"Show notes" for this episode…

Episode 002 - The Classification Service
Synopsis
James and Jon discuss classification, what it means in records management generally, how it is implemented in the classification service of a MoReq2010 compliant records system, and how this differs from earlier specifications.
Featuring
James Lappin and Jon Garde
When recorded
September 2011
Running time
59 minutes
Suggested background reading
From the MoReq2010 specification, Volume 1, Core Services and Plug-in Modules (Version 1.0):
- Chapter 5. Classification Service
- Chapter 201. Hierarchical Classification
Main topics discussed
- Classification is essential to records management – you cannot have a records management specification without it
- But what does classification mean?
- And how does this inform the implementation of classification in MoReq2010?
- Traditional classification schemes are hierarchical lists of transaction types
- Records need to be associated with a specific event or transaction
- MoReq2010 handles aggregation separately from classification
- It is traditional to limit users’ activities to a narrow area of the classification scheme
- Users, even power users, can only perform file and folder functions at the lowest classification in the classification scheme
- These types of scheme by their nature require all records in a file or folder to have the same classification
- Other mechanisms such as record types must be employed to distinguish detailed classification within a file or folder
- Within a MoReq2010 records system each record in an aggregation can have a separate classification
- MoReq2010 does not need the concept of a record type or other classification mechanism
- Instead of creating separate record types for special types of records MoReq2010 encourages the creation of specific classes instead
- The classes used by a MoReq2010 records system can come from different classification schemes
- A classification scheme can be published as a service that can be used by more than one records system
- In the future the DLM Forum will look at individual industry sectors to see what classification services might be common to more than one organisation
- Each class is linked to a disposal schedule which determines how long a record is retained by the records system
- It will also be possible in future MoReq2010 compliant systems not to have a hierarchical classification at all
- Keyword AAA is an example of a highly respected classification mechanism that does not adopt a purely hierarchical approach to classification it also has some elements from a thesaurus
- In Keyword AAA the path through the hierarchy gives uniqueness to the classification not the specific node alone
- In the future MoReq2010 may adopt an plug-in module that allows Keyword AAA as a replacement for hierarchical classification
- This will be driven by real word demand
- Across different industries we hope in the future to develop joint teams that explore different plug-in classification schemes
- Within a specialised industry only one classification scheme may be applicable so different suppliers can build a system that only supports one type of classification
- This is a feature of the plug-in approach
- And results in a simpler records system for the end user (that is also easier for the developer to develop)
- In a hierarchical classification scheme an aggregation must be classified at the lowest level of classification
- In other types of classification scheme this may not apply depending on how they are defined
- All records must be classified in MoReq2010
- What happens to documents in a mixed document and record environment?
- Does a record have to be within an aggregation?
- MoReq2010 defines a records system that can be part of many different types of line of business systems
- EDRMS is only one possible application covered by MoReq2010
- MoReq2010 does not seek to define elements of document management that are not also necessary for records management such as document versioning
- Different versions of the same document may be declared as records relating to different events or transactions
- Without a classification a record is not complete and not really a record
- Search alone is not a replacement for classification in a records system
- Search cannot be used as a piggy back to apply retention rules
- It is not the hierarchy that is important to classification it is what the classes themselves represent
Latest news
- The DLM Forum has announced its triennial conference for December 2011 in Brussels – here is the link http://www.dlmconference.com/
- The timeframe for presenters is closing soon
- In November 2011 there will be a meeting for translators of MoReq2010 – here is the link to the DLM Forum website
- Topics covered will include:
- What happens with records management in different countries?
- How MoReq2010 might be adapted to meet national requirements?
- The DLM Forum has approached potential test centres with a draft of the test centre accreditation contract this model agreement can be downloaded from the DLM Forum website
- Enclosed in the draft agreement is the process for testing and certification of compliant solutions not just the process for accrediting test centres
From the postbag
- Will vendors have much work to do to modify their existing systems to implement MoReq2010?
- This will make an interesting future case study
- There is a fine balance between flexibility under the specification and its mandatory elements
- All previous systems will require at least some modification, for example MoReq2010 uses UUIDs to identify elements uniquely within a records system and across records systems
- All systems will have to adopt the MoReq2010 XML export schema as part of their export service
- Data that is exported must have the same meaning for all systems
- We look forward to hearing through the postbag about the different experiences of suppliers as they set about implementing MoReq2010
- Suppliers can also now contribute to the specification through the technical panel of the MoReq Governance Board
![]() Previous Episode |
![]() Next Episode |
Copyright © 2011 & 2012, James Lappin and Jon Garde










