Categorised to a tee

[nmg]

Consider today's XKCD.

For the record, I like Isaac Asimov and XML, I've been thinking about buying a pair of Vibrams for the last year, I think that the Segway looks neat (and, as a child, quite the sort of vehicle which I was led to believe would be in my future), and I've been looking for an affordable head-mounted display for the best part of a decade. And yes, my favourite map projection is Dymaxion.

(although not because of Bucky Fuller, but rather because of my youth misspent on certain games)

Comments

[unwholesome-fen.livejournal.com]

Having worked in GIS and read around the subject in far too much detail to be healthy, I don't have a favourite projection. So I suppose that probably puts me into the globe category.

Also, FWIW, having worked in SGML for another job, I don't think much of XML. Not really sure what it's doing in that list, though.

[bohemiancoast.livejournal.com]

If you were writing an article where you wanted to include some maps on say the global incidence of poverty, rhematoid arthritis, budgerigars, iPads and animated conversations about maps, which projection would you include for those diagrams?

Note that if the answer is 'it depends what data I'm representing' I think there's a big risk of bias there.

[unwholesome-fen.livejournal.com]

What I was saying was that all projections have their flaws, so I wouldn't really be happy with any of them. In your hypothetical scenario, rather than depending on the data, I think it would depend more on the audience. If they were unfamiliar with the issues around projections, that would be problematic, to the extent that I might not want to use maps at all. For a more sophisticated audience, perhaps various different projections and copious footnotes. Or more ideally, an interactive GIS system where they choose their own projection(s), with appropriate caveats issued. That's barely scratching the surface of the issue though.