Improving existing visualizations

Aman gave some feedback on improving the existing visualizations on….

Aman: I’ve just copied-pasted the content of my e-mail here and we can take it from here. To figure out which tasks are more urgent / juicy / “low-hanging fruit”…
Broadly, I divided my comments into those concerning the phenomena being considered (which, how, time periods, space – where , who and to whom), and on the other hand, comments on visualization, suggesting how to make certain interactions more intuitive or informative.


  • Political timeline:
    • New event categories in political timeline: tax policy, international trade agreements, aid and national debt (foreign and internal), industrial policy, environmental policy, military procurement and recruitment, labour policy, energy policy, “agitational” events (by which I mean things like labour strikes and factory occupations, “payya-jaam” attempts by parties and movements, marches, rallies and dharnas (again, by parties and movements) )
    • Adding event data to the political timeline data file
    • Augmenting the information about the event: links to multimedia, recommended reads (likely to require some tweaking to the way the event info is presented)
    • Event scoring: adding components to the scoringĀ  (could be: economic significance, change in power relationship between rulers and ruled, change in relative strength of military and civilian power, change in balance between major national institutions, constitutional rights (freedom of speech, association, religion, checks on abuse of power by law enforcement institutions), socio-economic rights); some components may apply only to certain event categories; interface allows user to modify weights of components and final score is normalised sum of products.
  • Sindh Health and Literacy Comparision by District
    • Health workers per capita to the information on each district
    • Is data on numbers on midwives and trained “traditional birth attendants” (daiees) available somewhere? Shirkatgah or FPAP might have compiled such figures.
    • Related to the health rubric: mortality rates, prevalence of indicative diseases, some measure of transport availability, the statistics bureau report doesn’t have figures on nutrition/food consumption (only production) which is a pity,
  • Slums
    • Migration by reason: somehow it seems more intuitive to me to have the reasons on the left and the destinations on the right (unless they’re places of origin and not destination?)
    • Migration by place of birth: Punjab, Sindh and NWFP all have flows where the place of birth is the same as the province. Could this be a data entry error? Or maybe there’s something I’m missing.
    • Evictions and guttings: it would be so revelatory to have data on the characteristics of the projects and populations that replaced the evicted people. Kind of project (residential, commercial, military and/or governmental), income levels of new residents, new population density, new resource consumption levels, political tendencies of new population. But that would require some serious survey/report-trolling work.
  • South Asian vulnerabilities
    • Ecological indicators (forest cover, endangered animal and plant species, frequency, intensity and extent of flooding, water-carrying capacity of major rivers, encroachment of the sea into river deltas, extremes of temperature, precipitation and humidity, frequency of storms and cyclones)
  • Transport indicators (by country and province, by mode of transport): some measure of network density, passengers carried, km per capita, road vehicles per capita, cost / km / person on most heavily used routes, average travel time on most heavily used routes, percentage of population in areas with least service
  • Indexing Naked Punch articles (!)


  • Political timeline: possibility to select multiple categories, possibility to use a slider to specify the time window to filter events.
  • South Asian vulnerabilities: ability to flip the variables and the countries, so that one may choose multiple variables (as long as the scales are compatible) for one country
  • More work like this awesome visualization, i.e., linking phenomena.
  • Really advanced: animate spatial visualizations to depict changes over time, happening across regions and continents!