Why? All we are talking about is taking note of the first measurable snowfall of the climatological winter season at a particular location. For example South Lake Tahoe, California, USA had its first measurable snowfall of the 2011-12 season on October 3rd of this year (about 10 cm). That's fairly early even for this location and might be of interest to some people. That's all we are talking about.
Climate atlases of the USA have contour maps for the average date of the first killing frost and for first hard freeze. Such a map might be constructed for the average date of the first measurable snow based on date spreads. It's really a very simple concept. Some areas will be outside the snow zone entirely and some areas might have recorded measurable snow in every month of the year, and can be so indicated.
EDIT: I re-read the OP. Obviously that's not a provable statement, particularly in considering that in the Northern Hemisphere the first seasonal snow in the temperate regions will usually be in Oct-Nov-Dec and in the Southern Hemisphere in April-May-June. In places where it can snow in any month, there is no such thing as a "first snow" except as defined arbitrarily.