Deciding categories seems pretty important. We could run them along the lines of committees and categories that city hall might recognize, but we might discover more interesting segmentations by analyzing historic votes for indications of what has most controvery. We could rank historic votes by some object measure of "vote split", and see if we could figure out what councillors in toronto tend to get most divided on.
These are also the votes for which they might be more likely to consider the responses of their electorate, and it would also lead to feedback to citizens that made them feel "oh man, that was close -- if only my councillor had listened".
Deciding categories seems pretty important. We could run them along the lines of committees and categories that city hall might recognize, but we might discover more interesting segmentations by analyzing historic votes for indications of what has most controvery. We could rank historic votes by some object measure of "vote split", and see if we could figure out what councillors in toronto tend to get most divided on.
These are also the votes for which they might be more likely to consider the responses of their electorate, and it would also lead to feedback to citizens that made them feel "oh man, that was close -- if only my councillor had listened".