ever more refinement

So much further to go in the next few days... I think I can make this happen, as long as I stay focused on my end goal. Contrary to my sparse postings of late, I have been working on my project.

It seems that when I get deep into a problem in an unknown territory, I spin my wheels about 10x greater than I effectively move forward. This is frustrating, but at some level, I know that all that wheel-spinning has its fruits as well. The dynamic multiple choice experiment I was distracted with a few weeks ago was useful recently as I tried to duplicate dynamic text boxes as movie clips.

I spent about 5 hours on Saturday trying to set up a way to switch the kind of data is displayed in the bar graph. Right now, this is showing county data relative to values in the other counties. The highest value on the chart is equivilant to the highest county value for the dataset. A different way to view this would be to frame the data as it is distributed throughout the state, so what percentage of the state data exists in a particular county? I was able to create code that would do this, partially, but it would not toggle well between the two views. After 5 hours of struggling, I had to resign myself to keeping my graph view to only one option for now. Even as a second view would be valuable, the percentages of a single county in a state of 159 are very very small and mostly evenly distributed (well, for the data I tested). This can be accomplished or further explored in my next version, the after the studio version!

I've also been struggling with labeling each of the bars in the graph with the appropriate county name. I have tried every way I can possibly imagine to generate text fields that label these bars. What I have determined is that some part of my code is taking a major chunk of processing power and the code I wrote to dynamically generate this information was dropped. I could be presumptious with this... but I've tried so many things and proven that my code works in a clean document! I had to abandon this to some degree too. What I have done to solve the problem is to create a separate flash movie that contains all of the county names at the appropriate angle, size, and order. I'm importing this movie into my graph movieclip. It works, but it would be a pain to edit later -- Georgia does change its county names from time to time, but that has to be a pretty rare event.

I've also adapted the graph to mimic the color scheme of the map. I hope this will eliminate the need to make a legend that explains what each color gradient is equivilent to. When the user rolls over the map, the value in question is focused in the center of the graph and all the other information fades into the background.

I think I can move on from the graphing feature! The only refinement point for that is to find decent placement for the scrolling arrow buttons.

Here's the file in question: barGraphASedited

The next problem I need to solve is finding a way to effectively communicate the purpose of my project with my users. I've gotten quite a few comments on this, and I know that my project is not necessarily self-explanatory. I'm exploring methods to separate the dataset choice from the map and graph page. I've got this working halfway, but my map buttons no longer work. *sigh* I'll need to comb through my actionscript a bit closer and determine what must be displayed at what time to make everything to work seamlessly.

Here's that file: barGraphGettingBetter

What else do I need to accomplish?? Well, if I can make this entry page communicate the right information, I might be able to bring back some of my concept work from earlier -- this would primarily include using the map to interface with the Impact Statements system.

Does this project need to be pretty? I'm not so sure. It won't really serve as a stand-alone item in its actual usage. It does need to be useful and non-intimidating.

Onward onward! Tomorrow is going to be a busy day.