MINUTES OF REVIEW MEETING OF GROUP CHARLIE AT 2:00PM ON 1/3/2000 AT TP4

Present : Dave, Martin, Phebe, Karen, Nathan, Andy, Dr. Anderson

Demonstration

  1. A few examples were demonstrated at the Review meeting :-

    2. Expensive locks / Cheap tunnels case

    Round hill case

    Pathological case and a few other

    Dr. Anderson’s questions

  2. Q. Is there a functionality for terrain generation ?

    3. A. The terrain generation will be added later after the review meeting.

  3. Q. Have you looked at some physical canals, and compare whether the program produces similar routes ?

    4. A. The team may consider looking into this.

  4. Q. How long does it take to find the route for example 100 x 100 grid ?

    5. A. It depends on the terrain and distance between start and end points, and usually over an hour for large height range. A 30 x 30 grid takes about 2 minutes. An algorithm "DijkstraSurface" module was also implemented, in order to provide a quicker calculation for the route ignoring tunnels. A full 100 x 100 grid may take 4 to 5 hours.

  5. The team said that optimisation has been investigated, but to date, there is no feasible one capable to find the most economical route. Dr. Anderson said that in previous years, some groups had managed to optimise adopting calculation by cost. This is different from the methods the team had adopted, which is by vertices.

  6. Dr. Anderson said that the 3D display that Charlie produced is one of the best he has seen for the past few years. He asked what would be the group's plan in terms of presentation and demonstration in order to win the prize.

    7. Lesson Learnt

  7. The team members had learnt how to work as a team, and also the importance of good communication.

  8. In the area of programming, it is better to divide the tasks into more modules, though the team development strategy of 2 high level modules was still in the right track. Dr. Anderson said that the optimum module should be a few man-days work, optimum both in the sense of overhead and work amount.

  9. In term of planning, the functional specification should be written so that it covers the generic elements but allows flexibility.

  10. Most of the team members do not find testing appealing. Testing seems to be quite a high portion of software development, estimated to be about 40% in the development of Charlie. Dr. Anderson commented that it was normally the case in a commercial development as well. He also mentioned that sometimes working hours would not be the appropriate control of a project.

  11. The different areas of computing field include management, consultancy, development and testing. Apart from testing, most members of the group are happy to get into any areas of the above as a career.

    12. Questions from the Group

  12. Are we provided with any visual aids for the presentation ?

    13. Dr. Anderson replied that OHP is normally provided, but discouraged machine such as data projector etc., because each group only has 5 minutes to do the presentation, time management is crucial.

  13. What is the legal status concerning the ownership of this project ?

Dr. Anderson replied that, for Part 1B project, the product would be owned by the students involved in the project, not the University, though may be different for Diploma / Part II projects.

Next Progress Meeting : Saturday, at 11:00 am, after lecture at Cockcroft 4

Presentation/Demonstration : Wednesday, 8/3/2000 Babbage Lecture Theatre and Cockcroft 4