Forum .LRN Q&A: ANN: IMS Learning Design Package Specification (RFC)

Hi everyone,

As we promised, we are releasing the specification of the IMS LD package (the package name is imsld, if you have a suggestion about the name, please let us know).

You can read it at:
http://cvs.openacs.org/cvs/*checkout*/openacs-4/packages/imsld/www/doc/index.html

We would appreciate any feedback about it. We have just finished the first version of the spec, and we realize that there are some things that can be improved, like the datamodel, wich can be normalized. Anyway, any suggestion is welcome! (this is the right time ;o)

Regards,

Jose Pablo

Hi,

The data model (level A) has been normalized (http://cvs.openacs.org/cvs/*checkout*/openacs-4/packages/imsld/www/doc/ch04s01.html), and some errors found in the spec where fixed.

Please let us know any comment or suggestion (specially identifying which tables can be modeled using acs_objects or items).

We will now focus on the API, and let the normalization/optimization of the data molels of levels B and C for later, after we support level A.

Regards,

Jose Pablo

Is nice to see that you've done your homework doing such kind of documentation, but actually I would suggest to focus on the UI for the end user, the professor and the student. Since those users will benefit and use the application, and they actually do not care that much about design, datamodels and scripts. Have a look in moodle, and how a professor can easily create a basic activity sequence. Import / export as IMS LD will be good. Of course IMS LD can handle many more things, but if we don't focus on our actual professors who in most of the cases aren't learning design gurus, in those simple users needs as a first phase, this application will reach just an small number of people.

So, better to see how we can make this an easy-neat tool for professors & students in the short term, and how to make it compatible with IMS LD. Then learn from that, keep improving it and increasing the IMS LD support.

Collapse
Posted by Rafael Calvo on
Hola

Is this related to the alfanet work?
Have you talked with the people at UNED?

We are also using IMS LD in our 'beehive' project:
http://www.weg.ee.usyd.edu.au/projects/beehive

We should make sure that our work is integrated

Rafael

Rocael,

You are right, we will now focus on the API and UI of the first part of the package, wich will consist of the parser and sequencing activities using what we already have (assessment, evaluation, file-storage, etc, one by one), but we were waiting for feedback and meanwhile improving the datamodel ;)

We will try to make the UI as easy as possible, and we will work the same way they did with the assessment package (1. using an external editor, 2. parsing, 3. displaying, 4.editing...)

Any other comment/suggestion is welcome. BTW, moodle does not support IMS-LD yet, they will support IMS-LD on the next version that is planned to be released in Octuber.

Regards,

Jose Pablo

Rafael,

We already talked with the people of UNED (aLFanet - https://openacs.org/forums/message-view?message_id=301221), but what they have is not exactly what we want (it can be useful though), and the code has not been released yet.

About IMS LD and the beehive project, are you talking about dotTEACH (or other .LRN package)?, sure we would like to integrate our work (if our final goal is the same, of course)

Regards,

Jose Pablo

Yes, alfanet stuff seems to be a bit different, and not available yet.
Indeed, moodle started with what the users wants/use and then add ims-ld. I actually do not agree with doing the same stuff as were done before for assessment. We need something for the users within a month or so, then if we leave at the end "4. editing", which means actually what the user can do online for design a course and its activities, will represent no real impact for users.

As somebody pointed out in the past for assessment, it makes NO SENSE to create end user applications that can only be modified from outside, in this case might be enough to just have a subset of "edition for design of courses", we'll see in real scenarios what's is needed.

Also, for the professors and the learners, to have a UI like COPs were you can add many things around using a single template will be good, and probably more intuitive / easy to use. The key issue is figure out how to design that UI in a simple manner, and until you don't have feedback from real users, you will be in a circle with your own thoughts. This can be a killer application if we focus in the users experience.

The suggestion if the focus is the UI, is to leave imp/exp for a second phase, and simplify the datamodel while keep in mind future imd-ld needs.

Rocael,

Yes, indeed, the editor is a very important piece in order for the entire functionality to be useful.

I believe the compromise would be, let us first make a first cut at the implementation, not writing code, but planning the structure. Then we can proceed and propose a UI with user models, templates, etc.

All this before implementing anything. I believe once we have this solidified, we can deploy a first version with significant functionality and a basic editor.

So, following your suggestion, we'll include in the spec an editor definition and user scenarios.