Tuesday, November 25, 2008

Call for papers

Hello Fellow Gilmas,

Gilmaism has been a mystery to a lot of people. Everybody who hears the word for the first time acquires a wierd looking smile on his face for unknown reasons. Why does this happen? Is there a fibonocci spiral somewhere hidden inside the syllables of gilma that give it a strange sound? Is there a prime number pattern that gilma's follow. Below we present the call for paper invitation of Raman's Gilma Research Labs (RGI) on various cutting-edge research topics followed by the celebrated gilma_analysis paper's overview written by the reverend Raman himself.


We shall discuss further details about Raman S in the lecture series that would be published shortly. Here we don't waste any more time and present you the material coming right from RGRL.


------------------------------------------------------------------------------------


Papers invited under the following Categories for first edition of Gilma-Lectures:

1. Gilma Analysis
Here researchers analyse the core theory behind the Gilmas which have been an undefined topic so far due to the complexity it assumes. Here we search for the ground truth behind them, and analyse what feature that they have are having direct implication on their behavior and life. For eg, Is it their hairstyle, way of speaking, looks, etc which make them a Gilma. This will help designing efficient algorithms which correct the relevant features.

2. Gilma Dynamics
This concerns with the interaction that Gilmas have with the outer world, and the insults they get, and their responses. We need a concrete model to study the dynamics, which will enable us to predict upto certain extent what will happen if a Gilma approaches a girl/guy.

3. Gilma Algorithms
Everyone knows that converting a Gilma into a non Gilma is an NP Complete problem. Here we discuss various approximation, probabilistic, hueristic algorithms which will enable te Gilma to find their nearest guy/girl, however attu he/she might be. This also includes algorithms the other way, to successfully transform a Non-Gilma into a Gilma. Every algorithm published should be tested sufficiently on strictly human samples. And authors are requested to share their test samples so that other people can also do the testing on the same samples.





-----------------------------------------------------------------------------------
The difficulty in representing bangalore Gilmas
Raman, Arun, Dinesh

Gilmas[1] have always been a distinct set of their own. Though everyone has a Gilma within, it all depends on the coefficient of Gilma, which measures the extent to which he is part of Gilma. The first step with Gilma programming should involve declaring the gilma variables. The problem here is to find out, whether to declare as Boolean, Short, Int, Float, Long or any other type ? The natural answer of "Declare in any way u like" wont really represent Gilmas. We present a hypothetical reasoning for using Long Long as the data type, which can be the first step in doing large programs with Gilmas.

Background

Without loss of generality, we are restricting the space of search, and using Bangalore as test bed. Bangalore is selected since it has enough examples of Gilma and Non-Gilmas. By observation we see that Gilmas are interacting with each other as a sub group, and Non-Gilmas in a separate group. So, If one Gilma approaches a Single Bangalore Girl, that is more easily corrected by a Non-Gilma guy, which makes it almost impossible for the Gilma to pursue his aspirations. This observation means that the Male Single bangalore(MSB) and Lady Single Bangalore(LSB) are as much separated as possible.

Solution

Hence we propose to declare the Gilma as Long Long, which is very wide, meaning that it separates the MSB and LSB as much as possible. This facilitates in accurately representing Gilma, which can be used in further alogorithms. In Future, we also research in a better representation of Gilmas.

References:

[1] Gilmananda.glogspot.com Arun, et, al.
[2] http://www.orkut.co.in/Main#Community.aspx?cmm=38596396 Arun, et, al.

No comments: