Arnav's Blog

New Semester Notes

2006-01-12T19:43:00.000-05:00

A new semester has started. We have moved to a new building and I like my new office (will get a pic and post it here soon). It is good to have all of my classes in the same building for a change.

This semester quite unlike many others during my PhD, is going to be pretty hectic.

I am taking two courses this semester:

CSC707: Automata, Languages, Computability and Complexity Theory

This is an advanced course on this topic and satisfies the last remaining course requirement that I need to meet for my curriculum.

CSC720: Artificial Intelligence II (Rationality and Reasoning)

This course is taught by Dr. Jon Doyle and is mainly focused on Rationality and Reasoning in AI.

Apart from these, I am teaching Unreal Engine Programming in the Game Design and Development course for which I am the TA.

Two conference deadlines are also coming up so working away to get the papers finished:
AIIDE: Conference on Artificial Intelligence in Interactive Digital Entertainment (25 Jan) and
AAAI: Conference of National Association for Artificial Intelligence(21 Feb).

Summer Placement

2006-01-03T07:28:00.000-05:00

I recently got an offer for a summer position at Virtual Heroes Virtual Heores is a game studio best known for the game America's Army multiplayer FPS.

I visited their studio which is located (conveniently) only about 10 miles from my apartment. I met some really cool young game programmers, designers and artists chugging away on their high-end machines. I am really excited about this opportunity. I am gearing up to get my hands dirty with some real game code (rather than mods and research systems that I have worked on so far).

Introducing Darshak

2005-11-18T02:32:00.000-05:00

Darshak is the name of a Virtual Movie Director (yea, technically it should be DigDarshak or something, but it is complicated when converted to English) that I am writing as part of my dissertaion. Darshak will feature an AI planning algorithm with operators based on Cinematic Idioms. It will soon be tied to a modified Half-Life 2 game engine for execution of these camera plans that it will generate.

Here's the logo:

Shoonya

2005-11-18T02:26:00.000-05:00

Shoonya is a new desktop I got yesterday. I set it up as a server at home. It runs Fedora Core operating system and is powered by Apache.

I plan to use it for accessible online data storage purposes and also for doing little web-based projects. This also gives me a chance to run my own customized blogging software, rss reader software etc.

Chill sets in

2005-11-17T09:08:00.000-05:00

We are almost at freezing temperature today. Yesterday though was about 20 degrees higher. Weather here has been getting more unpredictable by the year. In 2001 I clearly remember the Fall season with bright colors (red, yellow, purple) replacing the thick green around the campus. Walkways full of dry yellow leaves with pearls of water on them shining in soft sunlight and a light comfortable breeze blowing. Gone are those days, we completely skipped fall last year. This year has not been better either. Due to the record number of hurricanes due south of us, the cold wave from the north west that usually brings rains probably got pushed back. So now we are in a near drought condition with water levels way below par. We did have very nice weather last week, but my plans of going around the campus with a camera to catch the fall colors are now frozen until(hopefully) the chill goes away.

Book Sale

2005-11-17T05:29:00.000-05:00

For all those in and around the Raleigh area. The Wake County libraries is holding their annual Book Sale this weekend. Last year I bought about 4 boxes full of books (mostly fiction - complete serieses of Robin Cook, John Grisham, Michael Crichton, Frederick Forsyth, Douglas Adams, Issac Asimov, Mario Puzo, some management books, a couple of books on Hitler and the world war, some biographies, a couple of science books, one asterix comic book and a Dilbert book). In short, they have a great collection of books and they sell them for $5 a box!!

For more information visit:

http://www.wakegov.com

Cricket, GarageBand etc.

2005-11-17T04:39:00.000-05:00

I am really enjoying the audio commentary of the cricket matches going on in India. After gaining a lot of confidence from their 6-1 series win against Sri-Lanka, India now face a much stronger opposition in SAF. It will be interesting to see how the 'new' team copes with this challenge. I am impressed by the way the tailenders played in the last match, but a little disappointed with the bowling in the opening overs of the SAF innings.

The England Pakistan test was the perfect example of why I love test cricket. The match was quite evenly poised at the beginning of Day 4 with england needing about 175 with 9 wkts in hand. Thanks to some aggressive and accurate bowling, Pakistan crawled in to the match and by lunch they were on top. At some point Geraint Jones with Giles and Udal did give a scare of running away with it, he eventually fell and Pakistan won.

On the other side of the globe, Australia are (as usual) on a killing spree as they bowled out West Indies on the first day (with no signs of giving away any wickets when they came to bat). All I have to say is 'tap tap tap... these kangaroos are crazy-and awesome'.

I am also playing with this software called 'GarageBand' - a software for creating musical compositions, only because I have so much time (yea, if you don't sleep you gain 7 hours for satisfying all your hobbies). Well, here is the result of using GarageBand for recordings. (well, these are some of my great flute recitals... okay, this is the only way I sucker someone in to listen to what I play)

Notes of Des

Fusion Music

Aarti Dhun

These are aif files so you probably need a QuickTime plugin to listen to these.

Okay, so we have talked about Cricket and GarageBand and here is the third topic ...

etc.

Halloween Meeting and Bowman/Crossbow Demo

2005-10-31T22:20:00.000-05:00

We have our research group meetings on Mondays. This monday being the
Halloween day(ref. previous post) some of the lab members were dressed
up in their costumes. I was not brave enough to dress up weird, but I
managed to hang on to a Longbow (which is the name of the planner we
work on) and I was Bowman (which is the user interface to the planner
web-service Fletcher). Dr. Young came in a Steelers Hockey(Ice) outfit (he is from Pitt), Joe was dressed up as an XBox, Yuna was a raccoon, Leo had a bear hat on and Jim came with a crazy devil mask. Jim's kids, Lincoln (Ninja) and Ben (Pirate) looked very cute in their outfits.

We were also supposed to demo Bowman and Crossbow (Partial Order
Planner implemented in C#). I was incharge of implementing the modal
literal support into the planner and the user interface for specifying
the inputs to the planner. The demo did not go well :(. We could not
get Bowman to talk to Crossbow with modal ops. The UI was mostly
implmented and working. Oliver showed a working single player
modification for Half-Life2 (source engine) that could execute the
action classes sent by the planner through an execution manager.

More about these systems: Zocalo Project

css revamp

2005-10-29T01:51:00.000-04:00

Just started modifying the css template for this blog. Figured I'd get a different theme from the usual ones. Also, this one agrees more with my new:

Website.

Tale of Two Celebrations (A Festival of Lights and A Festival of Darkness)

2005-10-28T19:00:00.000-04:00

Diwali and Halloween. Two festivals, same week, contrasting origins.

Diwali is a Hindu festival of lights. People decorate the exteriors of houses and apartments with little earthen oil lamps (and nowadays with electric lights). Sweets and festivities abound amid the din of firecrackers.

Halloween on the other hand is the Day of the Dead, when dead spirits come looking for bodies to possess. In order to protect themselves from these spirits, people dress in different types of ghoulish costumes. Children go about the neighborhoods dressed as little evil things asking for tricks or treats (candy!).

On Google:

More about Diwali.

More about Halloween.

TVS ( Transit Visualization System)

2005-10-28T18:46:00.000-04:00

NC State's Transportation division released a very cool real-time GPS based Online Visualization System.

This is developed by a group of recent graduates from NCState who founded their company with assistance from the university's Technology Incubator. The technology incubator provides start-ups with funding in the form of various resources (like office space on centennial campus).

The NCState campus is a collection of campuses spread across a very big area, students have to frequently travel back and forth between the historical campus (main campus) to the other campuses (centennial or vet school) for classes. I used to head to the bus stop atleast 5-10 minutes before the' scheduled arrival and buses usually (but not all the time) run 5-10 minutes late. Now with TVS you can just glance at the current location of the bus and make a run for it when it gets close by. Even if the bus is not late you can save atleast 5 minutes per trip.

This is a great application of the almost affordable GPS hardware (GPS software and maps etc. are still fairly expensive).

Doom - Review (no spoilers)

2005-10-28T01:50:00.000-04:00

Doom is a movie based on a series of games (Doom 1, 2, 3) developed by id software.

Story:

In the Doom universe, there is a genetic engineering lab (belonging to UAC corporation) on Mars accessible through a portal from earth. For some unknown reasons all hell breaks loose on the Mars facility and a team of soldiers from earth is called to investigate the situation on Mars. I will not spoil the rest of the story for you. For the gamers out there who havent already seen the movie --- Yeah there is a BOSS fight at the end :)

Sets:
The sets bore a faithful resemblance to the maps on the Doom 3 game environments. Mainly dark corridors and futuristic research labs.

Weapons:
The standard weapons - pistol, shotgun and chaingun were all there. One major difference between the movie and the game was that in the movie characters had a flashlight embedded into their weapons. Now anyone who had played Doom 3 would raise an eyebrow here since a key element of gameplay in these dark environments was switching between your weapon and your flashlight (yea you were not allowed to hold both at the same time).

Monsters:
Monsters looked good. They did not throw out fireballs though and I have mixed feelings about that (in the game some monsters fling fireballs at you). Some fights were particularly good, especially the one in the containment cell and one with the first person view.

SFX:
Special effects were routine and not very special. The game Doom 3 raised the bar higher in terms of dynamic lighting and graphics when it came out but the movie ended up just like any other sci-fi movie in terms of sfx. Sounds were good but not nearly so terrifying as the original game itself.

First Person View and other Game idioms:

What intrigued me about the movie before watching it was how game idioms would be translated into passive movie experiences. Now I could relate to different situations in the game as I was watching the movie. Here are the ones that I found in the movie:

1. Going around the labs to find access keys (or fingerprints) to gain access to protected labs/storage rooms for supplies.
2. Finding new (cooler) weapons as you progress through levels.
3. First person view - I really liked it on the big screen, non-gamers get an idea of what it is like to actually control a character against a huge monster :) What I feared was whether they had overdone it, but fortunately they didn't and the result was good.
4. BOSS fights. Every level usually contains a seemingly difficult boss fight after which you get refreshed for the next level.

Any others....?

Jorge Cham Talk

2005-10-27T23:29:00.000-04:00

Today I went to attend Jorge Cham's Talk. Although I had to go 'all the way' to main campus for it - it was totally worth it.

For those who came in late... (or didn't come at all) Jorge is on the faculty of the mechanical engineering department at CalTech, but more imporantly :) he is the creator of the popular comic strip PhD Comics.

The talk was very cool, he started off with the outline for the talk:

Outline

* Present the outline
* talk
* take questions

He did what every PhD student becomes good at - bulletizing complex ideas into powerpoint slides. I doubt there is any grad student who does not relate to the things that go on in the comic strip (funny as they may sound). He presented his theory of procrastination and the laws of graduation as follows.

* A graduate student remains in the state of procrastination until an external force acts upon it.
* Age(in years, of getting the PhD) = Flexibility of the advisor / Student's Motivation
Proof: Age = a, Flexibility = F, Motivation = m
So, a = m/F
Therefore, F = ma (we know this is true from Newton's law :)

Like any good research talk he did support his theory by giving examples of why his theory of procrastination works. 1. What was Newton doing under the apple tree anyways? (procrastinating) 2. What was Einstein doing working on physics at the patent office? (again procrastinating). There is a misconception that procrastination = laziness. Infact, it isn't. Why? You got to read the comic strip to know more...

Oh, and there was the complicated mind of the species 'homo procastinus'. Complicated as it may seem, there actually are only two things going on in that brain of theirs 'Free Food' and 'Keep Advisor Happy'.

I can go on and on about the talk. Every slide was interesting (and cartoony!).

To top it off, he signed a copy of his book (a compilation of all PhD Comics).

memeorandum

2005-10-21T13:28:00.000-04:00

href="http://tech.memeorandum.com This looks like google news for blog posts.

More about it follows...

I tag, You tag, We all tag - The 'Tagging' thing.

2005-10-21T13:19:00.000-04:00

After online web-logs (Blogger, msn blogs, et. al.) came RSS feeds and a barrage of readers that could update news and blogs. Around the same time whole bevy of social and professional networking websites (friendster, orkut, Gazzag, LinkedIn, Hi5 et. al.) grew by leaps and bounds in terms of membership. We have slowly (over the past year) evolved into photoblogging, podcasting (thanks to apple for spurring the sale of personal mp3 players) and videoblogging.

Now, there is data and then there is information and then there is the excess of it. So how does the human online community respond to this excess of information. By organizing it. How do you organize it, well, tag it. Everyone gives the content some 'tags' that they believe describe the content in some way. Honestly, when I heard about this tag thing I did not think it was going to be big. I am still not sure if it is going to be big, what with the credibility of the people tagging etc. but if wikis work, so could this tagging thing.

I was checking out this new firefox based browser href="http://www.flock.com" which required me to get membership to this website called the href="http://del.icio.us" This delicious thing looks interesting. You can create online bookmarks on it, tag them based on what you think best describes them and share it with everyone. The Flock browser now supports blogging, posting pics on flickr and also has a neat little scrapbook utility where you can drag stuff for blogging later. I am not sure how good the browser is yet in terms of actual working but will post a review once I use it for a little while.

Those who 'can', Do - Those who cannot , Teach

2005-10-18T21:43:00.000-04:00

I have always felt that the above average student in India is treated unfairly at all levels. At the high school level, there is this tension between knowing more v/s cramming more. Most exceptional students manage to gain knowledge and with the wisdom of remaining in the top of the class can be right at the top. Now, an above average kid, in her effort to get to the top on many occasion fails to grasp the real reason for going to school and that is 'gaining knowledge'. But even as this student enters the halls of universities, the same mentality prevails, nay, is aggravated by teachers who have limited knowledge and no motivation for increasing it in their own subject. Barring the IITs, IIMs and NITs there are few places where you can find teachers who teach, motivate and more importantly 'guide' the student to do research. Even some exceptional students end up not pushing thier limits during the critical years of undergraduate studies because of this lack of motivation on part of the teachers.

What we need right now is people who "can" and who "teach" to eradicate the widespread prevalence of "those who can, do - those who cannot, teach" thinking of the society.

I believe Dr. Narlikar's article (previous post) should insipire those above average students who have risen in their post-graduation lives to come back and change this mentality.

Reaching For The Stars

2005-10-18T21:42:00.000-04:00

Article by: Dr. JAYANT V NARLIKAR
(The writer is an astrophysicist formerly the director of IUCAA, Pune)

Higher education should fuse teaching, research to improve

A few months ago the finance minister announced a grant of Rs 100 crore
to the Indian Institute of Science, Bangalore, to help it raise its
standards to the levels of Oxford and Cambridge. Will a fistful of
crores help set up Oxbridge-level institutions of higher learning in our
country? How about offering a large sum to the Board of Control for
Cricket in India to generate a cricket team comparable to Bradman's 1948
Australian team? Or to Sahitya Akademi to produce another Indian Nobel
laureate in literature?

To get to where Cambridge and Oxford are, we could learn a thing or two
from them. Consider the following anecdote. In the middle of the 19th
century, two clever students, Parkinson and Thomson, were competing for
the Cambridge Tripos examination. When the results were out, Parkinson
stood first and Thomson second. Both stood way above the rest of the
pack. They were the only two to have solved a very difficult question.
The examiner of the paper, however, was intrigued to find that both had
solved the question in the same way. Did one copy the other?
He called Parkinson to find out how he arrived at the solution.
Parkinson explained that he made it a practice to go beyond the syllabi
and texts to read research journals and he had encountered the question
in a research paper whose author was anonymous. The examiner, who had
taken the question from the very same source, complimented Parkinson on
his preparation and interviewed Thomson. Thomson said: "I wrote that
paper". Thomson later went on to be a famous physicist and is better
known as Lord Kelvin.

In my batch there was an undergraduate who wrote a paper in the Physical
Review Letters, pointing out a serious error in an experiment performed
in the UK to verify a prediction of Einstein's gravitation theory. The
student went on to get a Nobel Prize based on his research work as a
graduate student. His name is Brian Josephson.
The likelihood of such brilliant students cropping up frequently
increases when the overall level of the students is high. Why does
Cambridge attract superior talent?

Good students are attracted by good teachers and researchers. I was
fortunate to attend Dirac's lectures on quantum mecha-nics, Hoyle's
lectures on electrodynamics and cosmology, Batchelor's course on fluid
dynamics,Davenport's course on number theory and Attiyah's lessons on
geometry in the course of my mathematical tripos at Cambridge. To learn
the subject from those who made important research contributions to it
can be very inspiring to a student.
These faculty members delivered their lectures as part of their
lecturing commitments at the university. They were not
exempted from lecturing because they were producing good research: It is
axiomatic that a faculty member in an educational institution will teach.

When Hoyle's proposal to establish a new Institute of Theoretical
Astronomy was under consideration at Cambridge, a condition was sought
to be imposed by a lobby of the faculty — that its staff members should
not teach the tripos students. Hoyle was unhappy with this. He explained
that teaching was complementary to research and that if we did not teach
undergraduates, we would not attract good students to research.

This explains why we are still a long way off from Oxbridge. We have two
streams in our educational-cum-research system — universities and
research institutes. In universities, teachers are overloaded with
teaching duties with research taking a back seat. The situation is
reflected in the membership list of our national science academies:
There are very few members from universities because the criterion for
membership is research.

In national laboratories and research institutes, the research
facilities are fine, but there are no undergraduates, not even students
for a masters' degree. A typical research scientist believes he will do
research alone and that teaching students is a distraction.
Organisers of graduate schools in these institutions will tell you how
difficult it is to persuade a faculty member to give even a short course
of 15 lectures.

On my return to India, I ran up against this attitude which I could not
grasp after my Cambridge experience. The University Grants Commission
should have made it a condition that these institutes have some minimal
undergraduate or masters' teaching programme, before giving them deemed
university status.

Our undergraduates fall between two stools, between professors who do
not teach and professors who do no research.
We lose motivated students who could be attracted to academia. My
experience as a scientist extends to other disciplines. We grope for
excellence in history, literature and other social sciences.
What is needed for our progress towards Oxbridge is a new set of rules
that enable teaching and research to go hand in hand, with no compromise
on merit, whether for faculty or for students. To achieve this, Rs 100
crore is neither necessary nor sufficient.

Doom - The Movie

2005-10-18T21:40:00.000-04:00

Yesterday I watched some preview videos of the upcoming Doom movie. Now, for those who came in late, I am a big FPS fan. The only games I have played through are FPS, Doom (the original) was one of the first 3D FPS I ever played followed by wolf3D, Quake II, Quake III, Unreal Tournament (the whole series including 2003 and 2004), return to castle wolfenstein and Half-life I and II to name a few.

In the last three years or so I have done some modding for UT and Half-life as well. And, I am interested in bringing in cinematic idioms from movies into games. So, it is obvious that I am really excited about a movie that brings in video game idioms (fps view) into a movie. So I am really looking forward to it and more so the reaction of the audience to it. My previous experiences with movies based on games make me take a defensive stance against too much excitement for this movie. Anyway, for the sake of 'research' I will certainly go to the movie this weekend and a review shall be posted here.

On Cricket SuperSeries (WorldXI vs Aus)

2005-10-11T17:31:00.000-04:00

I really have mixed feelings about the 3-0 series whitewash of WorldXI against Australia. On the one hand there are really great players in the world xi team and each one of them had the ability to give a matchwinning performance. I read a lot of comments about how world xi did not play as a 'team' and that 'if only one person had clicked' things would have been different.

I believe that australia's strength is that none of the australians really are uncomfortable against any of the players in the World XI. They have players who can play shoaib or murali or both at the same time. I don't believe this is true for the 'individuals' in the World XI. Early overs at McGrath's accuracy have proven fatal a number of times for the likes of Sehwag. What was really amazing was that Sangakkara of all people came out as the best batsman (he was even not selected in the test team)? Although Kallis and Lara have had a great year, they are on the lower end of their form graphs right now for the year. Dravid has not had too much quality cricket in a while and really the only other player who the team could have counted on was Flintoff (he must be getting used to the feeling now).

If I were to breed a team solely for the purpose of defeating the australians, this is not the team I would select. There were others who did well 'against australia' who did not make it to the World XI purely because of either their inexperience or their lack of form early this year. Also, teams must be selected not just by selecting the best players but also by looking at their styles and effectiveness working together.

Anyway, it is certainly an interesting series and I would like it to be repeated once in a while (although not against australia every time but the current number 1 team-doesn't look like it will change in the coming 4 years... time will tell).

Mac Share/Free-ware Applications

2005-10-11T17:09:00.000-04:00

I have been following a number of posts on the top 10 mac software and I have to say that some software that I found were very useful and free too!

Here are some software/utilities that I use regularly (in no particular order)

TeXShop - For LaTex editing all my papers etc.
VoodooPad - Great notes software
Desktop Manager - Multiple desktops
NetNewsWire - For all my rss feeds from blogs, news, forums, etc.
ChickenOfTheVNC - A VNC Client for Mac
FFView - For viewing comics (btw check out http://thecomicproject.blogspot.com if you ever came across indrajal comics)
Fugu - ssh/scp gui client
SideNote
Adium - A one stop chat client for all your chatting needs (AIM, MSN, YAHOO, etc.)
MenuCalendarClock
GmailHD (applescript - very cool)
iTerm - Terminal emulation program with tabs and other cool features
VLC - media player
ComicLife - Make comics out of your pics (I'll post a couple one of these days)
eclipse - Development environment written in Java for Java (and then some)
Mono - C# on Mac
PageSucker - For downloading webpages for offline view

Links to some blogs/pages that list out top mac software.

Om Malik's Thread
HicksDesign
MacZealots Article

Navratri

2005-10-11T15:48:00.000-04:00

Nav-ratri (9 nights) is a festival to worship the goddess Durga (or for some to worship three goddesses Durga, Lakshmi and Saraswati). The festivities that last for 10 days end with the burning of an effigy (usually large) of Ravana (the evil king) on a day called Dushhera(10th day) to commemmorate the victory of Rama over Ravana which he attained after asking for blessings from Goddess Durga.

Navratri [1] is a popular festival in my home state of Gujarat [4] in India. During these 9 nights, people dress up in typical gujarati dresses to perform a very popular dance forms called the 'Garba' [2] and the ' Raasa' [3].

I attended the Navratri celebrations at a local Hindu Temple in Morrisville, NC (yea, I was all dressed up too with the traditional koti and dupatto). The orchestra was good and they played older garba songs rather than the noisier new ones. There were about 80-100 atendees (less as it was a monday night). A brass statue of goddess Amba was kept in the centre of the hall around which we went in circles performing different styles of garba (1 clap, 2 clap, 3 clap) for about an hour (9-10) at a slower tempo. After a short break at 10 we moved on to perform Raasa, which is another dance form that is played with two wooden sticks and is usually performed in pairs. Raasa could be a more aggressive dance form and we enjoyed doing it at a higher tempo. The event was wrapped up after the aarti (prayer) around 11 PM.

One of the great things about festivals is that it brings a lot of people together. That said, I believe that we are losing the main reasons for why the festivals are celebrated and are merely looking at them as 'fun' events. Most festivals have some story behind them and a lot of these stories were written to teach people how to behave and live in the society and improve as individuals.

References
------------
[1] Navratri Links
AhmedabadCity.com
Blackburn Hindu Centre
[2] Garba Links
GujaratOnline
Garba
[3] Raas
Dandiya Zone
About.com
[4] Gujarat
Official Gujarat State Govt. Site
Gujarat State Tourism Site

ARGH! Blog Spam

2005-08-15T12:26:00.000-04:00

Man! these marketing guys really get on my nerves.

After phone spamming, email spamming, IM spamming... welcome to the world of BLOG SPAMMING...

My post about attending a national conference on AI got this comment from somebody named vyht2dpga04rmlg, what the $@%#&@%#

http://cyclonurb.blogspot.com/2005/07/aaai-2005-conference.html#comments

hrmph!!!

AAAI 2005 Conference

2005-07-15T00:09:00.000-04:00

I was lucky to get my paper accepted for presentation at the National Association for Artificial Intelligence (AAAI) 2005 Conference held in Pittsburgh, PA from 9th July through 13th July.

The highlights of the conference were a series of invited talks, doctoral consortium, competitions and robot demos, poster sessions and a chance to interact with top researchers in the AI community.

Students who attended the conference are running a blog for publishing their views on the conference. Please visit AAAI Blog for more information.

TA Evaluation

2005-04-05T17:39:00.000-04:00

I just found out the student evaluation scores from the class that I was a Teaching Assistant for (scores in parenthesis are departmental averages for TAs):

TA attends office hours: 4.44 (4.16)
TA is well-informed: 4.75 (4.21)
TA understands subject: 4.88 (4.24)
TA understands platforms: 4.69 (4.21)

TA overall: 4.56 (3.92)

It will be very difficult(almost impossible) to maintain these for the current semester but the target would be to remain well above the departmental averages atleast.

Video games for education and training: Part I

2005-02-13T14:57:00.000-05:00

I hate cleaning up my apartment and taking out trash. Most people do. Especially when they live in a bachelor's apartment. Why anyone would enjoy doing this, even in another reality, beats me. Nevertheless, THE SIMS[1] is one of the best-selling games ever. Not saying that all you do in the game is clean up the apartment; no; but all that you do is really what you should be doing in real life and enjoying as much as doing it in front of your computer screens. So, does that make people playing the sims for a long time(it is addictive to some) better people? There is no support for this claim. But that is because The Sims was developed as a game, not a training program. If it were marketed as a training program, I probably would not be writing this article.

The motivation behind writing this article came from the daily blog post dated 12/23 in the biotech section of the MIT Technical review magazine[3].

The purpose of this series of articles is twofold. First, I would like to analyse games, successful as well as unsuccessful ones, to see why they appeal so much to people of formerly only teenage and now pretty much all ages. My second purpose in writing this article is to identify the various competencies that could be developed through games during the years of formal education all the way from kindergarten to professional graduate level curricula.

Through online games, I believe, people living in most modern societies can learn the codes of social behavior better than in real life. I once participated in an experiment carried out in a popular Massively Multiplayer Online Role Playing Game(MMORPG) where two teams - one comprising avid gamers, and the other comprising soon-to-be army officers were given a search and rescue assignment in a city populated by strangers(yeap, real strangers in the virtual world) and the result was that gamer team was much more organized in its maneuver, and were successful in their mission, while the latter failed to maintain proper communication which led to their failure.

References
----------

[1] The SIMS, Maxis

[2] EVERQUEST - Sony Entertainment

[3] Video games used for training surgeons

[4] Everquest meets Sopranos

Music - The Time and Mood Connection

2005-02-13T12:51:00.000-05:00

Over the last month or so, I have been experimenting with my choice of music in different situations and for different moods. Here are the observations...

In the morning when I wake up: silence
During my Bath: 96 Rock on radio (new rock)
When I am alone driving: Indian Classical Flute(especially on >30 min drives, since I can listen to complete ragas) - Yes, I believe that your driving is affected by the kind of music you listen to which implies that I am strictly against aggressive rock music on the road.
With friends driving: ... varies from Hindi Film Music to soft rock ...
While working:
Programming: Metallica, Creed like stuff
Writing : Western/Indian Classical(santoor mainly)
Cooking etc. at home: Ghazals (Jagjit singh and Ghulam Ali)
Weekends or evenings: Talat Mahmood or Mukesh

I guess the choice of music according to mood is really very intuitive:

Peace/focus: Indian/Western Classical
Rest: old hindi film or ghazals
happiness: soft rock, some film songs, some ragas
frustration: hard rock

I like musical ambience, so more often than not you'll find something or the other playing when I am around.

Neighbors Clash

2005-02-13T12:01:00.000-05:00

I read about the India-Pakistan series while browsing the Cricinfo site yesterday. Usually the series comes with a lot of controversies surrounding various issues that are totally unrelated to the game itself. For instance, there is always a big threat from certain religio-political organizations about digging up pitches(recently the Mohali incident). Then there is also the big security issue for the players at certain cities that are prone to have communal riots. It is very unfortunate that such things happen. Many would agree with me if I said from my observations so far that all this off-the-field confusion significantly dies down once the series starts. Most of it could just be marketing gimmicks - I am not sure.

I was lucky to get a chance to watch the series in Pakistan last year(I will not be able to this time :(). This series, I believe, is going to be interesting because in contrast to last year when pakistan were at a disadvantage of having a relatively new team with lots of inexperiece coupled by the fact that the experienced players were not very successful during the series, this time they have gained some experience especially after the australia tour. On the other hand, I am not satisfied with the Indian camp's performances in the past few tours. Bangladesh was easy but you never saw every indian batsman consistently performing well even against them. The bowling was never upto the mark. Even though India has the home ground advantage I would expect a very tense series(I know that bookies as well as sponsors would want that too...but in a different vein than I do).

Talking about neighboring clashes, two others are also coming up in South Africa vs Zimbabwe and Australia VS New Zealand. I don't have much to say about the former because of the dismal form Zimbabwe is in(for various reasons). It would be fun if they were to provide respectable resistance to SAF. The latter is going to be an interesting series because I have always had respect for the NZ team. The great thing about their team is that they have never had big stars in their team (major exceptions MCrowe and Hadlee) but they have had very good team performers. It will be interesting to see if they can pull of a shocker against Australia.

I think I have written too much for now. Please excuse my disorganized ramblings in some of my posts. There is a lot to say on this topic and comment-discussions are most welome.

Intelligent 3D OS Interfaces

2005-02-13T01:39:00.000-05:00

I took a Human-Computer Interaction course and thought it would be cool to think about how some of the anti-mac principles are broken by 3D game interfaces.

Here is a paper that I wrote about it. The paper may not look very coherent, but I don't have time to refine it now... maybe I'll pick up research on that in future. Right now it doesn't feed my VAL.

Here's the link: Issues in designing Intelligent 3D OS Interface.

old article revived: Morphological analysis of design space of input devices

2005-02-13T01:27:00.000-05:00

*****************************************************************************

Morphological Analysis of Design Space of Input Devices

A REVIEW

(I do not have the names of the authors of the original paper(s) that this review is based on, I will post it as soon as I recover a copy of the paper)

*****************************************************************************
Arnav H Jhala

Date : 01/27/03

*****************************************************************************

1. Systematic Study of Human-Computer Interfaces

* Toolkits
* Taxonomies
* Performance Analysis
* Morphological analysis of design spaces

2. Design Space

* Construction
* Analysis
* Examples

3. Conclusion

******************************************************************************

A large number of Human-Computer Interface devices are available today and they seem to address different aspects of HCI, some in a general sense and some specific to the application for which they are designed. This techonology has developed to such an extent that it is now possible to systematically study and compare these devices to come up with certain design standards that could then lead to development of devices from these designs. A discussion of such a method is the central theme of the paper.

There have been various types of methods employed for the study of such techonolgy, namely: Toolkits, Taxonomies and Performance Analysis.

Toolkits:

Toolkits basically provide the user with a wide variety of design tools to systematize the device knowledge. They do little to help users to come up with design decisions and need additional information like technical abstractions about the user, devices and tasks.

Taxonomies:

Taxonomies basically attempt to provide a classification of input devices based on
different aspects like computer graphics sub-tasks or physical properties or dimensions that they can sense. Although these approaches to take us closer to describing the design space, they are currently for very restricted domains.

Performance Studies:

Performance studies are based on actual experiments with interfaces. They provide user abstractions for various devices which is not provided by any of the above methods. The downside to these is that the studies have not necessarily attempted to disentangle task, subject and human performance variables.

Morphological Analysis of Design Spaces:

Another method of analysing the Design Space for input devices is to plot the
performance of input devices and their description into a parametric space. This
can be typically used to find abstractions for generating and testing the designs.

Construction of Design Space

Human-machine communication that basically occurs through the input devices can
be basically thought of as a dialogue between two systems where the language of
the dialogue is hand movements and clicks on one end and charts, etc. on the other
end. Focusing on only the issue of input devices, it can be formulated as a problem of three interacting agents, namely the human, the machine and the application.

A language of communication can be modeled using these primitive operators:

1. A primitive movement vocabulary
2. a set of composition operators

Formally the language can be represented as a six tuple

-M Manipulation Operator
-In Input domain
-Out Output domain
-R resolution function from input to output
-W set of device properties that describes properties of devices

The physical properties like the Position, Displacement, Rotation, Turning,
Force, Delta Force, Torque and change in torque can be used as primitive
Manipulation operators. These are operators that the user uses to express something to the system.

There are three types of compsition, the merge composition where merging of two devices leads to access to physical output domains for both the devices. the layout composition is the collocation of devices on different places of a common space. The connect composition where the ouput of one device is the input to other device.

Thus the design space is described as the set of possible combinations of these operators.

Analysis and Testing of Design Space

This is basically done using the techniques described in the Performance analysis approach earlier. The designs that are created using the above mentioned design space can be evaluated using two different measures:

1. Expressiveness, is a problem arising out of displays not being able to prevent expression of unintended meaning.
2. Effectiveness, that deals with, to what degree is the intended meaning conveyed to the user.

Effectiveness is a feature that can be tested quantitatively through various metrics like,
-Desk footprint
-Pointing speed
-Precision
-Errors
-Time to Learn
-Time to grasp the device
-User Preference
-Cost

The paper discusses two of these in detail, the footprint is basically tested as the amount of space the device occupies on the desk and since this is directly related to what kind of a device is actually used, it still tells us that a device having larger footprint is bad. Also, the performance depends on the sequence of actions of the device and not on the device itself. This can be resolved by choosing an extreme task.

Another aspect discussed is the Bandwidth or the speed and extent of use of a device. Every device relies on some transducer on a human and this factor relies on the effective bandwidth that such an associating muscle has, it also depends on the application's precision requirements and the devices' effective bandwidth.

There have been studies on both the human muscle groups and the precision requirements of tasks and there have been established the Fitts index of Difficulty that gives the movement time for muscles of the body and the wellford's formula of this index as a function of distance to width. The device bandwidth is standardised with respect to the movement time of the "Hardest Easy Task" for a mouse since it is kind of a ubiquitous computing device. This hardest easy task for a mouse is movement time to click a word and comes out to be around 220ns using the fitts law. Thus precision of a device is the equivalent of the hardest easy task for a mouse.

Conclusion

This paper has essentially presented an approach to systematize the knowledge about input devices to come up with a representation that enables us to evaluate the performance of various input devices in formal terminology and also at the same time come up with designs based on the analysis of unexplored areas of the Design Space
where potential improvements are theoretically observed.

The approach combines the different previous approaches and their advantages and brings them under one hood while generaizing the concepts derived from specific studies in the field. But it is not wrong to reiterate that the most interesting part of this research is the ability to analyse the design space to look for potentially better performing devices and to practically see them in action !

On Attention

2005-02-13T01:25:00.000-05:00

Attention

The Meaning of Attention

Attention can be loosely defined as "selection of a particular stimulus from a group of stimuli for further processing". For example, A person attending to a conversation with another person in a crowded environment. In this case out of multiple auditory inputs, the atendee chooses to attend to the voice of the person who he/she is conversing with.

Attention as Resource Allocation

There are different kinds of stimuli that humans can perceive, namely Visual, Auditory, Tactile etc. It is easy to observe that attention is sometimes exclusive to one stimulus, shared between multiple stimuli or switched among different stimuli. This process can be explained by viewing it as Resource Allocation. The varying degrees of resource sharing bring about the types of attention allocation like:

Exclusive Attention, Shared Attention and Switched Attention. It is observed that better processing of one stimulus leads to poorer processing of other stimuli, this is known as performance trade-off. Detailed discussion on how this trade-off affects attention is presented in the following discussion.

The Study of Attention

The study of detection and identification of visual targets is among the most active areas of attentional research. The results from the visual search experiments are used in this document for demonstrating various concepts involved in the study of attention.

Definitions

Visual attention involves attending and orienting to visual targets.

There are Overt forms of attention involving shifting of gaze to bring the target into view(in most cases the focus or center of view).

In Covert forms of attention, there is no such shifting of gaze but more attention is paid to the target.

In some cases simultaneously presented stimuli are attended to without any processing trade-offs, in such cases a phenomena called "Pre-Attentive Processing" comes into play. This can be explained by noting the fact that some target processing requires much less than one's attentional capacity and in some cases only the primitive processing is required for identification.

Visual Search

The experiments that require subjects to identify a particular target element from an array of visual elements are refered to as the Visual Search Experiments. The results give an insight on the process of visual attention and factors involved therein.

One important consideration in such experiments is the "Set-Size effect". For some targets it is observed that the response time of identification increases linearly with number of elements. Yet other targets are identified in constant time independent of the number of elements.

Other metrics used in the analysis of experimental data are:

AOC/POC: The Attention Operating Characteristics and Performance operating Characteristics are plots of processing time of various targets with different visual features. These determine how much attention allocation is done on the targets.

Cost-Benefit Analysis: This is a function that plots the average time to detection of targets considering valid, invalid and neutral cues for detection. It does not show explicitly whether or not the subjects are given a chance to attend to each element equally or not.

While considering experiments with visual cues, speed-accuracy tradeoff is also measured. This metric gives an idea of how accuracy of detection is affected with changes in speed of identification.

Conjunction of Features and Discriminability

It is observed that set-size effects are highly pronounced in cased where multi-dimensional feature identification is carried out. This is explained by Treisman's Feature Integration Theory(FIT). According to FIT, conjoining of features is a process that "requires attention". For example, Searching an n-element array for a conjunction target leads to a linear set-size effect.

POP-OUT EFFECT

Certain multidimentional targets seem to have "higher-level" or "emergent" properties that also produce flat search functions. Such targets seem to "pop-out" of the display

These results have led to the discussion about Discriminability of features. Certain features are more easily discriminated than others. Thus more discriminable features are used in User Interfaces for better detection response functions. This has led to alternatives of FIT that take into consideration the heterogeneity of the targets and distractors.

Order of Search Functions

A brief note has to be made regarding the linearity of search functions. In certain cases if the subjects are exposed to detection practice prior to actual detection then such practice tends to flatten out search functions. Significant research has been carried out by Shiffrin and Schneider using "constant mapping" and "varied mapping" between targets and distractors. Also, practice leads to pre-attentive processing where perception seems almost involuntary. For example, identifying the string "DOG" for an experienced reader.

On Perception

2005-02-13T01:22:00.000-05:00

What is Perception?

The Dictionary of Science and Technology defines perception as "the conscious mental awareness and interpretation of a sensory stimulus". For centuries there have been studies in the field of psychology and physiology on perception. More recently there has been a lot of research in AI to study perception through computational models. In this document an attempt is made towards a study of the understanding of perception as perceived by researchers in the aforementioned areas.

How do we perceive? (The AI Approach)

The AI approach to study of perception breaks the problem into three levels of analysis. The questions that correspond to these three levels are:

What are the psychological mechanisms involved in perception?

What is the process of perceiving?

What are the properties of the physical world that allow us to perceive?

These questions are clearly tackled in the field of AI as study of:

Hardware, algorithms of operation and theory of task to be performed.

OBSERVER THEORY

Observer theory is a promising development that has emerged from the AI approach. According to the observer theory, and 'observer' performs inferences from a relation between the class of perceptual premises and its characteristic set of perceptual conclusions or interpretations. This theory promies to resolve many issues in perception due to the fact that observers can be chained in a variety of different networks. Observers can not only take primitive sensory perceptions but can also take higher-level inferences from other observers. One very significant claim made by observer theory is that most of the observers cannot be modeled by a Turing machine; thus challenging the assumption that human perception can be modeled completely by AI techniques.

The Neuropsychological Approach

The neuropsychological approach tends to study perception by observing the behavioral patterns and associating them with the modules within the brain. This is generally done by using results from studies of brain damage caused by various agents. It provides a window to the biological hardware underlying complete acts of perception and cognition. This is in contrast to the neurophysiological approach wherein a study of neural substrate for registration of simple neural stimulus is done. One very important contribution of this approach is its capability to provide strong behavioral models that provide independent and strong constraints on modeling. The only questionable issue in this type of research is the modularity assumption, as it does not provide for an overlap of biological hardware for different stimuli.

The Gestalt Theoretic Approach

The issues related to study of perception inherited from the Gestalt movement include perceptual organization, figure-ground segregation and perceptual combination of attributes.

Ecological perception is based on the Gestalt hypothesis that perception is determined by the stimulus configuration presented to the attendee; it also shares the Gestalt rejection of any process that mediates construction of perceptions from sensations. Another school of thought developed by Rock et al is based on the assumption of intelligent perceptual processes.

Comments on the unification of approaches

While each of the given approaches provides a different perspective to the study of perception, individually they fail to give a strong theoretical foundation to the theory of perception. It seems that for a complete theory of perception, various factors need to be considered like stimulus information, an abstract functional description of behavior without regard to the underlying biological machine and the study of low-level perception mechanisms on the biological hardware.

An AI approach using the above three layers is a sound approach to a complete theory of perception. An attempt has been made in the form of an area called "Natural Computation" developed by Richards et al that addresses this issue.

Hibernation, Work, Ski-Trip, Qualifer and all else...

2005-02-13T00:55:00.000-05:00

Last few weeks have been quite hectic, and so will be the next one.

I am preparing a couple of papers for the ACE 2005 conference. Writing axioms for cinematic rules, implementing planning algorithm in LISP, writing an OGL based Interface in C++ and C#, writing code on Unreal Engine for path planning are keeping me busy.

I still managed to squeeze through a 2 day ski-trip to the Wintergreen Ski-Resort in Virginia. A colleague of mine, Jim Thomas, owns a very nice lodge in the resort and he invited the lab members for a mid-week ski adventure. I did pretty well to come out with only a bleeding nose and aching neck and thighs, considering it was my first time, and that I was on my own (since the others were experts) on the beginner trail. By the end of the day, my legs were itching for more, but even though the mountain was lovely, white and sleek I had to get back to work the next morning. I will write the details in a subsequent post.

My qualifier committee was assigned to me last week. Including my Advisor Dr. Young, the department assigned Dr. Munindar Singh as the topic expert(AI - Agent Communication) and Dr. Alan Tharp as the departmental representative. I plan to take my qualifer ... soon.

Some articles that I wrote and saved as drafts but never saw the light of the monitors other than mine, are in the process of being revived so look out for them.

We are THE people

2005-01-01T02:35:00.000-05:00

ABC News announces Bloggers as 'People of the year'. Blogging has really picked up this year and it is true that it gives people a new way of expressing their thoughts, ideas and even fantasies.

I believe that blogging might even evolve to being a psychological therapy because it is a well-known fact that "It helps to talk.". Regular blogging to me also brings about an inspiration to do and know things I can write about. Also brings a sense of organization in your life. It gives you food for introspection when you go back and read your old posts.

I think it is turning out to be one of the finest contributions of technology at an individual as well as social level.

Here is the complete story.

... sometimes I like using exaggerated language in my posts as you can see from this one.

There should be negative aspects of bloggin too....

Like privacy and security, if you are careless about disclosing stuff available for everyone to see freely. It has a potential for becoming an addiction, which is also not good.

Any other comments?

anti-corporate activist pranksters

2004-12-31T19:13:00.000-05:00

The Yes Men are a group of pranksters who record videos of their issue based practical pranks related to real issues.

Science's top stories of the year

2004-12-31T18:59:00.000-05:00

Seems like astronomy and biotechnology were the two areas where there were major breakthroughs.

Here's the whole story from MSNBC

Homepage renovation

2004-12-13T01:27:00.000-05:00

I figured that my homepage needed to be brushed up. I have two big posts that I need to publish, but they still need a lot of work. There's a feedback form that sends me an email with your feedback, feel free to use it for suggestions so that I can continue improving the site. Here's the link:

HOMEPAGE

What's in a name?

2004-12-02T03:02:00.000-05:00

I have been noticing for quite some time the importance of naming your machines on a network.

For instance, our liquid narrative lab machines are all supposed to be named after video game characters(no wonder) and they all log into our server zion.csc.ncsu...

My two machines so far have been lara.csc.. and alice.csc..., other machines are johanna, kakarotto, gundam, niobe, samfisher(xbox) and the new entrant masterchief(Halo).

Anyway, I named my new ibook 'ekaagra' it just sounds great and has a great meaning too. It translates to something like 'focused' in english. I like my roommate(Nirmit's) machine name which is 'Samyak' from jain religion.

Gloomy weather, whether or not I am, or am I?

2004-12-02T02:34:00.000-05:00

This morning after I woke up, couple hours later than when I had planned for, I had already started having a bad feeling about today. The drive to work was terrible, gloomy weather with thick black clouds hanging around and fair amount of rain; the wind chill factor would have been nasty too. The only good thing about the long and slow drive was that instead of listening to NPR(national public radio: FM 91.5)-more gloomy news/discussions- I instead went for a soothing raga bageshree and then raga durga on flute.
Couldn't finish my computational logic project that I had planned to today. Will have to continue working on it. Very hard to concentrate while reading a logic paper, I forwarded one that looked remotely like our work, to my advisor, who did not take more than half an hour to respond to me by saying - "Arnav, this paper makes my head hurt, is yours going to be anything like this?" - well, not really the only head that my paper would hurt will be mine(while I am writing it).

Conclusion:
1. [THEREEXISTS] x. ComputerScientist(x) AND Logician(x)
2. [FORALL]y,x. ResearchPaper(y) AND Author(y, x) AND Logician(x) -> HurtsHead(y, arnav)
3. [FORALL]y,x. ResearchPaper(y) AND Author(y, x) AND ComputerScientist(x) -> Understands(x, arnav)

4. [FORALL]day, y. HurtsHead(y, arnav) AND Today(day) -> NOT Productive(arnav, day)
5. [FORALL]day, y. Gloomy(day) AND Today(day) -> NOT Productive(arnav, day)

6. [FORALL]day, y. 4. AND 5. -> SUPER NOT Productive(arnav, day)

--Today was a super non productive day.

Wish list for mimesis 2k4 release

2004-12-01T14:40:00.000-05:00

I have been given the charge of preparing the Mimesis Unreal Tournament 2004 client, alongwith our overlord-eric pelkey. I am aware of a few problems with the 2003 version that we released last year and some are really annoying. Here's a wish list so I can check all these off before release.

1. Server name as a default property.
-- primarily so that anyone working only with lisp/c# version of planner does not have to recompile the UT code every time the server changes or if they want to test it on a different server.

2. step-by-step guide to developing Mimesis worlds - for dummies ... and newbies

3. discuss the possibility for using one of the older lab machines as a dedicated server running all/most of the (lisp and c#)components as well as fletcher.

4. Allegro has to be restarted after every execution of a ut story. registry::(shutdown) does not close the port 2500 after shutting the registry down.

... more as they occur to me

To DOT NET or NOT .NET

2004-12-01T14:28:00.000-05:00

For the last few days I have been thinking about this interface that I am building for an intelligent storyboarding tool. The components that I have to use are: the mcapi and mwapi mimesis components, now since the dlls are created for .NET I have all kinds of stupid msvcpdblah.dll errors when I use not dot net. Now even though I can just recompile the thing again in not dot net and create a dll it still doesnt work, just doesnt, the function call fails and I have no access to it. So why not just do it in .net, well, obviously theres a good reason for it. the other big module that I am trying to fit in is the geometric constraint solver, and I dont have the courage(or rather the time) to get behind the few thousand line solver code and try to recompile it in .net. there's also some talk of .net having 'managed' code and 'unmanaged' code, but I have decided to (atleast for the next couple of weeks) stay away from all that and just finish it off in MFC-VC6. whew! can't believe I wrote all that in less than two minutes! the frustration shows!

2004-11-30T01:26:00.001-05:00

Closed ibook

2004-11-30T01:26:00.000-05:00

ibook pics as promised...

2004-11-29T22:02:00.000-05:00

Link to article and raaga list

2004-11-26T14:53:00.000-05:00

Article about Raagas and Time in carnatic music written by Sudha Jagannathan on Sify.

Also an alphabetical list of Raagas that I have on either tape, CD or MP3,

- Ahir/Nat Bhairav (Flute, Santoor, Guitar)
- Bageshree (Flute)
- Bhairavi (Sarangi)(Violin) (Flute, Santoor)
- Bageshwari (Flute, Santoor)
- Bilakshani Todi (Sitar)
- Bhoop (Santoor)
- Bhoopali (Flute, Santoor)
- Bhimpalasi (Flute, Santoor)
- Chandrakauns (Santoor)
- Darbari Kanra (Sarangi)
- Dhani (Sarangi)
- Darbari (Flute, Santoor)
- Des (Santoor, Flute)
- Desi Todi (Flute)
- Durga (Flute)
- Hamsadhwani (Flute+Santoor)
- Jait (Flute)
- Jog (Flute)
- Kamalshree (sarangi)
- Kausi Kanhra (Sitar)
- Kirwani (Flute, Santoor)
- Kalyan (Flute)
- Manjh Khamaj (Flute)
- Madhu Malati (Sarangi)
- Mand (Sarangi)
- Mian Ki Malhar (Sarangi)
- Malay Marutam (Flute)
- Mishra Bhairavi (Sarangi)
- Mishra Piloo (Sitar)
- Malhar (Sitar)
- Miyan ki Todi
- Mishra Kirwani (Santoor)
- Malkauns (Flute, Santoor)
- Marwa (Flute)
- Megh Malhar (Flute)
- Pilu (Flute)
- Panihari (Sarangi)
- Patdeep (Sitar)
- Pahadi (Santoor)
- Rageshwari (Flute)
- Rageshri (Flute)
- Sindhu Bhairavi (Sarangi)
- Vachaspati (Flute)
- Yaman (Flute, Santoor)
- Yaman Kalyan

'Twas chilling and my filthy toes did tremble and nimble in the air ...

2004-11-26T13:06:00.000-05:00

It was the great thanksgiving sale this morning.

woke up at 3:30 in the morning and left for bestbuy with Shikha, Pragyabhabhi, Jayush, Bhavanabhabhi and Rachana.

We waited in practically a freezer-only windier than the One in a long ...no loooooong queue outside of the Best Buy at Crossroads Mall. And, then on an instinct I decided to move to the store next to best buy, Office Max, coz I thought I'd seen better deals at office max online last night.

We ended up buying all crap-but free (two usb keychain hard drive, couple 100 CD-Rs and DVDs, one cordless phone set and something else I can't seem to remember). I had a coupon for PS2 and when I asked a best-buy employee standing at the games aisle about the price, he replied, "149, but you're guaranteed to get one if you present the coupon before noon", wow, and I thought I'd get a greate deal on a PS2 ... dreams are not like glasses - they break, yes - but they are more like clouds, which is why we see cloudy or foggy dreams(well, atleast in old Hindi films). Anyway, there's nothing solid about them, hence the simile.

so, now in the middle of the day I am feeling very sleepy, depressed as I have a lot of work to do. Wish I could copy my mind into a portable usb drive and send it off to space or something.

'kay sleep's taking over now ...

the words are fuzzier ...

fuuuuzzzzzziiiiiieeeerrrrr....

.......

Moving ...

2004-11-22T19:59:00.000-05:00

Today we started our couple of week long moving exercise. Since we have net connection at the new apartment, I think I'll just come over here at night.
The new place is in Morrisville, NC (The Legends at Preston)and we have an awesome view of the Preston Golf course from the first floor apartment's french window. Will post some pics soon. It is going to be a 20 min drive everyday to campus from here. Good thing is that Hindu Bhavan(our local mandir) is only abt 5 mins from here, and so is the airport. And in december there is not going to be much work(ah really!) at the office, it will be easier just working from home (except for the fact that I will not be able to do any Visual Studio stuff on my ibook (well maybe not, I have Virtual PC). Anyway, have to rush back to avery close now for dinner, I am very very hungry right now.

pics shall soon follow.

Pic from AAAI Fall Symposium, Workshop on Style and Meaning in Language,Art and Design

2004-11-21T11:25:00.000-05:00

Front

From far left:
1.Shlomo Dubnov - UCSD - Music
2.Dalia Cohen - Israel - Music
4.Harold Cohen - UCSD - AARON's creator (Invited Speaker)
5.Roger Danenburg - CMU - Music (Tutorial)
6. DUH!
7. Michael Gamon - Microsoft Research Natural Language Group

Back

From Far Left:
1. Bill Swartout - ICT(USC) (.. My academic grand grand parent) - Guest
2. Arshia Cont - IRCAM France - MUSIC

Flickr

2004-11-21T11:21:00.000-05:00

This is a test post from , a fancy photo sharing thing.

About Raga Des

2004-11-19T13:40:00.000-05:00

It is a beautiful night melody of the Khamaj thaat. It complements the romance and poignance of the semi-classical thumri. Omitting the third note or ga and the sixth note or dha in the ascending aaroha, des has a sampurna avroha (descending scale) that consists of all the seven notes. Hovewer in the avaroha, a flattened ni is used as opposed to a shuddha ni in the ascent. Pa and Ri rule over the raga as its vadi and samvadi swaras.

Acknowledgements for this description.

-Living Media India Ltd.(Music India - Krishnadhwani series)

About Raga Jog

2004-11-19T13:37:00.000-05:00

Jog is a night raga of the Khamaj thaat. It has a pentatonic structure that omits the second(ri) and sixth(dha) notes. It is identified by its characateristic use of both the natural and flat note ga. with the fifth note(pa) as its vaadi swara and the tonic sha as its samavaadi, it is a raga of pain and longing.

TA for CSC481

2004-11-17T21:11:00.000-05:00

I have been assigned the coolest course on campus to be a TA for,

CSC481: Game Design and Development

and, I think I'll have the students do some UT2004 programming - how about that!

Book collection !!

2004-11-17T19:22:00.000-05:00

There was a HUGE no.. HUMONGOUS (whichever is bigger) book sale of I believe all of Wake county's library books.

We bought FOUR BOXES full of books, isn't that very cool! Here are some of the books that we bought:

Management: The dilbert principle, in search for excellence, the IBM way, biography of thomas J watson Jr., from socrates to satre, Hitler's biography, ...

Science: LIFE magazine: Time, COSMOS and CONTACT by Carl Sagan, Exploring the Universe, Planets of the solar system.

Sci-fi: Asimov(8), Adams(4)

17 Robin cook novels, 11 Tom clancy, 9 Ludlum, 4 Puzo, 7 Sheldon, Archer, 9 Crichton and several others.

Oh and all this was for $5 per box in 4 boxes :) haa haa haa

Mommy's visit

2004-11-17T19:19:00.000-05:00

Today my mommy is going for her visa interview. I am eagerly looking forward to her visit. I will also be taking my MS graduation walk this december. we are moving to morrisville for a little while. Hope to go to Disneyworld over christmas. Last christmas (2003) was awesome with the disney christmas parade etc. ( which reminds me to get the pics from kurang ).

I can post all kinds of stuff regularly now coz of my ibook.

On Introspection

2004-11-17T18:51:00.000-05:00

so, I was thinking today about why the human race is intelligent and why are robots not. you can tell that I am an AI researcher to be. But thats not what I want to write about here. I am still in the process of thinking and the thoughts are raw-not very organized but fearing their loss from my working memory, here are the pieces that I am recording.

- The one thing that I believe might have played a key factor in not only the development of the human race but in evolution is - Introspection. The reason why evolution has been a slow process is because introspection is not something that we do well. For instance, I was thinking about all the conversations I had today and listening to tv and reading a sci-fi novel how we, very conveniently, leave all the philosphical reasonings to the philosophers. Infact if each one of us were to do some introspection into the things we do and have done every moment in our lives, we might be able to find answers to many questions that puzzle us. So, you'd argue, if that is the case then why do you not have answers to all such questions... maybe some day I will have them, thats why I live.

I think therefore I am. It's all in the mind. But it is very robotic. hmm... we try to make artificial robots intelligent but we do not observe how many real people live away their lives just like robots. We can do better, therefore I think.

Swami Vivekananda has said that if you meditate and focus yourself to one point in nothingness which is the problem you have, then you are bound to find an anwer to it. Reading partway through the "Hitchhiker's guide to the galaxy" and pondering over the behavior of Marvin, then sitting in the logic class thinking about non-monotonic logic and the philosophy of non-monotonic logic, it seems all this is connected somehow and I only need to find the link.

The thing that AI lacks is introspection, in a generic sense, something like commonsense introspection and reasoning from that.

Even though I think all of the above is a bunch of blabberings thrown together, for some reason I am still missing a link.

I'm still alive

2004-11-17T18:35:00.000-05:00

It's been a while now that I've posted something on this blog.

There's your standard excuse for any graduate student doing research and then there's a nother excuse of not being able to handle courses+research together(really it is hard) and on top of that being a TA.

But no, the deal here is different and here's a story - again from my advisor - to illustrate my situation....

Every grad student starts off with a boat and a couple of oars, in a lake. As he/she moves the oars and goes deeper and meets the river, the river current suddenly takes the boat further. Everything feels great and you defend your MS thesis. But now you are not satisfied. You see the shoreline receding beind you and say, well, I can go further and since the river current is taking you ahead at full speed it all feels great. But suddenly the water becomes calm again and the boat stops, no shore to be seen in any direction, no current to guide you. This is the ocean, and all you can do is thrash around with your oars until you find some shore.

Hence, the lack of frequent postings on the blog. But anyway, I'll still pen down a few whenever I take a break from thrashing my oars around.

Narsinh Mehta - The first gujarati Philosopher/Poet

2004-11-05T18:41:00.000-05:00

"... tej maa tatva tu, shoonya maa shabd tu..."

Here is the translation of his poems thanks to Sachin Ketkar. Thanks Sachin!

The following poem is my favorite:

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

You are the divine in the human flesh,

The fire you become among the elements,

In the void, you become the Word, which the Vedas laud!

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

O Sustainer of the Earth! You are the wind!

You are the water and you are the Earth!

You are also the outstretched tree blossoming in the sky!

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

Only to taste the nectar of being manifold,

You created the jiva and the siva and countless other forms!

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

Ornaments differ not from what they are made,

As the Vedas and other scriptures truly say,

Only their names differ once their forms are cast,

Gold is always gold in the end!

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

But the books messed up this truth and left it unsaid,

So the people worship whatever they like,

With all their hearts, words and deeds,

Thinking what they understand is true.

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

You are the seed in a tree and the tree in a seed,

I see you close just behind the veil,

You will never find him with your mind, says Narsinh,

Love him and he will manifest himself before you!

In this entire universe, you alone exist, Shri Hari,

Yet, in infinite forms you seem to be!

About Raag Manjh Khamaj

2004-11-05T17:36:00.000-05:00

This is my personal favorite.

Raag Manjh Khamaj
-----------------

Favorite raag of Maihar school, it is obtained by shifting shadaja(tonic) to madhyam(fourth). This changes all the internal consonances of the Khamaj, lending it new color and variety. A raga redolent with eroticism implicit in Sringara Rasa, Manjh Khamaj also has an undercurrent of wistfulness-a longing for someone not present, a nostalgia for times gone.

About Raag Brindabani Sarang

2004-11-05T17:28:00.000-05:00

I am writing this as I listen to the performance on a flute by Pt Hariprasad Chaurasia accompanied by Anindo Chaterjee on Tabla and Rupak Kulkarni on flute.

Raag Brindabani Sarang
----------------------
An afternoon Raag from the land of Krishna(Vrundavan or Brindaban), this is a charming pentatonic Raag which omits the gandhar(Ga) and dhaivat(Dha) while making use of both the nishad(Ni) - shuddha(natural) and komal(flat) in the descent. It evokes the mood of a langurous summer afternoon, with its bright sunlight and deep shadows.

Brindabani sarang is a relatively expansive raag, lending itslef to systematic note-by-note elaboration.

I would appreciate if people who know more about it could expand on this.

Acknowledgement: Music India, Sangeet Sartaj collection.

IBook :)

2004-11-03T11:37:00.000-05:00

Got my new iBook YAY!!

It is awesome, installed Unreal Tournament 2003 on it ... Hope they make dev tools available for that on macs.

Will soon post some pics ...

Classification of Indian Classical Raagas

2004-11-01T15:36:00.000-05:00

This was requested by Vimal. I recently read a book "Understanding Indian Classical Music" by Baburao Joshi (NCSU library: ML338 J68 1974).

This classification is taken from this book:

Group I - Light-Licentious

1. Bhairavi 2. Pilu* 3. Kafi 4. Tilakkamod 5. Mand 6. Khamaj* 7. Jogi 8. Pahadi

Group II - Scientific - Easy

9. Yaman 10. Bhimpalas 11. Bhoop* 12. Basant* 13. Durga* 14. Bageshri* 15. Puriya
16. Shankara 17. Bihag 18. Sarang 19. Bahar* 20. Tilang

Group III - Scientific - Serious

21. Malakounsa* 22. Todi* 23. Bhairav* 24. Jaunpuri 25. Poorvi 26. Miyamalhar*
27. Jaijayvanti* 28. Kedar 29. Adana 30. Bilawal

*I have these on tape/cd, mostly on flue but some on other instruments as well.

Flute Concert at Durham

2004-11-01T15:25:00.000-05:00

Went to the flute concert of Pt. Hariprasad Chaurasia on Flute organized by CRY America.

http://www.sjce.net/cryrdu/hpc/

All I can say is "simply superb". He started with Raaga Jaijayvanti, an exquisite evening raaga.

Blog Summarizer

2004-10-12T09:46:00.000-04:00

Here's another idea that I have: I need to create a report of all the blog postings that are there on someone's blog. Wonder if it's possible. Will have to look for it. A web services interface maybe ... hmm.....

My Win2K Desktop after installing the Longhorn sidebar

2004-10-08T01:35:00.000-04:00

xbox tv

2004-10-07T17:35:00.000-04:00

Games blog. Thanks to Ajay for the pointer.

Article on XBOX TV announced by Bill Gates. That rings a bell in my mind... hmm

http://www.ferrago.com/story/4555

eBay item 2491943747 (Ends Oct-05-04 15:44:43 PDT) - ARNAV MFD5200 COLOR W/BUILT IN GPS SET UP FOR WX500

2004-10-05T16:48:00.000-04:00

eBay item 2491943747 (Ends Oct-05-04 15:44:43 PDT) - ARNAV MFD5200 COLOR W/BUILT IN GPS SET UP FOR WX500

Arnav on ebay....

the top search item for arnav today costs $355.

Attending the AAAI 04 symposium

2004-10-04T21:49:00.000-04:00

So, looking at the people who will be present at this symposium and when you know Ed Hovy is going to speak right after your talk, is scaring me. Also the fact that most papers in the working notes are more like "finished work"+additional working notes ... means that my working notes would mostly look pathetic.

Anyway, it will be fun especially coz its challenging.

Makes me think of a geeky reality show where you get to give a talk in a very reputed conference not directly related to your topic ... he he ... any producers out there who wanna give it a try???

Reading and playing

2004-10-03T13:31:00.000-04:00

gotta do a lot of reading and playing this month.

Trying to install far cry right now so that I can test the c++ dll calls to it, but that has to wait till I play it atleast part ways through. Looks promising. After RTCWolfenstein and D3, seems like a natural progression - I guess.
Might even have to work

On the reading front also lots of stuff going on, mainly:

Semiotics of film language - Christian Metz
AI programming Wisdom -2
Possible worlds, AI and Narrative Theory - Marie-Laure Ryan
And a bunch of papers for the 582 class.

hopefully I'll have time to post some summaries/reviews soon.

'ite, Far Cry installation is complete and is screaming for my attention, so back to that....

ISRO's EDUSAT Lauched !!!

2004-09-28T01:20:00.000-04:00

Finally EDUSAT was launched. I am proud of the fact that we were working on one of its itty-bitty module's prototype 4 years ago as a senior design project(final year project).

It is a satellite dedicated to providing basic education to remote villages in rural India. Interactive video based classrooms with real-time audio feedback to a central studio. Very Cool!

Here are the details for interested readers:

http://www.isro.org/Edusat/Page2.htm

Google

2004-09-28T01:12:00.000-04:00

So, here's another posting that is related to google. For some reason I can't get the right pages to appear on top of the seraches for my name. Google completely misses this blog and it really beats me. Probably because there are not enough arnav jhala arnav jhala arnav jhala arnav jhala entries on this page. And maybe the rank of this page is not very high, but how will it be if people can't find it on google. bad google.

-Arnav Jhala
Arnav Jhala
Arnav Jhala
Arnav Jhala

SIGGRAPH Submission Guidelines !!

2004-09-22T21:03:00.000-04:00

This is a good one... What does SIGGRAPH reject?

http://www.siggraph.org/publications/instructions/rejected.html

Thomas Kuhn's: The Structure of Scientific Revolutions

2004-09-22T20:46:00.000-04:00

Michael mentioned the name of Thomas Kuhn's book during our meeting. I found it online so here's a reminder to take a look at it when I get a chance.

http://www.emory.edu/EDUCATION/mfp/Kuhn.html

-Arnav

2004-09-01T15:50:00.000-04:00

Arnav on Google !!

TO DO

2004-09-01T15:14:00.000-04:00

Here are the things that I are currently keeping me occupied:

Research

Translating Plan actions into Camera constraints (Priority 1)
Logical representation of thesis problem for static worlds (Proposal due Friday Sep 3)
Writing the IUI paper draft (Deadline Sep 20)

Development

C++ MClient for IUI submission
OpenGL world for demo (animating character, building a new scenario, testing out action classes)

Job

Preparing presentation for Rimmon-Kenan book chapters on Narrative Fiction
Updating course web-pages with slides
Finding the remaining readings and posting them on the page

-Arn

The World As I See It - Albert Einstein

2004-07-22T01:19:00.001-04:00

The World As I See It: "distance and a need for solitude...' "

Albert Einstein: Principles of Research (1918)

2004-07-22T01:19:00.000-04:00

AE: Principles of Research (1918)

Cargo Cult Science - Richard Feynman's address

2004-07-22T01:18:00.000-04:00

Cargo Cult Science

hamming's advice on research

2004-07-22T01:17:00.000-04:00

hamming advice

found this on nirmit's blog.

2004-07-20T16:18:00.000-04:00

What is this life if full of care, we have no time to stand and stare ...

2004-07-20T04:16:00.000-04:00

This is how I look at 4:00 AM at my office desk...

ok, I need a shave, and a haircut and a lot of other grooming stuff, but all of that after the journal submission...

2004-07-20T04:08:00.000-04:00

It is 4:00 in the morning and I am still working ... I was a little bit disappointed by the ending of Prince of Persia - Sands of Time. The worst thing that could happen to you is the gal who accompanied you through all the zombie crap finally forgetting who you are ... tch tch tch ... you can't just leave her like that ...

2004-07-20T04:04:00.000-04:00

I have started saving up chat logs with ppl, now this motivates me to chat more which is not very good, but still has its own advantages. I will probably go through these chat logs after 5 years and laugh at stuff :)

Dissertation Advice

2004-07-19T04:20:00.000-04:00

Dissertation Advice

I still fumble when I ask myself: "How would you describe your thesis in one sentence?". I know that I have finished my dissertation when I am able to answer this convincingly to myself.

Here's a funny link:
http://www.kuro5hin.org/story/2004/5/12/21333/7169

HindiSongNotes .... Hindi .... Music .... tabs .... chords .... midi .... lyrics

2004-07-19T04:00:00.000-04:00

HindiSongNotes .... Hindi .... Music .... tabs .... chords .... midi .... lyrics

Seems like a nice collection. Will check it out when I get time (VRJ article needs to be finished before the guitar is picked up again) :(

New tool idea for research to be built using JAVA

2004-07-17T05:20:00.000-04:00

I have always wanted to develop some real cool application in java. And as the saying goes 'Necessity is the mother of invention'. I finally have found a little tool that I am going to write in Java. This might actually end up being a very handy tool for researchers.

About Hello program and photoblogging

2004-07-17T05:04:00.000-04:00

This hello program is very cool, I can now publish posts as well as pics through this single interface. I will have to ask my blogging friends also to use it as an IM. I love integrated applications.

My Car: Saturn SW2 wagon

2004-07-17T05:00:00.000-04:00

Arnav's first car (well second actually: the first one was a 95 Mazda 626 partly owned with Sandy for 3 months from June-August 2001) ... 98 Saturn SW2 Wagon.

My office desk pic

2004-07-17T04:57:00.000-04:00

This is Arnav's office desk in Venture I Suite 110, Computer Science department.

New Sci-Fi novel idea

2004-07-16T22:46:00.000-04:00

Writing is a good habit. And I have cultivated a faint desire for a long time to write a full length novel. I like the idea of writing a novel. My inherent interest lies in science fiction. So, the decision is made to write a Sci-Fi novel. My roommate listens to the plot and thinks it should be based more on scientific theories than on fictitious and wonderous ideas that I have. No wonder he is going to be a good scientist, he likes publishing in scientific journals. I am writing an entertaining novel, and not proving any theories. The idea of basing the wonders of my novel-world to scientific theories sounds preposterous and is rejected.

So, blabberings aside, 'What's the plot?' - you might ask ... so I will start outlining the plot in my subsequent posts. Here are the few things that I need to do before I can start working on the actual plot.

Character bibles
Location description
--> I have positions for artists to give life to all of these - any takers?
Some study of neurology (for some basis to the scientific wonders)

thats all for now, keep visiting for more updates.

-Arnav

Slashdot | Videogame : Graphic Advances - Not That Important?

2004-07-16T18:43:00.000-04:00

I agree, gameplay is very important. Although any enhancement in graphics that add to the gameplay experience are always welcome.

I was playing Prince of Persia-Sands of Time and was very impressed by the fact that all the little intricacies of gameplay (Like the crumbling ledges) in the original Prince game have not only retained their place, but also look more awesome in their 3D rendering.

I am also looking forward to what the id guys have done to the original doom models, I am getting geared up to beat the crap out of their true 3D versions :)).

Slashdot | Videogame Graphic Advances - Not That Important?

Guitar classes at NCSU

2004-07-16T18:10:00.000-04:00

Three beginner classes TUE 6:30 PM, WED 4:30 PM and 6:00 PM. I am planning to sign up for the WED 6:00 PM class. so I get enuff time to cook afterwards.

Bett Padgett

Programming Guide for Linux USB Device Drivers

2004-07-16T02:13:00.000-04:00

Programming Guide for Linux USB Device Drivers

So, I've been thinking about a very cool little usb toy that might make a millionare out of someone who starts selling it.

But first, I need a prototype. For a prototype I need to know how to program USB drivers first, then go back to a drawing board and design the hardware for the toy.

It is not easy to become a millionare (unless you decide to participate in a reality show -- Any geek/nerd dating shows around??)

-Arnav

Extracting expressive elements from motion pictures

2004-07-16T01:42:00.000-04:00

Text based discourse generation systems have been founded on extensive research carried out by disocurse analysts (linguists as well as psychologists). Large corpora of text have been mined for finding patterns and relationships between discourse segments.

Now, there has not been too much work on mining video databases for finding such patterns. One of the main hindrances in undertaking such a task, is that there are no strict linguistics rules that could be applied to sequences of shots and frames. But movies are created from scripts, and scripts are written in natural language.

I propose a hybrid model for analysing films. The rhetorical structure of the discourse is first extracted from the analysis of the script. The script is annotated with discourse markers with the types of relations. Now, frames corresponding to these markers in the script are specifically analysed for a particular expressive element.

Of the expressive elements in films, tempo, rythm and tone have already been studied. I believe this approach will yield more accurate results because by annotating the script, we are incorporating the writer's storytelling intentions explicitly in the analysis.

...

The Economic Times

2004-06-16T23:28:00.000-04:00

The Economic Times

Welcome to Arnav's blog

2004-05-30T16:00:00.000-04:00

Just testing my new blog.