Of course, I’d love to see full details, with all personally identifying information, for every Canadian census ever taken, up to and including the day before yesterday. But: I also realize there are genuine privacy concerns relating to the public release of personally identifying information.
The fact is, you can learn a lot from the recorded facts of the census, if you pay attention to the details. “Illegitimate” births; unofficial (and perhaps unacknowledged in the adoptee’s lifetime) adoptions (including, e.g., British Home Children whose birth names were erased/subsumed under the surnames of their adopters, or, perhaps, employers); “widows” who were not actually widowed; bigamy … look, I don’t mean to suggest that the census is just a big scandal sheet in tabular form, but I’ve seen examples of all of the situations just mentioned, and more, in currently publicly available Canadian census returns.
About a month and a half ago, when the US federal census of 1940 was first released, people were posting pages from the 1940 on facebook: ‘Here’s my grandmother in Detroit, Michigan!'; ‘Hey, Dad! Here you are, just five years old!’ Needless to say, I felt a pang of, well, envy, I suppose: under current rules and regulations, it won’t be until 2033 that we see the public release of the 1941 Canadian census, after all, and even the release of the 1921 Canadian census is a year away. At the same time, though, I couldn’t help thinking that 1940 is awfully recent (in historical terms, it is almost the day before yesterday): still well within living memory, and, indeed, a date that many people now living have actually lived through.
On the other hand, I have very little patience with the idea that the census returns should be destroyed just as soon as the data has been extracted and compiled into socio-economic profiles of the aggregate. I want to see the details; I think we should all have access to the details, once a certain amount of time (72 years? 92 years?) has elapsed. My perspective here has admittedly been coloured by the loss of the 19th-century Irish census (which, sob! I’m still not over…): a great loss not only to Irish genealogy and family history, but also to Irish social history and demography (but see Fiona Fitzsimons on Griffith’s Valuation as a census substitute).
Is the American 72-year rule too lax? is Canada’s 92-year rule too strict? What is a reasonable middle ground between access and privacy?
UPDATE (7 June 2013): Poll removed due to technical issues (basically, the poll plugin was slowing down the site).