Two things.
- I'm not Nolan, so this story is going to be bad.
- Next, this might be my last rant, since I'm getting out of dev/null in a few days.
My project is going to be big. I will never even go remotely close to a Project centre. I have time. I have a computer that I could talk to. I should build something incredible and call it my project.
- http;//google.com -> Greatest graduation projects
- http://news.ycombinator.com -> What could be the final year project of a Geek?
- http://joelevis.wordpress.com -> Dreams of my final year project
What's this projects list catalogue that these project centre guys give away right inside my classroom? Isn't it something strange?
So, I get invited from my friend to select a good Project from a centre for him and I go with him.
That day, I went to around 6 centres and all I got is a little afraid and nothing else. Because, projects were nothing like I thought.
There weren't any Javascript frameworks as Projects. No Django Apps as one. No mobile app bootstrap technologies or cross platform App development fraameworks as one. No, they were nothing like them.
Most project titles started with 'Automated..' and ended with '..system', labeled with domains. We weren't asked what we would like to do. We were only asked 'Java or .Net?'
The day was a nightmare which I'd never like to think of.
That day, I knew why Microsoft or Google or Facebook never sprung up from Chennai, India.
That day, I realised how project centres and colleges together generate lambs for TCS and Infosys.
Ok, big day. They are asking names for project teams in college.
I proudly make a vow to set a difference and call up #Benjamin and tell him that we three along with #maharaja are gonna register separetely as individual projects. I tell him that we'll do these together and we do projects on our own, three of them.
[#Benjamin] always was a lot geekier than most of them. almost all of them. He just never knew. we used to talk about how our project should be like. He hates getting my help and vows to do on his own. we talk about how broken the college system is and try to come up with something based on it. but we never knew it would remain a talk. I knew it when we recieved the first mail regarding our project.
The first mail regarding project hits my inbox.
It never had any details as to how useful or different it should. It didn't contain any words of encouragement not to go to project centres.
It had instructions.
Instructions on how small the font of the report should be. Soft Copy details of abstracts of our projects. And, more about them.
That day, I was scared more. I knew it's not going to be what I expect.
By the time the second mail hit , I was convinced that I'll only be able to code things. I never knew what IEEE papers are. I hated doing documentation works, which ofcourse were nothing like real documentations.
Lots of ppt slides and lots of papers. So, we three reconfigured our minds, went back to college and registered as one single project, all three of us together.
Still, I had hope that we could build something and get away with documentation works with a little help from my friends. but it wasn't like that. There was this thing called IEEE papers.
We had to choose a paper and work on it as a Project and it was compulsary.
That day, I knew my project is not even going to be half of what I thought of.
That day, I knew why yuvi mentioned about fighting with faculty to get my thing done in his mail to me.
Situations changed altogether. I rolled up to an Internship programme @ iThoughtz and I was too busy to fight with faculty and come up with something big.
My friends had meanwhile talked to our faculty guide, and she has given us a project to work on. It was something related with detecting emotions from textual data. Thats all I could make out then.
Though I went mad for letting her choose, I never had a strong project idea to fight against. One fine day ,myself and Benjamin planned to come up with a good project idea and pretend it to be an implementation of an IEEE paper and tell our guide that we could not work on her project.
But, the sketch failed. We never came up with a solid one and I was too egoistic to tell her that we couldn't do her project.
That day he went mad at me, and I decided that this idea sounds pretty good. If I could dream of writing a compiler, this is nothing for me. And I began seeing usefulness out of it. With emotions detected out of text, we could do lots of things. Get emotional quants out of the current playing song and suggest songs based on it? Get their emotional state from the messages they send and give them something? This and there were lot more.
I started reading the IEEE paper for the first time. It used a Bayesian Classifier, a supervised learning algorithm which when modeled will learn and classify spams apart. Well, the classifier could also be used for labelling documents with emotions as well.
It was not the end of it. While continuing reading with a little more enthusiasm, I confronted a big long equation, something related with LDA. That moment, everything messed up. I could never imagine what a drichilet process would be like. I never had much convolutions in my brain. I couldn't digest higher order mathematical models all of a sudden. I was not good with numbers, but definitely I was easy with thinking mathematically. I knew how recursivity is a power that lisp utilised. I could relate bayesian prior with likelyhood of an event. But not this. This was something far more. I knew the concept is far away from my grasp. It was something, that needed me to brood about all day and night, which I couldn't. Cruelly speaking, I just had lots of other things to do.
I knew that my area was building something using this and not building this itself. So I started my search for any implementations of any kind of textual emotion recognition algorithms.
As days rolled by, accidentally, I found this open source implementation of emotion recognition from text, which itself uses a decade old wordnet based technique. With wordnet, I could have written one easily myself, but I just knew it wasn't worth a shot. So, I found it, stole it and ported it and called it my final year project, shamelessly.
But there was one thing I still was proud of. We never stepped into a centre. I convinced myself that this libary exists so nobody has to re-invent the wheel. I ported it to android and used it to create an Application that scraps news articles from sites and presents them classified on emotion. I spent a couple of nights finishing it and found it elegant.
I was not there when #Benjamin and #Maharaja showed the Application to my guide. I was sleeping at home, bunking examinations as I always do. I got a call, from them and when I rushed to college, I found out that she was not satisfied with what we did.
He said she wants to see input and output. yes, just like our lab programs.
So, I felt like first I have to show her an input out interface and go on to build a better implementation.
We went to our so called R&D Lab, took a few minutes to write a Swing GUI to input a text and print informations along generating an emotion label for the text and presented it to her.
Few more minutes of general questions and she said it's enough. SHE SAID IT'S FINE. THIS.
An Android App didn't amuse anybody there, as just an input output interface did.
Everything was strange, we showed it to another faculty who asked completely irrelevant questions and laughed on us when we said we can get lyrics from currently playing songs as input. He chuckled as he was proving how songs are based on mp3 files and we never could get lyrics from them. He does think this rationally, at times ;)
That day I remembered how I started and how it ended. That day I knew I would be writing my last rant.