A small detail.
I've read several articles recently berating country X for being ranked position Y in broadband penetration, where Y < Z for some values of Z (all of whose names are alarmingly foreign sounding).
It doesn't matter, these tables are bunk at that level of granularity. Yes, the "big picture" stuff is true, and developed countries come up high compared to developing ones. But get real, and re-read your stats books.
Many small companies like to pronounce on their explosive growth, forgetting to cite the law of small numbers; the day you got your second user you were experiencing 100% daily growth, which compounds to an annual rate of over 3 bazillion.
Likewise, the opposite occurs when you get to big numbers. If we separately ranked US states, or divided the UK up into regions, the picture would change radically. London and southeast England would look great. So would a few US states like Virginia.
Some of those little regions with greater connectivity have familiar name tags. Like 'Norway', and 'Luxembourg'. Yes, they're real countries. (My wife comes from a small, real country, with a real language and real culture, so I know how sensitive people are. Don't worry, if size matters, it's only to statisticians. Mine is only bigger than yours when viewed through a spreadsheet.) Yet their position in the rankings is an artificial by-product of how we divide up the statistics. It means nothing to the median European or North American.
If North and South Korea had a massive reconcilliation tomorrow, and thus the Unified Koran Republic plummeted to #102 in the world, do you think things are getting better, or worse? (Hint: begins with 'b'.)
Furthermore, such comparisons barely begin to hook together the bigger picture of mobile penetration and adoption, 'bit literacy', barriers to commerce, etc. Each country is on its own cultural and economic journey from 'here' to 'there'. Each started off from different 'heres'. A better measure might be comparative rates of increase of adoption from normalised base levels. But really, there is no simple index of connectedness. It's a complex, multi-dimensional problem.
So personally, I won't be losing any sleep over the matter.
“Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted.” -- Albert Einstein
PS - If we found a negative correlation between country population and broadband penetration, the same naive reasoning would lead us to recommend either massive secession and regional government, or simple genocide of people without PCs to help the stats. Sounds silly? It's the same statistical logic that makes Americans, Britons and Chinese read endless articles on how they're 'falling behind'. It sounds silly, because it is.
Posted by Martin Geddes at 9:46 PMTrackBack URL for this entry:
http://www.telepocalypse.net/cgi-sys/cgiwrap/mgeddes/MT/mt-tb.cgi/740