Re: [OSCOM] Open source CMS / documents are content/

On Thu, 17 Mar 2005, Seth Gottlieb wrote:
> The author needs to understand that their output is not just what they 
> are able to achieve on their screen but has other uses as well.  The 
> developer needs to understand what formatting the author needs to 
> enhance the meaning of the content and give them an area for that 
> expression.

Authors are generally focused on the task at hand (producing information) 
and have little interest in structuring the information to be parseable by 
the system.  Database systems and XML document authoring tools enable 
authors to produce highly structured content with defined semantics by 
enforcing validation rules and integrity constraints.  Word processors 
don't.

I think two concerns push organizations away from using structured 
authoring tools that enable semantics ... cost and flexibility.  Most 
organizations have a lot of unstructured information on paper, in Word 
documents, in HTML and probably a dozen other formats.  Turning that 
information into reusable content may or may not be a goal of the 
organization.

On Thu, 17 Mar 2005, Paola Di Maio wrote:
> What I am trying to say is that with the same investment, you could have 
> a flexible system that can be extended to handle all content formats and 
> functions, if and when required.  A well designed CMS can handle DM as a 
> matter of course, if thats part of the specification, the contrary is 
> not true.

There is nothing technical that precludes the creation of a document and 
content management system that can usefully manage both reusable content 
(let's say for the sake of argument XML), and non-reusable content (again 
let's take scanned PDFs as the canonical example).  Both types of system 
generally support:  versioning of objects, workflow tools, meta-data 
management, roles and permissions, indexing and searching, hierarchical 
directory structures (and other tools to organize objects).

To enhance a good content management system to do document management, one 
would need to add support for mass importing and indexing large 
collections of documents, an OCR tool to index scanned content, and a UI 
for internal users that emphasizes adding and versioning documents, and 
organizing and searching the library.


Oliver