As usually, I started the first day of the week with a short meeting with Sarah. Dr. Aragon joined us later. We discuseed the first prototype and the next things that have to be done for it.

Things to do:

1. put dynamic queries into the demo
2. look into panning techniques.
3. increase the performance

After the meeting I started working on the fisheye effect for the vertical axis. There seemed to be more target names than the number of dots; the order was not good because the names were not matching the correspondent SN, although I made all the names fit the height. I calculated the size of the names accordingly to the measure of the main frame and I updated the code with this information together and the offsets. In the mean time, Sarah put a scroll-bar and the queries panel inside the big frame. The panning "to-do" was solved.

Before the Independence Day the layouts for both axes were fixed. Sarah decided that is better to have the phase query implemented first and she helped me by creating the table with the data. She provided me with a snippet for removing one small dot. Using that and my code for the fisheye-calendar slider I implemented the first query; the user can remove and add graph-dots. The same problem from the tabular calendar - with skipping days in between, applied to the phase slider. The problem appeared when you combined the text fields with the movement of slider: the updateRange was skipping some nodes.
Sarah had a look at the code and sent me a snippet for the mousePressed method; the problem might have been with not updating the old values and the extents.

On Thursday I implemented the real fisheye effect for zooming in and out by only rolling over the numbers/names with the mouse.
At the end of the day I had a meeting with Sarah and Dr. Aragon to discuss the comments received from the astrophysicists about the first prototype.

Here is my list with the things I have to do in the next 5 weeks:
1. create the fisheye rollover effect for both axes.
2. create dynamic queries for redshift, spectra and others that the astrophysicists may ask for.
3. look into performance
4. select a target and make the entire column or row visible.
5. put the small target graphs on a box.

By the end of the week, I finished implementing the fisheye effect for the vertical axis; the text was colored differently when the mouse rolls over the names (i.e. red for the focus target, blue for the neighbours). The next step was fixing the layout when we were in the focused case, when one target was selected. The effect worked fine until you selected a target and you rolled over the other ones; all the offsets were shifted to the left and the space in between them changed. The first thing that had to be done in the following week was re-writing the layout that also has the special effect implemented.

Also, by the end of the week, I fixed the slider problem. I needed to change the mousePressed method, the updateRange and all the other calls made to the updateRange method. The problem I had was mostly given by the extent and the end of the slider.

Regarding the performance I found a profiler plugin for Eclipse. I talked to the sys admin of the workstation I was using with Eclipse and hopefully, soon I would be able to profile the codes and to improve them, to make them run faster.