[MediaBot] Publishing Admire in Software Practice and Exp.

Guys:

I was looking around for conferences which would be suitable for 
publishing Admire, when I though of "Software: practice and Exp" 
Journal. I had previousy published a paper in this journal and I now 
think that this may be the best one for our project. Take a look at its 
requirements 
(http://www3.interscience.wiley.com/cgi-bin/jabout/1752/ForAuthors.html)
I have also attached one sample paper from this journal.

Since we did not use any new algorithms for the search or retrieval 
part, I think we might as well publish our experience in collating 
already existing components (Jade, TexttoOnto, MySQL, QuickTime and 
GnuMP3d) into the java fabric and creating a versatile product i.e. 
Admire. We will stress on the fact that we strived to take advantage of 
as much available code as possible (in the form of third party libaries) 
to weave the Admire System.
This is how I plan to organise the paper:

Section I: Introduction:
*Why do we need multimedia search engine?
*How would an agent based search engine help our cause?
*Why content-based (or semantic) and how is it different from 
audio/visual based indexing?
*Whats the reason for choosing java?
*Are there any existing prototypes that are similar to ours and where do 
we differ from them (at least 3 to 5 such products are required)
*If agent based computing is an old idea, how does it put a fresh 
perspective in ADMIRE? (Indexing algorithms can be changed on the fly)
*We will be describing our experience of building a java-based working 
prototype, the choices that we had to make while selecting the 
components and our experience with using them.

Section II: Architecture of ADMIRE. (We will need to change Preyas's 
Diagram)
*Should clearly indicate the following components that make up Admire:
-->Mobile Agents
--> Indexer
--> Database
-->Query formulation
--> Streaming Server
--> Renderer
*The function of each of these components should be well-defined.

Sections III: Choosing the right components:
* What  choices were available to us and why did we choose to use the 
particular component for :
-->Mobile Agents (why Jade?)
--> Indexer (Why TexttoOnto)
--> Database (Why MySQl?)
-->Query Formulation (How did we design the query formulation? What are 
its limitation (example, audio/video queries cannot be entered))
--> Streaming Server (Why settle on Gnump3d)
--> Renderer (Why Quicktime)

*Section IV: How to effectively use the running prototype in an actual 
system.
(We will have to give the exact steps that are required for using 
Admire).. Stress on the fact that this system can be effectively used in 
a small group of peer-to-peer users.

*Section V: Screen Shots of the running system and Future research work.

More thoughts on this will be added as they come.
If you think of more, please add them and keep everyone on CC.
Once we finish with the basic design, i will start assigning work. I 
plan to complete this paper within 5 to 7 days..
All comments are welcome.

Regards
Shashank

Attachment: Design and Implementation of Video Browsing System.pdf
Description: Adobe PDF document