[annodex-dev] Using CMML without Annodex container

[Silvia.Pfeiffer at csiro.au]

Hi Brad,

CMML has definitely been developed with the "storyboarding" paradigm.
Each of the clips identifies a segment of content overt time. With the
sort of "presentation"-style content, we were already thinking of having
an audio track accompagnied by a timed image track and a CMML track.
Thus, if each slide is linked to exactly one CMML clip, chopping this up
into the storyboard type content that you are after is fairly simple.
That can then be packaged in a tarball.

I can see that your ideas are working. However, I don't understand this
bit:
> CMML (with or without the annodex stream) might 
> be better shown as the image (still or moving) with the appropriate 
> annotations shown in a text box, along with some kind of 
> master control. 

Cheers,
Silvia.

> -----Original Message-----
> From: annodex-dev-bounces at lists.annodex.net 
> [mailto:annodex-dev-bounces at lists.annodex.net] On Behalf Of Brad Hards
> Sent: Friday, 9 September 2005 3:02 PM
> To: annodex-dev at lists.annodex.net
> Subject: [annodex-dev] Using CMML without Annodex container
> 
> 
> G'day, 
> 
> I'm interested in whether CMML might lend itself to storyboarding.
> 
> As I see it, annodex is focused on the continuous media, and a set of 
> annotations (the CMML) which are auxillary to the media. 
> There is some value 
> in applying CMML to sequence and annotate  non-continuous media - for 
> example, a series of PNG-encoded screenshots with some 
> explanatory text. 
> Annodex appears to support this, as "time-instantaneous 
> bitstream", however 
> there may be an alternative packaging that makes mores sense in some 
> circumstances (eg just a bzipped tarball might be very useful as an 
> interchange format during the authoring process, although 
> that has obvious 
> disadvantages such as not being able to be delivered as a 
> temporal subset).
> 
> I'm starting to see some interesting ideas in useability 
> coming out of the KDE 
> community, however moving around very large files (eg an MPEG 
> encoded full 
> screen recording) is impractical for some of our developers. 
> I'm thinking 
> about CMML as a tool to help collaboration. 
> 
> Of course, XHTML can do some of this already, however CMML 
> (with the concept 
> of events happening at a particular time) might make for a 
> better viewing 
> experience. Typically the rendering for XHTML would be a 
> linear page, with 
> alternating text and images. CMML (with or without the 
> annodex stream) might 
> be better shown as the image (still or moving) with the appropriate 
> annotations shown in a text box, along with some kind of 
> master control. 
> 
> Not sure if this makes any sense - appreciate your thoughts on it!
> 
> Brad
>