What should I do ? I'm a worthy person but I need to take a break. And Sanmitra won't answer my calls. I need to take a break in some other way. I could call Phatak.

OK, so I need to get this issue sorted out. I have a problem in setting goals. My goals are pretty vague – e.g. I want to get married, I want to get a job, I want to get a girlfriend etc. – not specific – like I want to marry this girl, I want to get this job and so on. This causes me problems I have come to realize. So what is the way to get past this?

I can start of by focusing on this job at Apple.

I am hugely relieved to find that the course Formal Methods in Software Engineering does indeed satisfy breadth requirements. I should have killed myself had it not. Tinelli gave me unwarranted stress just minutes before class. Wonder if he comes up with his fake British accent again today.

Yelena is sitting with Eric and I don’t like it. I wanted to ask her to be my project partner. Oh well.

OK, so to decide what to start off with.

OK, I decided to start off with TA work – Padmini’s reminder email virtually left no choice. I was going to begin with that anyway. Right looking at the crawler code. Must remember there is a lot for me to learn here as well. First the importance of a good readme file. Placement and visibility of instruction comments. Precise instructions on how to run.

OK this thing works but the question is how to interpret it. What is the Maximum Frontier ? It is the maximum allowed number of URLs from a page that are yet to be crawled. Here it is set to infinity. Then it says Frontier: 1. It means there is only 1 seed element in the frontier. i.e. http://www.uiowa.edu.

OK I need to refocus on what I need to do. Padmini has asked me to

a) Go through the crawler code

b) See if I can run it

Well, a) seems harder. Reading the README file it asks me to make some change in the Tester file (email address) and clearly recompilation is an issue.

When I try to run it, it does run to completion and spit out the stats but gives a couple of OutofMemoryExceptions while creating threads. It is hard for me to figure out what exactly the problem is.

Question is – do I have enough to get back to her? I can tell her that I looked at the code.

OK, first pass done … question is now what do I do? I need to understand the code better.

OK I need to get up to date with compilations and stuff with java.

It is time to turn to research work. I had a meeting with Kasturi today. He seems intent on going into the details of the bad example. This should be interesting. Let me look at the bad example. Oh crap! I may have been wrong again! Perhaps not, I have the figure now.

The plan for today was to get up to date with research work. What do I mean when I say I want to get up to date? I want to do enough to be able to report to Kasturi on Monday. OK, I feel full, I shouldn’t have had that much root beer. OK, back to work, what will I tell Kasturi? What do I want to tell Kasturi? I will talk about the extension of the bad example to the general case. And I will talk about use of combinatorial methods. I will talk about a) finding the shortest path from x to its infinite exterior and then for each disk on the shortest path finding a shortest path along the belt from one end of the disk to the other. But this is what he mentioned last semester. What is new here? Honestly, nothing. I have to tell him that all I could come up with was what he had already mentioned and that I don’t see any problem going in this direction.

But that is not good as I should have done some work before. Well, I haven’t.

What are his objections likely to be? They are most likely to be that, well, what about the case where the point may be contained inside a disk? Well I want to tell him I want to handle that as a special case and then focus on this one. This is sure not to please him.

There sure was something else that was a gap in what he said last semester. It has to do with figuring out the properties of essentialness etc. of the cycle w.r.t the surface at hand.

So what do I tell him? I definitely feel that for Monday I should focus on presenting to him better rather than trying to do anything new in a rushed manner.

So here are the points I want to talk to him about:

1) Bad example

For the bad example, I want to first remind him of the bad example we discussed. What was it? It was showing that a path $\pi$’ could enter a disk several times. OK, the figure is drawn. The bad example is recollected.

2) Combinatorial Methods

As far as combinatorial methods go, I want to begin by talking about the paper – what paper? – the Jeff Erickson paper. Say I went through it, but the main idea seems to be finding shortest paths from which we go on to find the shortest cycles and then ensuring the essentiality property. Say that I haven’t got the details of this work; but I remember discussing the shortest path approach before we left for break and would want to start off by recollecting what is the void in this particular approach.

3) Also maybe it might be an interesting idea to go home and look at the notes I made in the small pad which had some slightly mathematical content.

Ask yourself why you are interested in doing a postdoc.

Every time I start writing something like this I feel sleepy.

It is weird - if I THINK I don't feel sleepy, but if I WRITE I feel sleepy.

I want to also feel good and feel sexy

Do I REALLY want to have sex?

I also want to have sex.

I will do it.

I have to do everything on my own

kasturi is not fucking helping me at all

how can i get out of this place ?

I have this in front of me - my work stuff - fuck I hate my work. I am in an uncomfortable position. I should be at my desk. Or at least get my laptop closer. Fuck my sofa. It is getting old. I really really need to get out of this place - and soon.