aculich
commented
10 years ago Over time, developing good collaboration and communication means that all 4 roles (data curator, analyzer, visualizer, and presenter) should be able to work in parallel. Though the data flow passing through this model is linear as we drew it on the board, you can work with a simplified, mocked-up version of the data so everyone can get started building their part, and then as you get the real data in place you can incrementally replace the mocked data with the real data.
At first it may not be obvious how to do that. It gets easier if you can precisely and completely understand the problem and what the end-result will look like-- though that's not always easy or obvious with all projects.
For this particular project it should be possible to get a rough idea of what the end result will look like, as well as the data that you need. The data we are working with now is somewhat dirty, but fairly well-bounded (~40 rows of data is manageable without resorting to unusual tricks).
In the workflow specified in class, the analyzer does her work, and then the visualizer. However, once the data curator has finished her task, shouldn't the analyzer and visualizer be able to start their work at the same time? We should know the format of the visualizations before starting, just not the specific numbers in the visualization. If anything, isn't it easier to do analysis after the visualizer has finished her work?