Better, I think, would be an approach that is more routine and less time-sensitive in prioritizing among responders -- the thing that would differentiate demand would be people's *own* prioritizations, in relation to *all other* possibilities for demands. This means that only those most focused on Product 'X' or Event 'Y', to the abandonment of all else (relatively speaking), over several iterations (days), would be seen as 'most-wanting' of it, for ultimate receipt.
My 'communist supply and demand' model, fortunately, uses this approach as a matter of course:
consumption [demand] -- Every person in a locality has a standard, one-through-infinity ranking system of political demands available to them, updated daily
consumption [demand] -- Basic human needs will be assigned a higher political priority by individuals and will emerge as mass demands at the cumulative scale -- desires will benefit from political organizing efforts and coordination
consumption [demand] -- A regular, routine system of mass individual political demand pooling -- as with spreadsheet templates and email -- must be in continuous operation so as to aggregate cumulative demands into the political processhttp://www.revleft.com/vb/blog.php?b=1174
I'm also realizing that this model / method of demand-prioritization can be used in such a way as to lend relative *weight* to a person's bid for any given product or calendar event, if there happens to be a limited supply and a more-intensive prioritization ('rationing') is called-for by the objective situation:
Since everyone has a standard one-through-infinity template to use on a daily basis for all political and/or economic demands, this template lends itself to consumer-political-type *organizing* in the case that such is necessary -- someone's 'passion' for a particular demand could be formally demonstrated by their recruiting of *others* to direct one or several of *their* ranking slots, for as many days / iterations as they like, to the person who is trying to beat-out others for the limited quantity.
[A]ggregating these lists, by ranking (#1, #2, #3, etc.), is *no big deal* for any given computer. What we would want to see is what the rankings are for milk and steel, by rank position. So how many people put 'milk' for #1 -- ? How many people put 'steel' for #1 -- ? How many people put 'milk' for #2 -- ? And how many people put 'steel' for #2 -- ? (Etc.)
*This* would be socially useful information that could be the whole basis for a socialist political economy.
So, by extension, if someone was particularly interested in 'Event Y', they might undertake efforts to convince others to *donate* their ranking slots to them, forgoing 'milk' and 'steel' (for example) for positions #1 and/or #2. Formally these others would put 'Person Z for Event Y' for positions 1 and/or 2, etc., for as many days / iterations as they might want to donate. This, in effect, would be a populist-political-type campaign, of whatever magnitude, for the sake of a person's own particularly favored consumption preferences, given an unavoidably limited supply of it, whatever it may be.