Mathemagenic

Struggling with thinking on finding who belongs to a weblog community for a paper: there is no good way to define weblog community boundaries.

A very cynical perspective - what is included or not in a particular weblog community depends on:

Definition of a community used. There are all kinds of problems here, especially those about distingushing group - community - social network.

Dataset analysed. Including all weblogs in the analysis is almost impossible; in all other cases it's about "where to start from?" and "when to stop?". Selecting weblogs from a specific platform? Those self-selected in a list of "weblogs on X"? Snowballing?

Measures used to identify relations. For example those are likely to give different resutls: community as defined by linking, by common topic, by network of social relations between bloggers...

This post also appears on channel weblog research

Just an example: how selecting units and color-coding for visualisation can amplify one perspective. Is the US really a nation polarised as much as it seems?

Compare US election results:

• Red/blue by state vs. purple scale by state
• Red/blue by county vs. purple scale by county

This is a good example of the case where black and white (red and blue ;) lenzes would do more harm than good.

It's interesting why do we slip into binary thinking so easily?

I'm guilty of binary representations myself. It's so strange: even given my beliefs in complexity, continuums and multi-dimentional nature of personal knowledge management I often slip into binary mode in my texts, making my own arguments vulnerable and stirring polarisation.

It seems that thinking in binary/linear/tree structures (context) is more natural for our brains than embracing complexity, so we need some conscious effort for getting beyond simplification and polarisation.

Lois Ann Scheidt on this in a context of research:

As human beings it is very common for us to look at new ideas, technology, etc. compare them to their older antecedents and then slot them into a linear continuum between two older examples of similar phenomena. By so doing we position the new idea, technology, etc. as somewhat less then the exemplars that anchor the continuum.

[...]

In my own research while I am forced to background some discussions with linear models so I echo the point of view found in published literature, I quickly try to move to more dimensional modeling that symbolizes the complexity of the ideas without making the ideas I am expressing overly complex and difficult for some of my audience to grasp.

See also:

• similar associations & more examples - Theresa Senft on maps and a follow-up by Helena Kvarnstrom
• even more maps