4.3.3.3 Adaptability for Multimedia
The issue of how to make multimedia courseware adaptable is not easily addressed.
The separation of data from applications is one solution that will improve the
adaptability of multimedia. The following passage demonstrates a commitment to
the separation of data from applications for multimedia databases by the
AthenaMuse 2 development team:
AthenaMuse currently has an internal representation of multimedia
objects but lacks an accepted, external (file-based)representation.
This current limitation makes exchange of multimedia information across
Muse applications in any standard way extremely difficult. We intend to
use existing database systems to create a Muse-standard representation for
multimedia "documents", thus allowing storage and retrieval of components of
Muse applications. (CECI, 1991, p. 5)
The separation of the database from the application is again currently among
the goals of the AthenaMuse 2 software currently being constructed.
But a different issue associated with the availability of multi-media
within a distributed paradigm was highlighted by Janet Daly
(personal interview, March 4, 1992), Information Officer at Athena:
Daly: The two major issues regarding truly distributed multimedia
are network capacity and data compression tools. We weren't sure whether we had
adequate bandwidth to accommodate full motion video as well as data across the
same network. If you then look to data compression tools, you must find proven
effective and reliable data compression tools, otherwise when they don't work,
you've hosed everybody on subnet and you make no friends. Thus, the Project
Athena multimedia work involved the local delivery of video from a video disk
player connected directly to the workstation, and the local digitization of
that analog video signal by a digitizing board installed into the workstation.
The lack of dependable data compression tools was clearly an obstacle of adaptability
for availability standing in the way of the implementation of some multimedia in the
distributed networks. While the above describes a limitation of networking structures
a few years ago, CECI at MIT is currently preparing for the networking infrastructure
of tomorrow, and designing tools that will be able to adapt to those structures when
they are available. Part of the CECI agenda for the AthenaMuse 2 software under
development is image video compression, which will play a key role in allowing
digital video to be "usable," "available," and "adaptable" when the networking
infrastructure matures to the point where digital video becomes common.
The descriptions of the goals for the AthenaMuse 2 software in this area is
described below:
Image and Video Compression: New optical media and networking standards
(e.g., DVI, JPEG and MPEG) involve image and video compression. AthenaMuse 2 will
support standard compression formats as they stabilize and as the associated hardware
peripherals/boards become available.
For projects like the Geology Tutor, it is important that the multimedia control
software used to construct the courseware also be adaptable, because of all the
fields in computing today, the digital multimedia field probably changes more
rapidly than most. The issue of how to adapt multimedia software for change
was also discussed in a conversation between Hopper and Lerman
(personal interview, March 4, 1992):
Lerman: Well there are several areas where we decided we didn't
have capability. One of those things we didn't do with AthenaMuse 1 is we
didn't think what it would mean to make it adaptable. In AthenaMuse 2,
we're much more aware for example, one of the issues is things like
color palettes. What is a computer's color palette? What is AthenaMuse's
capability to adapt to different workstations' display sizes, physical dimensions,
and resolutions? Ideally, applications written in AthenaMuse should be able
to adjust to that. AthenaMuse should know how to remanage the real estate
on the screen for a different application.
Hopper: For new capabilities that haven't been conceived of.
Let's say an entirely new digital form came through, a new MIDI form,
or something like that, how easy will it be to add? Are you going to
need new additions, or will the expandability allow that?
Lerman: We hope to design modules, but of course the proof is will
it eventually work? when new technologies come along, are they easy
to integrate or hard to integrate? There's some ideas, for example,
in Apple's Quick Time. They designed the compression-decompression
in a place that allows you plug in new compression routines relatively
easily. It's a designed in effect, a place to put in a decompression
algorithm. That sort of extendibility is what we hope to design in.
Will we be able to foresee all conceivable? Probably not is the bottom line.