GSoC Day 56

It’s exceptionally windy outside

• And rainy. I’m staying with three other people this week. I think it’s all girls. I really like it so far. Everyone’s super friendly. I have my own room, and internet seems fine, so I think it’s pretty much like I were at home (except for the shared-refrigerator thing).

I spent the day locked up in my room mostly

• I took an afternoon nap and pretty much spent the evening looking at Haskell and Elm.
• Trying to deconstruct the stamp example, which I think is making a list of the the pentagons based on their X coordinate. The X coordinate is the centre of the pentagon.
• So something like if Mouseclick then store X and Y coordinate, where each Mouseclick is of form [x,y]. So we end up with a list of lists. [[90, 200],[285, 20]], etc.
• It then makes a timeline based on the closest X coordinate and the furthest one, and that becomes the “timeline”. To go back “in time”, we look for the X coordinates collected in the list. So say, if one pentagon is drawn with a centre of X coordinate 89, on our timeline, if we go to the point set at 87, that pentagon does not exist. But I could be wrong :D
• So my question was if it’s using Mouseclicks and geometry to create a timeline, how can this be tracked as an event without mouseclicks or geometry? Is this possible?

I spent the other part of the evening

• Reviewing Higher order Functions and Typeclasses and also trying to understand the State Monad.
• I really like the examples of probability, because they are essentially the way you would expect to do them in statistics.
• Oh, I also joined the Haskell-art list forum for fun. One guy I know already started posting on there. He’s into ontology, Haskell, and live-coding.

Other reading

• SO
• MR paper on State in Haskell