first commit

1 files changed, 71 insertions, 0 deletions

diff --git a/thoughts.schoolties.txt b/thoughts.schoolties.txt
new file mode 100644
index 0000000..bd0f914
--- /dev/null
+++ b/thoughts.schoolties.txt
@@ -0,0 +1,71 @@
+I've been waiting really to get some guidance on what the dataset should be, has the client been given any idea of what questions we will ask the members? It strikes me that the more personal the data is the more interesting a picture can be drawn of it? But its a decision what that data should be.
+
+Moving ahead, I've been looking at the demographic classifications from http://www.statcentral.ie/
+It could also be interesting to compare some of the results with stats from that source - for example averages.
+
+I propose that we work with the following datasets:
+
+Year of joining the school
+Year of leaving the school
+Education level
+Ethnic or cultural background*
+Religion* (still a practicing catholic?)
+Household income* (within bands)
+Marital status*
+Number of children - ages & sexes*
+Children who have attended Oatlands college
+Current place of residence
+Number of previous places of residence
+Years at present residence
+Professional sector (based on heirarchical classification system in http://www.cso.ie/en/media/csoie/census/census2006results/volume8/volume_8_occupations_entire_volume.pdf p39-43)
+Access to the internet
+Access to broadband
+Lifestyle - hobbies, participation in sports
+
+The ones I've marked with a * would be ones that we should think about. 
+We could decide it by ourselves, ask the client or do a little user survey of what data we think we should be looking at?
+There are other areas that we could conceivably look at but I'd imagine that we won't- such as
+
+Health - incidence of serious diseases, psychiatric treatment, incidence of smoking etc
+Access to private health insurance
+Longevity - age of parents, are parents alive
+Children with disabilities
+
+I think the main ways we will look at this data is in the form of maps and timelines that graph the other datasets.
+Maybe a good approach is to utilise the visualisation as the input method also? This might reduce the amount of boring data the user will have to input to participate in the database, making the process more interactive: first the user inputs a data item graphically, and then is hown how the input results compared with everyone elses. Diego, I guess that would break your assumption that the data access model is one-way only.
+
+-------------------------------------------------------------------------------------------------------------
+
+question: on a page which is illustrated around the subject
+transition to answer
+colours: candy
+
+start: ask a question, be shown your answer fitting in
+skip to the graphics
+default ways of graphing the data: by place, by time
+ways of browsing the data:
+	-selecting data/ axes from lists
+	-jumping to the next preset graph w/transitions
+
+requirements of the library:
+	-read JSON
+	-draw scale sprites, vectors
+	-support for graphs, menus, buttons
+
+processing.js: 
++++ I know it
+--- No native charting
+--- Scaleable?
+
+highcharts.com
+--- jquery
++++ pretty by default
+--- difficult to customize? ie to transition between graphs? as it just seems to draw a single graph
+
+	
+design of the program
+
+------------->
+a list of questions
+
+is ie compatibility a strong need?
\ No newline at end of file