[Reader-list] Gap Minder

[Isaac souweine]

Dear Taha,

Thanks for your enthusiasm and sorry for the late reply - I just finished
final exams.

Anyway, to start off, here's what I understand as the "bad case" for
visualization technologies, drawing on what you write below. Statistics
themselves do not contain truth per se. Rather, they are simply a pallete
with with which we paint whatever truths we find convenient. Such stories
are invariably told more effectively by the powerful. Therefore,
visualization technologiers just represent another tool in the arsenal of
oppression and power. Damn lies told with a bit more flare. The soft version
of this thesis simply calls into question the conclusions based on
statistics, i.e. their official intepretation. The hard version calls into
question the method for defining or gathering statistics, i.e. their
construction. The extremely hard version disputes the role of statistics in
effacing individuals.

Now, here is what I understand as the "good case". The modern world produces
phenomenal amounts of data at every level. Within this data is contained all
sorts of truth - scientific, sociological etc. Accessing this truth is
itself a vexed problem that requires sophisticated technology and training.
Rendering these truths then sensible for larger audiences is further
complex. With the advent of technologies like Gapminder and really with the
whole revolution is information design that we see daily displayed on
websites everywhere, fundamental facets about our world are made more
sensible. Implict truths become explicit when places in time sequences or
rendered in creative displays. In the political context, important issues
are thus opened more clearly to debate and discussion by a broader array of
people. In the classical Millian liberal formulation, public discourse
produces public goods.

To proceed, I think we would need to dispense with the hard and super hard
versions of the bad case, since otherwise visualizations are fundamentally
flawed no matter who makes them. To make this move, we have to privilege the
picture you paint in your first paragraph, which involves intepretation and
counter-intepretation of statistics in Britain, rather than the arguments
made later about rationality etc. I'm guessing that this might not sit
entirely well with you, so I appreciate you humoring me.

If we do make this move, we are left with a question of who will benefit
more from visualization technologies - the "powerful" or the masses. In
other words, whose narratives will be more amplified? Considering the larger
social patterns we have seen with information technology, I would feel
fairly safe betting that the masses will win this battle. So, for example,
imagine the power of Gapminder coupled with data sets provided by
governments, like these: http://www.data.gov/. Isn't this setting the stage
for a time where more and more people can craft their own narratives and
then insert these narratives into the political process? In other words -
Gapminder as the visual soundtrack to an age of information dissemination
and democratization.e

I know I've sort of skipped over the epistemological and aesthetic nuts and
bolts of data visualization, not to mention many issues that you raised, but
perhaps this gives a very simple version of what I am thinking about? I hope
you find it interesting.

Best,
Isaac


On Mon, Jun 15, 2009 at 10:14 PM, Taha Mehmood
<2tahamehmood at googlemail.com>wrote:

> Dear Issac (and All )
>
> It is nice to see some posts on emerging 'nice/pretty nice' ways of
> looking at data. In this regard, the overlap of aesthetics and
> epistemology is indeed interesting.
>
> Some preliminary questions came up while thinking about these issues,
> I have tried to list them as follows-  for instance, I do not know how
> notions with respect to truth claims or validity of representation
> could be reasonably dealt with a certain amount of confidence while
> using quantitative data? This is a contested territory both
> epistemologically and aesthetically.  Even in a decision making
> process which relies on quantitative data one wonders how questions
> related to ruling ideology of the day and its role in molding
> seemingly 'rational' decisions are addressed? For instance take the
> case of policies related to urban planning in the UK. For the last 60
> years official statistical organizations have been churning numbers
> which suit the ideology of the day. Hence in the name of
> Reconstruction, Revitalization, Renewal, Redevelopment and
> Regeneration, the policy makers have been able to create a seemingly
> rational rhetoric to push their agendas. While in the hind sight many
> so called experts have successfully refuted the claims of such
> propositions and were able to generate a counter imagery using almost
> the same set of numbers.
>
> Do these visual representations by their very nature not tend to iron
> out various underlying variables which make up this data? Do all
> political process everywhere appeal to a so called 'rational' form of
> decision making process? How should one approach the question of
> 'rationality' in political decision making? Is there some universal
> system or process through which political discourse or policy
> decisions sprouting from such a discourse could be referred to and
> could one conclusively locate 'rationality' in such a process? Does
> questions related to data and its visualization or representation
> appeal to only political structures and leave no room for some times
> 'irrational' agency of operators, agents or actors?


>
>  In this  regard, Isaac, could you please articulate a bit more on
> your claim, that such data could be  'a boon to political discourse
> and rational decision making'. This claim might be true in some places
> but from a south asian perspective I  think one would hesitate before
> advocating a blatant use of data to boost political discourse or to
> make 'rational decisions' :)  For instance, on a broad level, in the last
> thirty years at least, in
>
a country like India, we did not have a single instance of mass
> political mobilization by some political dispensation which was able
> to use some correlations based on 'nice/pretty nice' data or analysis
> and make a serious bid to power or emerge as a serious long term
> player.



> At the same time I think numbers are an essential tool for governance
> related issues, or to analyze the purported impact of a policy in this
> regard I do not think I would have pitched my tone to the tune of
> 'lies, counter lies...'.
>
> Having said that, I would be very interested if you could please share
> with us any instances where data or resultant correlationships were or
> are being used to aid either 'political discourse' or 'rational
> decision making' process. In this regard maybe you could also
> elaborate more on the overlaps of aesthetics and epistemology. It
> would be indeed very informative to know more about ways of looking
> and think about questions related to visual rhetoric and claims to
> valid forms of knowledge about society.
>
> I look forward to your reply.
>
> Warm regards
>
> Taha
>