These statistics are relevant to Paul Graham's recent essay. They show that the number of firms with large numbers of employees is growing faster than the number with small numbers of employees. The years covered by the data are 1988-2001; it would be better to have more recent data, and also to have data broken down by field. It's possible that the US Census Office (which is the source of these data) has more recent figures, but I haven't checked.
With these limitations understood, the data here still contradict the essay's assumption that there is a trend toward smaller companies.
Thanks. I had submitted a bug report, but didn't realize that it was the ticks that were causing the trouble. They may want to fix that anyway though...
and Here's a visualization using the time-treemap to compare sector growth: http://manyeyes.alphaworks.ibm.com/manyeyes/visualizations/t...
and here's a stacked area graph that shows it all at once, although it's not as nice for comparison purposes: http://manyeyes.alphaworks.ibm.com/manyeyes/visualizations/u...