This is a key point Pierre is making here. Thankfully this is also one which
Phil on the RELOAD X4L project has been working on.
For the RELOAD framework, a lower-level representation framework provides
factories to get interface widgets which directly reflect not only the
underlying IMS schema, but additional constraints and usability assistance
using a helper schema derived from the IMS original, but containing things
like friendly names for elements and attributes, tooltips and so on.
This makes for a pretty powerful set of tools for building all kinds of
IMS-based applications, and is focused in the short term on building editing
tools for the X4L projects.
He's working on Java Swing interfaces for these, but there is no reason from
my understanding of what I've seen so far that would prevent using the same
set of libraries for building servlets for a web-based user interface if
thatıs what you needed.
Anyway, Phil's coding it so you can pester him for the details!
As Andy said - the schema is really a complicated mess, which makes it a
pain to implement, hence the need for some abstraction layer to make it more
accessible for people developing their own tools, repositories.
I had a go at implementing editable metadata for my Python lib and it drove
me spare! (How many levels of nesting is that?!)
- S
On 20/2/03 6:09 am, "Pierre Gorissen" <[log in to unmask]> wrote:
>
> The abstract method you describe is more flexible, but potentially much
> slower, more complex to code etc.
> If the interface parts, where you edit/add the metadatafield, aren't as
> flexible as the databasedesign, a change in the metadataschema would still
> mean codechanges in your application.
>
|