[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Minutes July 26
Learning and Training Subcommittee Minutes, July 26, 2007
Attending:
John Accardi, acting chair
Bob Doyle
Wayne Gafford
Allyn Radford
John: Reuben has donated a schema to support assessments and we want to determine today if it has enough interaction types to provide an adequate starter set. We need a DTA/Schema, an assessments information type plus an interaction domain.
Stepping through the proposed
interactions, we have:
True/False, which includes:
Title
Instruction Text
Feedback.
Our first question for the SC is whether feedback should be very simple, i.e., Yes for True, or whether we should support "per option feedback" which would support richer, e.g., multimedia feedback.
Reuben showed an implementation in Authentic by Altova (which is free), where you could associate an asset, e.g., an image as feedback. Maybe we could have smilies, etc.
Wayne: We did some work with SCORM/S1000D. The beta module schemas were not that good for interactions. We found limitations. They were text only, maybe text popups.
John: We need basic XML primitives,
that can drive the pointers to assets. Altova did something with SPS. What is that?
Bob: Those are Altova's proprietary stylesheets, developed with their Stylevision ($700?) product.
John: Does Authentic support some native assessment types?
Allyn: We are creating content to be consumed by an assessment application.
Bob: Can we separate the content from the (presentation) interactivity mechanism?
John: When Reuben played this, he used a radio button for T/F, had a radio button for multiple choice (single correct answer).
He did matching with two columns. He dragged and dropped items on left onto items on the right – in Authentic.
Action: We should ask Reuben to record these interactions with Camtasia and post the movie to the Wiki.
The important point is that the interactions.XSD contains the right elements .
Allyn:We must structure the content so the platers can play it.
Wayne: It must read some tags or attributes. In a SCORM environment you also need to manage the sequence and navigation piece.
Allyn: Yes, through a standard player.
Wayne: The right content model can drive an LMS session or sequence, using Javascript or whatever in a runtime environment.
John: We can get Reuben to tell us more about his company's implementation.
Allyn: I don't understand. Are we building this set of interactions from the ground up? Isn't it possible to DITA-ize a standard like the QTI? Wouldn't this be easier?
John: We've been down that road. Invested a month's time There was a way in vanilla DITA to allow non-DITA structures.
But QTI is an aircraft carrier when all we need are a couple of marines. For example, it handles how many chances you might get to answer a question correctly, how many hints you might be provided before closing the question.
We came away dizzy. QTI too heavy duty. They found QT light, which got rid of a lot, but was still too hard for our simple "starter set." for DITA.
Wayne: There must have been (good) reasons for all that complexity.
John: That's a very good question. It was discussed at the ADL Summit by John Hunt's people. They concluded QTI was not in much use. They also found some difficult politics going on between ADL (SCORM) and IMS (QTI?). There might be difficulty getting the IP to be available to OASIS.
Allyn: QTI gets built into a lot of LMS's. The ya re consumers of QTI content. It has built a community (huge in higher education).
Wayne: ADL wants a SCORM-conformant QTI. There is a big need in the testing and assessment industry. Medical school applications, standardized tests, etc.
John: John Hunt and I found it hard to locate real implementation examples.
Wayne: Who do know that is a member of IMS that's on this subcommittee?
Allyn: I can send a note to Colin Smythe (in the UK). He is a specialization strategist for IMS.
John: Can we get it donated to OASIS? If not, can we work with it by changing the tag names a bit?
Allyn: At ADL it seems that IMS has developed a non-constructive position recently on use of its IP, even though development and funding resources came from outside IMS.
Wayne: ADL and IMS are in a tussle over SCORM. It's how IMS regards sequencing and navigation. OASIS use is not likely to go over well with IMS.
John: We tried. And we struck out.
Allyn: If QTI is useful, there is a process to deal with this. If it does the job. I will talk to the chairman of IMS.
Bob. Can we take a viable subset of QTI for our starter set?
John: That's QTI light.
Action: Allyn to get Colin Smythe on a future call to explain things.
Action: Could QT light serve our purposes?
John: In the remaining time, can we get a sense of the subcommittee on per option feedback?
Allyn: Can we ask for both the simple yes/no and the expanded per option feedback?
Wayne: The question is – does the schema support this?
John: We have the current schema and an ideal schema.
Wayne: I don;t like something as specific as Reuben's red X's and green check marks. That is presentation, it should be separate from the content.
John: We need Reuben to explain more.
In the meantime, are there any obvious interaction types missing?
Bob: I reviewed those offered by the Moodle LMS I want to use to instantiate our deliverables, once we get them. It also has an essay type, and two sophisticated types that accept variables for numeric parameters, and for a range in an answer value.
Wayne: Sophisticated assessment does not require a complex schema.
Allyn: Agreed We just need basic primitives, like premises.
John: So we are happy with these types?
Allyn: As a starter kit. For the DITA environment.
Wayne: (objecting) Don't suggest that the DITA standard will be limiting in any way.
Bob: Specialization of our starter set can provide for more detailed interactions?
Wayne: Lots of concern about seeing a DITA specialization that could do S1000D.
John: We are running overtime. Let's adjourn.
_______
-- Bob Doyle
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]