DaCoPAn Software Engeneering Project

- Home  - Overview  - Members  - Documentation 

- Resources/Links - Project Website

 

Dacopan project                                MEETING MINUTES
                                               Feb 13th, 2004

Meeting
        Feb 12th, 2004, at 8:15

        Department of Computer Science

Attendance
        Alejandro Fernandez Rey
        Carlos Arrastia Aparicio
        Jari Aarniala
        Vesa Vainio
        Jarkko Laine, secretary
        Jonathan Brown, chairman
        
Absent
        none

1. Start
        Mr. Brown started the meeting at 8:25.

2. Assessment of the situation
   - The Russian group has returned to Petrozavodsk on Feb 11th.
   - The project plan is almost finished; schedule and size estimate
     still need to be done.

3. Discussion

   In the meeting the main focus was on brainstorming for different 
   animation types and creating the size estimate for the product.

   3-A Animation types brainstorming

      Three different types of animation were discussed:
       * Message sequence chart (Main animation type)
       * Encapsulation 
       * Event timeline (A possible additional animation type)

   3-B Message Sequence Chart

      - It was pointed out by Mr. Vainio that in order to
        make reading the text presented in the animation easier 
        the screen should not scroll all the time but instead 
        only glide upwards after reachings a certain point in
        animation.
      - Mr. Aarnio also suggested that information on host 
        variables and states could be written in columns next 
        to the packet animation.
      - It was reminded that we should only have one layer visible
        at a time.
   
   3-C Event Timeline

      - The event timeline is an animation type presented by
        Mr. Vainio. See [1]
      - The benefit of this animation type as stated by Mr. Vainio
        is that it shows information from many layers at the same time.
      - Mr. Fernandez noted that this is not necessarily desirable, as
        showing multiple layers can be confusing to the user.
      - Mr. Laine suggested the possibility of rotating the 
        Event timeline view 90 degrees so that time would increase 
        horizontally instead of vertically. That would give more space
        for the information and make the animation look more 
        "traditional". It is also more intuitive for the user to 
        scroll downwards than to the right.

   3-D Encapsulation animation
   
      - It was noted by Mr. Brown that the same animation type 
        should be able to show both encapsulation and fragmentation.
      - As Mr. Vainio mentioned, there is no need to be able to show
        encapsulation animations showed together with other types 
        of animations. As encapsulation happens rather in layer dimention 
        instead of time dimension, it would be hard to orchestrate 
        these animations together.
      - Mr. Vainio suggested that a static diagram should be produced
        during the encapsulation animation.
      - Mr. Laine noted that there is no need to animate decapsulation.
      - It was finally suggested that an animation type with a divided 
        screen could be used: on the other side there would be a static
        view on the encapsulation and on the other side an animated version
        maybe with more information on header fields.
      - In animation the following issues should be taken into 
        consideration:
         * How to show the header structure so that it is 
           clear for the viewer that the headers from upper
           layers are only payload for lower layers.
         * As Mr. Aarniala pointed out the relative sizes of
           the headers and payload should be someway visible.
         * Which header fields to show and how.

      Mr. Arrastia and Mr. Fernandez left at 9:50 and the meeting was
      continued in a different class room.

   3-E Size estimate

      A rough division to modules and a very early version of the class 
      diagram for data structures was sketched by Mr. Aarniala. Then
      the sizes of different parts of the Animator were estimated separately. 
      For each part a size estimate was made by lines of code and effort 
      needed to complete them. 

      This information was added to the project plan immediately after
      the meeting by Mr. Aarniala.

   3-F Desicions on requirements
        
      It was decided that Java 1.4 will be used in development of the
      Animator as it contains direct XML support. If possible, XML
      output will not be supported in the final product. Mr. Aarniala 
      however will do some research on the possibilities for doing that.

4. Action points

   - Cancel the Tuesday (Feb 17th) meeting with customer. 
   - Write rationale on the size estimate creation.
   - Add a section for meeting blackboard images to the TWiki.
   - Start the glossary section.

5. End
        Mr. Brown ended the meeting at 11:15.
        

Back to DaCoPAnDocumentation