Just recently I visited DMS Expo in Stuttgart/Germany which claims to be “Europe’s leading trade fair and conference for enterprise content, output and document management”. It was a large trade show, and of course I didn´t expect to see the Semantic Web playing a central role there but on the other side it became much clearer what´s still missing in most enterprises of today to be an “Enterprise x.0″: Open Minds which consider digital contents as a source to create knowledge out of it.
What´s obvious for most of the Semantic Web evangelists, isn´t clear for at least 75% of all exhibitors (and their clients) at DMS Expo. For these people who are dealing with core systems of today´s enterprise stacks it´s not quite clear that documents could be a valuable resource for enterprise knowledge management. They still focus rather on the basic idea that documents have to be revision-proof, they should be long-term archived and should be put into a safe. That´s quite the opposite of how content is organised in a (Corporate) Semantic Web. In such an environment each little piece of information at least has the potential to get linked with another piece of information.
Open World Assumption is not only about the way we put ontologies in place.
It is also about the basic assumption that people intend to get their content published and linked in a way that this creates an extra value for their colleagues and their organisations.
Documents are containers, and containers tend to be put into containers which are even bigger. In a world where documents are the atomic elements to get information organised the question always is: What should be in there?
On the Semantic Web information is no longer locked inside documents, same with wikis: The idea is to organise every little piece of information in a way that it can improve constantly because it´s out on the (corporate) web. Part of this evolution are mechanisms which help to get pieces linked in a meaningful way. On the Social Semantic Web this job is partly automatically executed and partly done by human beings. In this world which is based on the assumption that people would like to have their information out there on the stage the question always is: How can this piece of information get linked to other pieces in a meaningful way? Which metadata should be put on top?
It´s the people who make the difference, and for many of them there is still no “business use case” based on the “Open World Assumption”.
Here is my proposal: Appreciating each one´s work as a valuable resource for the whole organisation!
And I can hear the question already: Great, but how can I put this into my Excel?
“How can I put this into my Excel?”
I think a great, small demonstration of that is the <a href="“>google sets integration with google docs. Apparently, loads of the freebase data has been used to improve google sets – so it is a way to show fragments of typed data, free of any container, being re-used in a spreadsheet
I know the requirements from the corporate DMS & archiving world, which are exactly the ones you mentioned. The issue is that many of the requirements come from outside of the enterprise, and enterprises take consciously or unconsciously risks in not answering these demands. (e.g. demands based on SOX, eDiscovery, etc.)
But in technical terms I don’t see not the basic incompatibility you mentioned in the blog entry: even tools from the semantic web area can be made fit to cater the needs of traceability, archiving requirements etc.: e.g. you may think of archiving the information which was valid at a certain point in time. Or you may think of attaching the metadata in RDF format to documents before archiving them.
Before introducing semantic solutions (as any kind of other IT solutions) one has to think the functional and non-functional requirements a solution has to cover: doing a compliance check on e.g. which traceability, documentation and archiving requirements need to be covered, is one of the tasks.
(…and somehow you might be surprised, that not all of the rigid requirements apply all the time)
Thanks Daniels
@ Daniel O´Connor: With “How can I put this into my Excel?” I actually meant that managers always tend to make decisions on top of numbers – but thanks for your freebase example – where can I find it?
@ Daniel Koller: You´re right from a technical point of view. But many IT-managers are still not in the position to see how valuable information could be if processed, published and linked in the right way. With metadata rarely “semantic” metadata is meant but rather metadata which helps to process a document as fast as possible so it can end up in its SAFE.
Interesting discussion. Am currently working with a client who wants to push forward with both Knowledge Management and Document Management. Interestingly they see a document management implementation as one of the foundation stones for a knowledge management system. Would you agree with such thinking/ philosophy? I note your reference to ‘revision proof’ thinking – it is important for the client to know where the ‘final’ copy of the document used in a contract is. Is this type fo thinking what you are referencing?
Barry, I agree with you and your customer that documents should be one of the foundation stones of a knowledge management system (beside wikis, cms, crm etc.). But there have to be some pre-conditions in place before it actually becomes part of the kms:
- people and organisations are willing to share information (thus people should not be afraid of putting information “on the stage” anymore)
- metadata management is not only about metadata which is used to manage processes but also about the meaning of the content
- documents should not be put only in containers to store them due to law regulations but rather to make an asset out of them which flows to the place automatically where it could contribute in a valuable way (and these kind of technologies already exist!)
- innovation vs. standardization: management, especially in the western industry countries should start to realise that with such a low level of innovation in many industry sectors it won´t take a long time until we have lost our leadership position. innovation oriented organisations tend to put their staff in the centre of their considerations, thus such kms look completely different to thos which were invented in the early 2000´s. They are not about documents anymore but about PEOPLE, their networks, their knowledge and their ideas. Documents are really just a small part of the whole system in this version of a KMS.
Pls mind the correct formation of a question in English:
What DO the Semantic Web and Document Management HAVE in common?