VisSym 2003 Trip Report
Really interesting conference,
nice place (Grenoble), interesting people. I have learned a lot, not
only in the Visualization field, but also on how to deliver a talk and
use at best the congress time. Here are some very quick notes to
remember my discoveries.
Really good talks…
- Real problems, real data.
- A lot of images (we work in Visualization, isn’t it?)
- They show immediately which the motivations of the work are and why I must hear to the talk.
- Provocative, make the audience think.
- Leave the details to the paper.
…and the ones to forget
- Read the slides, the formulas.
- Present an algorithm step by step. It is far better in my opinion to make the audience understand why you propose something.
- A comment I heard: “Too enthusiastic talk means poor paper”.
Ideas on talk structure
- Put something that remind you to the audience on every slide (email, URL).
- The speaker name is not always the first one in the list. So please repeat your name at the beginning of the talk.
- I think is nice to spend a slide presenting colleagues and a quick context (institution, project).
- The classical objectives slide is always boring. Please let me know were we are heading.
- Every talk finished with an Acknowledgment slide (who pays…). Maybe this can be combined with the context slide.
General ideas on the congress
- Read carefully the program before the congress starts. Try to understand from the titles which talks will be of interest for you.
- Take a look to speakers’ home pages. Try to memorize which are the top people.
- It is difficult to meet everyone if you only react (i.e. after the talk). You must talk to this people the first coffee break.
Afterwards the grouping is already crystallized.
Well, back to the talks
Here are the talks I enjoyed more:
||one of the first talks focused on
visualization usability and not on a new technique. Explores why
a certain technique should be used and which are the drawbacks.
||use of InfoViz techniques to help solve SciViz problem.
||puts the basis for building a scalar field summarizer.
||another approach to a problem I had (volume intersection in computer vision)
||nice survey of the field.
||importance of GPU to offload CPU workload.
See you at VisSym2004!