I’m starting to wonder why I was planning to use Java for this project–most of the SolarSoft stuff is already in IDL. And, it seems like writing my stuff in IDL would help it to interface with future additions to SolarSoft. Sure, this means learning a modicum of IDL, but I should be learning languages in my free time anyway. Plus, IDL has a Python API if I really need to revert to a familiar language.
November 15, 2009 reading notes
“The fundamental unit of analysis is the dyad, i.e. a pair of vertices which may be adjacent or not. Dyads are classifies according to various types of adjcanecy, distance, connectivity, and other traits of direct and indirect connectedness”
–Brandes, “Social Network Analysis and Visualization”
TODO: find what architectural style is HEK
network inference
The more I think of it, the more finding similarities in solar phenomena seems like finding similarities among people in an on-line social network. TODO: find some academic papers on graph theory re: Facebook and the like.
Posted in Uncategorized
Tagged cliques, Facebook, graph theory, graphs, network theory, networks, similiarity
Leave a comment
brainstorm: possible OOP hierarchy
abstract class SolarEvent has sub-classes:
specific solar events, e.g:
Flare
AR
Sigmoid
Sunspot
——
Abstract class Values has sub-classes:
specific types of values, e.g. location, time, human- or software-tagged, etc.
——
the specific events like Flare and Sigmoid will reference specific value types
——
example code:
public class Flare extends SolarEvent{
StartTime thisStartTime
EndTime thisEndTime
ArNum thisArNum
//etc
}
yes, naturally I am using my “birth language” of Java
Posted in Uncategorized
Tagged brainstorm, class hierarchy, classes, inheritence, Java, oop
Leave a comment
network theory
Network/graph theory is a powerful branch of mathematics that seems to pop up in so many interesting but, on an un-abstracted level, dissimilar systems. I have though somewhat about this project in terms of graph theory, possibly using an object-oriented paradigm to implement it as a weighted pseudo-/multi-graph with objects representing nodes and links. Nodes would be specific events and/or specific quantities, e.g. location on the solar disk. Links would be weighted by some meassure of similarity. Cool thing about using a graph theoretic mindset is that there are many other examples I could draw from, e.g. social network graphs on Facebook.
Posted in Uncategorized
Tagged brainstorm, graph, graph theory, network, network theory, oop, social networks
Leave a comment