the ultimate solution to all customers.  Is that realistic or science fiction?
We prefer to focus on one area at a time and be the best at it.  Right now, that’s identifying files / records and extracting as much background information and metadata as possible.  The better the solution, the more you can automate the process.  The biggest complaint that I hear about today’s Document Management solutions is the laborious process of checking in your documents.  One alternative is to automatically inventory files company wide without your intervention.  That would protect against accidental or malicious  supression of valuable company data.  Some companies already do this with email.  It would prove very valuable when faced with litigation.  But, then they would need an effective Data Reduction method to avoid data overload.
Isn’t it better if your company’s solution volunteered the most likely input for those manually entered fields based on the type of file and data found within that file?  You could simply skim the results and make small corrections before continuing with the check-in process.  We provide that technology today, but the solution providers need to hear their customers request that feature before they will add it.