Musings on mArgazhi contest
Like everybody else, I too was looking forward to see how the contestants fared. My hearty congratulations to the winner. Of course, ikOlam and latAji must be congratulated for the great efforts they put in conducting the event. The 44 contestants as well as the countless judges too deserve a pat on the back. I do not know how many people went through ALL the entries, but I did. I gave 8 five-stars, 21 four-stars, 13 three-stars and 2 two-stars. I must say the quality of the entries is very good. In fact, everyone deserved a bouquet! However, in any competition, there are winners and also-rans. The same entry in another contest judged by another panel may win a prize. Of course, participation is the most important facet. Unfortunately, I could not participate as I cannot draw with powder or wet flour. That is a blessing too as I could enjoy the event without any personal bias.
Here are some of my observations: (1) Why is everyone so much bitten by the colour bug? There were only three entries predominantly white, one of which is extremely complicated to draw. Sorry to say that all these were placed very low. That means these people had a difficult task. They were competing against not only others but also against indifference and neglect. I felt sorry. In the future, it may be a good idea to have a contest only with powder or wet flour and red clay without any colours. Then the playing field will be even. (2) There is a good number of new entrants. That is very encouraging. On the other hand, I find one thing rather discouraging. Some people draw rangOlis throughout the year and upload them whereas some others are active only during the contests. I suggest that as an encouragement there must be an annual prize for non-contest daily upload entries too. (3) While the method of judging is the prerogative of the organisers, I beg to differ with the methodology. In my personal opinion, the number of votes must not play any part in the final assessment. Is it the competitor's fault if others do not view their entries? Either weighted average must be taken or for every contestant, the same number of votes must be chosen randomly. The playing field will be level if this method is adopted. I was really sad to note that the person who got the maximum percentage of five-stars is placed low.
Once again, congratulations to the organisers, participants, voters and the winners. May the ikOlam community thrive and grow from strength to strength!
Regards! - mOhana
Comments
Lata
Fri, 2010-02-05 09:08
Permalink
mOhanaji, I read your quick assessment of the contest-voting process. But, I can't get rid of the gnawing feeling that has to do with answering questions. I think its because I still haven't answered to a couple of your previous questions/comments, posted by you after the last contest. I'm sure you remember. So, I would like to finish with those, before getting to this one.
jayamohan
Sat, 2010-02-06 04:54
Permalink
All ikolam members know that contests are just for activating them periodically. We enjoy the joy of participation. Winning a prize may be a bonus.
jkmrao
Sat, 2010-02-06 12:29
Permalink
This is perhaps my final message on this topic unless otherwise it becomes absolutely necessary. The main argument of my thesis was, and is, is the following. Depending upon the whim and the fancy of each individual, certain rangOlis are visited by more viewers and certain others by a smaller number. That is fine with me. It is the individual's choice. May be rangOli A is more appealing and rangOli B is not thematic. All people, for whatever reason, have the liberty to visit or not to visit a rangOli, to vote or not to vote for a rangOli. I respect that individual right and the individual choice in a free country and in a free world. That is perfectly democratic and corresponds to free will. My only worry and complaint is - in determining the winner and the ranking, not only the value of the votes but also the number of voters is considered. Let me give a hypothetical example. Let us say rangOli A has been voted by 40 individuals and the total stars obtained is 140 (out of 200). The average is thus 3.5. As a percentage this is 70. If one multiplies it by 40, the total value becomes 2800. Let us consider rangOli B. It was visited by 30 people, a comparable number, and the total stars it got is 129 (out of 150). The average value is 4.3 and as a percentage it is 86. Its total value 86 X 30 = 2580. If one takes the average, B is the clear winner and if one takes the total value, A is a narrow winner. The artists who created A and B are not aware how many are going to view their creations. [In principle, it is possible to have a voter drive and influence the total number of visitors. Once again, I am not asserting that such vile acts take place on our respectable forum.] In this scenario, we are penalising B because it was not visited by more people. Does the choice of the average or the total number alter the results? Yes and no. Once it did not and once it did.
Regards! - mOhana
Lata
Sat, 2010-02-06 22:07
Permalink
People in the south make white or plain rangolis on a daily basis. Making colorful rangolis during the Margazhi season has been the norm for the majority of the rangolis enthusiasts. About the 3 entries in white, scoring low; in one of them the picture was too bright, the one drawn with concentric borders might have lost out because of being somewhat low on the crispness/brightness factor, in comparison. And the last one, the ones with paisley-birds, was crisp and clear, with nice strokes, but maybe scored low because of lack of color, again in comparison to the colorful ones (takes more effort and time to color). I think people judge a rangoli based on the effort put into it too. This reminds me of Ms.Hema’s Ganesha dotted kolam from last year’s competition. We know she won for that entry in the Times rangolis contest, for the idea as well for her effort in it (very elaborate with most number of dot counts). Whereas when the same entry was submitted in our site, the colorful peacock done by Darshana won out, with Mr.Sandeep’s Jain rangoli coming in the second (spiritual/holy looks). So, it is clear, that if it is a white design, it has to be a compelling one, in order to stand up to their color counterparts.
I go through mixed feelings after such a contest. I feel happy that there were a considerable number of participants who had taken part to show their skills. And yes, this year we’ve had more number of new entrants. What most members do not know is the amount of back and forth “talking” that goes on behind the scenes, in order to produce valid entries. We are talking about weighted and unweighted numbers for the voting process, whereas I’m still dealing with training most of our members’ in uploading, adding appropriate titles to separate out entries appropriately, and to stick to submissions time-wise. In other words, I’m still building a field here. I didn’t plan on being stuck at the building part for 2/3 yrs in a row, but that’s where I am, so to speak.
According to our initial plan, the voting process should’ve evolved too, over these 2-3 yrs, but progress has been slow. We have to work on revising it, and we’ll consider the points that you’d made. As you clearly state, there are pros and cons to both total value and average value approach. The total value method is the system that we have in place. Any changes to it, would require system level modifications, which I don’t have the bandwidth to do right now. Please thank your colleagues too, on my behalf. We are going to breakdown our future voting procedures into a two step process; and most probably, the first step will be handled by ikolam, where we accept only 50% or so of the total submissions. I will not be doing some of the correctional procedures that I had taken up with our participants, during the previous contests. A considerable amount of time is spent on contacting the participants, and making them re-submit, for various reasons, despite allocating ample time during the entire contest-days. Anyways, more rules are in order, considering all aspects.
I like the bright red elements (asbestos look alike) on this rangoli. It looks good against the black background. But, why do you feel the need to include a design with every blog post? Is it necessary?
Jayaji, Mani thinks the same too
mani
Sat, 2010-02-06 23:00
Permalink
All this talk about the process to use to count the votes and the fairness behind it, reminds me of the current grid locked dysfunctional US senate.
Every state in the US sends 2 senators (that is 100 of them totally), irrespective of the size of the state. And the senators play a major role in shaping the direction the country takes. So, one would expect that the senate body be representative of the people that elects them; at least the majority of the people. But, that is hardly the case. It is some of the small state senators, who really have a big say in what gets done in Washington.
While US, one of the oldest and strongest democracies is still grappling with it, I am sure, the republic of iKolam would get it right, in one of these polling cycles!
jkmrao
Sun, 2010-02-07 03:38
Permalink
If it is of interest to anybody, the whole pattern was done with sine curves.
Regards! - mOhana
jkmrao
Sun, 2010-02-07 04:50
Permalink
Just a note. No big modifications are necessary to get the weighted average. Divide the last column (the total votes) by the number of votes to get the weighted average as a percentage.
Regards! - mOhana