The Maami Matrix will identify who is a Maami

There was a rather serious discussion at work today. It had nothing to do with the economic crisis. Nor did it have anything to do with how we should go about our daily activities.

The discussion was fully centered on whom in our circle of friends, was going to be given the "மாமி" title. In our "sangam" in the office, comprised almost exclusively of TamBrahms and an Iyengar, there are 5 women. The eldest amongst them was called மாமி. She announced to us today that she has put in her papers and would be relieved of her official duties by the end of this week.

And in true spirit of our gang, we gave her a simple farewell mail and then got onto business - the business being the need to identify the next மாமி. I asked the 4 remaining ladies to volunteer for that title, but no one did. There was a challenge thrown at us men to identify the மாமி.

That got me thinking. While it is easy to call people names, the tag "மாமி" is usually given to TamBrahm women of a certain age/position etc. The "etc." was what no man has ever attempted to define till now.

I decided to take the serious step of actually listing out the items that probably influence us into associating a lady with "மாமி" moniker. I listed out a few parameters and gave each of them a weightage. I then did a calculation that involved multiplying the weightage with the score in each parameter. These were then summed up. The lady with the highest total was then declared as the next "மாமி".

So here is the calculation sheet for your perusal.



So how does the scoring work? The score has to be in the range of 0.0 to 1.0, with 0.0 representing a true NO and a 1.0 representing a full Yes. For example, if the person is Married and Has Kids, she gets 1.0. If she is unmarried (and without kids) she gets 0.0. Anyone who is married and has no kids gets a 0.5.

Similarly, for the Iyer accent parameter we have to take into consideration the usage of words like "aathukku" more than "veetukku" or "nanna" in place of "nalla" or eases into "nee" very quickly from "neenga" etc. In this day and age it would be almost impossible to get a 1.0 in this scale, and hence this parameter carries a good weightage.

There are a few parameters like "Kid-Look". If someone does look her age and comes across as being much younger, then she gets a rating closer to 1.0. Conversely, if one looks their age (or even older), she gets a score closer to 0.0. But while calculating the weighted score we can’t do a direct multiplication. If you do, someone who actually looks younger would get a higher weighted score whereas she should get a lower weighted score. So, the workaround is to deduct the score from the basis point of 1.0 and then multiply the resultant with the weightage.

After I shared this scoring sheet with my friends, I received a suggestion to include the residential area of the individual also as a parameter. Someone living in Mylapore, Nanganallur, Triplicane etc. would get a higher score whereas someone living in say Anna Nagar or Chengulpet is likely to get a lower score. We can probably assign a weightage of 5% to this parameter.

If you have any parameter in mind that can help us arrive at a more robust and scientific model, please share the same.