Okay, Day 1 started with me trying to schedule some sleep in vain. While still trying to setup a distcc farm, build Qt, I went at 10 AM for a break, followed by Atul Chitnis’ talk on “FOSS and Technology” at 10:30 AM. Atul was down with Chickengunya, but gave an excellent talk nevertheless. I don’t know about others, but I did have something to learn from it.

He started by saying that the talk would appear a little biographical to begin with, and went on to talk about his younger days, when he pulled apart a grandfather clock, and about his college project – they had quite a lot of learning in them. The punch line of the first bit of the talk, if I’ve got it right, was “Understand everything deeply, to the core. Because, if you want to develop something new, you had better know the inner workings of the system at hand”. Trying to take things apart is sometimes the way to understanding the inner workings of a system, and Atul stressed that this is of educative value, even if it fails. Even while putting things back together, you can learn something. He then explained how the FOSS way of doing things – open interaction, open source code, a nice community that is always willing to help – supported the user to dive deep into the working of technology. He asserted that it is through Free and Open Source Software that new technologies could be born, because building new technologies implies the requirement of understanding how things work on the inside – not just how to use an API. He also explained how FOSS was beneficial to students of technology, in enabling thinking, and in helping them to work with a large team. (On a side note, I was wondering whether the community-interaction in the scientific community would be analogous to that in KDE and other FOSS projects would, when he said this). He told the audience several anecdotes – I found the one about Harald Welte’s GSM stations the most exciting. The question-answer session at the end was very interesting. Someone who had attended one of Atul’s talks before could easily say that Atul had Chickengunya, because he wasn’t moving about as much as he would, but I don’t think someone who was blindfolded would! His talk very good in my opinion.

We quickly broke for lunch, because Shreyas’ talk was next. Shreyas re-did the “FOSS Foundry” talk after guaging the audience’s skills, although he had originally intended something different (and more exciting!). It was very interactive, and everyone in the audience was actively participating. He managed to keep people who had not slept the entire night awake and absolutely active. I’m sure everyone enjoyed his talk.

Okay, now let me talk about the pre-hackfest talks. I assumed the responsibility of building KDE this hackfest. Prakash has been running around doing many other things, so he goaded me to join him in the pre-hackfest talk for KDE. I didn’t know what to say – we didn’t even have presentations prepared – but I think we did something reasonable. I couldn’t afford to stay through the rest of the pre-hackfest talks (that were intended to give an overview of various organizations), so I can’t comment.

Then, the hackathon began. The kernel hackathon seemed to be the most attractive and the best in my room. Aneesh is a charismatic speaker, and from whatever I could gather during my compile-time breaks, he, Kamalesh and Prasad really had motivated their participants. Participants stayed up as late as 4 AM. I heard that they talked about the release cycle of the Linux kernel, helped people build the Linux kernel, went over some of the options in the kernel config, and talked about how to debug the kernel running on an emulator using gdb.

As for KDE, we had a few people staying as late as 4 AM. Kashyap was probably busy with other responsibilities, while Prakash gave an introduction to the KStars code-base and I sat trying to build KDE. (I made a lot of blunders, which is why it still hasn’t built). Later in the night, I live-fixed a simple bug to demonstrate the thought process and the procedures involved in fixing a bug and commiting it to the repository. Here’s a link to the commit. The participants made several suggestions, and we filtered the good ideas from the “unclean” ideas. Some folks were still hacking on KStars early in the morning with Prakash’s help, while I went back to the failing build.

The Mozilla JetPack hackfest seemed to be going really well, although they were in a different room, and I didn’t get to hear much. Siddharth was very enthusiastic. He was teaching people JavaScript, which was a pre-requisite for JetPack.

The GNOME mentor Arun found out that people didn’t know about function pointers, so they couldn’t understand event-driven programming. So he opted the route of “get your fundas right” and strengthened participants in the basics, so that they can hack GNOME today. I don’t know much of what happened there.

I was completely isolated from the Sugar hackfest, so I have no idea what went on.

Jai was conducting the ffmpeg hackathon. They submitted a patch for review. Today, they will be writing a decoder! That’s a lot of nice news . Kudos to Jai.

So that was Day 1. We knew what to do, unlike last year. Things were completely under control and went off fine, although KDE didn’t go as good as expected.

We should be hanging out on #iitm-hackfest on FreeNode. Do catch us!

Apologies: This post is coming MUCH MUCH later than it should have!

Wow! I wonder if you remember last year’s “Hackfest” at Shaastra, the technical festival of IIT Madras. Last year’s hackfest, despite failing miserably in a LOT of ways, had a whole lot of impact in other ways. I learned that it was partly responsible for the creation of two new LUGs, and the activation of one dormant LUG. It also indirectly played a role in bringing two new KDE developers, apart from quite a few things that I’m sure have escaped from my sight.

Last year’s hackfest was something that we coordinators placed really low on our priority list. S V Vikram, Sanjeev Sripathi and I were coordinators last year, and SVV and I had our most heavy academic semester going on. We really didn’t want to do the hackfest, but well, no one else pursued it. I did expect some juniors to do that! Well, nevertheless, we hardly spent any time working on it. The whole thing was a terrible disaster in terms of organization, but it actually did something significant!

Considering that this was one of those events which would actually motivate people to contribute to FOSS and to learn deeper, this event has entered the mainstream of Shaastra! We had four third year students – Prakash Mohan, Kashyap Puranik, Vinay Hegde, and Kashyap Garimella – applying as coordinators for this event. That felt really nice – something I wanted to do, but someone else (mind you, they are equally inclined towards the event as I am, or probably more!) to organize it! I decided not to pull myself in until those 4 days of Shaastra (we’re on Day 2 now) where I promised to be available to help with KDE – some contribution to KDE from me after my commit rate has been dwindling . It’s really awesome to see motivated juniors taking care of the event. They have worked REALLY HARD on it. Thankfully, their fifth semesters are lighter than ours was.

We got Atul Chitnis (tech. guru), Shreyas Srinivasan (GNOME, RadioVerve), Aneesh (Kernel), Kamalesh (Kernel), Prasad (Kernel), Jai Menon (ffmpeg), Vamsi Davuluri (Sugar), and Siddharth Agarwal (Mozilla) to give talks / mentor people during the Hackfest. We also have our very own Arun Chaganty (GNOME), Prakash Mohan (KDE), Kashyap Puranik (KDE) apart from me (KDE) mentoring folks at the hackfest.

I was hoping to publish this post before the hackfest, but I couldn’t rest because KDE was not building. Not that I’ve built it successfully now, but I’ve decided to do something else to keep me entertained while it builds. I was hoping for a better Day 1 than what actually went, but unlike last time, Day 1 was pretty good this time! Hopefully, Day 2 will see the real work happening.

I will describe Day 1 from my POV in the next post. Till then Tschues.

I simply couldn’t resist penning down my views on this subject after reading Smt. Lakshmi Sriram’s article in The Hindu (Chennai):
http://www.hindu.com/ms/2007/12/01/stories/2007120150210600.htm

I must admit that I am hardly familiar with forms of music other than Carnatic (and maybe to some extent Western Classical). However, I have attended a Hindustani concert or two, and I hope I do make sense when I talk about Hindustani music or other forms. Please correct me if I’m wrong. Of course, this post might be highly biased, but that is natural because it’s a personal opinion – take it or chuck it!

While not purely so, most forms of music have some aspect that results in ‘intellectual enjoyment’. One must be able to understand the nuances of the music to wholly appreciate it.

In my opinion, the ingredient of music that requires the least amount of musical training to appreciate is simple rhythm patterns. Somehow, I’m inclined to believe that this is why the likes of Shivamani are extremely popular – because they come up with simple, but yet innovative, rhythm patterns which the common man can understand and appreciate. The other ingredient is Lyrics, preferably in English. I attribute the popularity of those forms of “music” that I would love to derogate by branding them as ‘noise’ to the fact that they are rhythm-intensive or Lyrics-intensive, and thus understandable by most people.

Then there’s harmony. I found out from my friend that not everyone can actually appreciate harmony. Western music primarily relies on harmony. Western Classical music utilizes ensembles of instruments to produce harmonious tones. Other ingredients like the tonal content or rAga, are harder to pick up unless one is trained in the particular form of music.

Carnatic music involves not only a lot of rhythmic complexity (eg: Pallavis, or stuff played in a taniyAvartanam) that is hardly understandable to the untrained ear, but also a lot of tonal complexity (heavy bhRgas and gamakams) that requires a really fast ‘Fourier Transform’ in your brain! What is suprising is that training can make the brain capable of ‘parsing’ every single note in a tonally complex phrase of music.

While Hindustani music might be tonally complex, it is mostly mellifluous and doesn’t involve as much rhythmic complexity (unless I’m mistaken) as Carnatic Music. Carnatic music is sharp, whereas Hindustani is mellifluous; and unless your ‘Fourier Transform’ is fast enough, you can’t appreciate T N Sheshagopalan’s 3rd kAlam swaras in the Alapanai (I still can’t!), which will just sound like drab nonsensical oscillations with no tonal beauty, explaining the popular impression of Carnatic Music. Hindustani Music, especially when rendered popularly, on the other hand, doesn’t require as fast a Fourier Transform – except for some parts of the concert. Thankfully, Hindustani music has these slow phrases which the untrained ear can appreciate, making it more popular.

So I would conclude that the popularity of non-classical forms of music, and above that, that of Hindustani over Carnatic is mostly because Classical Music, and in particular, Carnatic Music doesn’t pander to the average unacquainted man but offers a steep learning curve and calls for appreciation at various levels – emotional and intellectual.

I had to dig through several pages of tweets to find this command:

sox -d -d reverb 80

It basically feeds back from the mic to the speaker with the reverb effect. So if you like a lot of reverb while playing a musical instrument, for instance, but your room doesn’t have the reverb, then you could run this to get the required reverb.

I’m putting this down here, so that I can search for it the next time I want it

There are other methods, like:

rec -t wav – | play -t wav – reverb 80

But that requires a lot of processing time, so there’s a significant delay between the input and the output…

After some internet searching, it wasn’t hard to find enough inputs to write a IRC bot using Python. I hardly know any Python, but the language being so simple, you could write code in it just right away. Needless to say, the best part was that there were libraries for both Feed Parsing (duh!) and for IRC clients. A simple one hour long digging through documentation resulted in the following put-them-together code:


#!/usr/bin/python
#
# IRC b0t that keeps track of RSS feeds
#
# Licensed under the GNU General Public License v3
#
# Copyright (2009) by Akarsh Simha

while 1:
while len(msgqueue) > 0:
msg = msgqueue.pop()
for channel in channel_list:
server.privmsg( channel, msg )
time.sleep(1) # TODO: Fix bad code
irc.process_once()
time.sleep(1) # So that we don't hog the CPU!


Forgive me for writing ugly code. I’m a newbie pythoner, and this is my first python script that actually does something useful.

On a side note, I attended a semi-classical concert by Anil Srinivasan (Piano), Unnikrishnan (Vocal), and B S Purushottaman (Kanjira), and it was one awesome experience. The auditorium (the Lady Andal School auditorium, where Margazhi Raagam was shot AFAIK) was simply superb. The ambience, and the stage lighting were amazing too. It was the confluence of the music of all three artistes, the auditorium and the lighting that created the paradise that we experienced.

Snow-capped-peak-2, originally uploaded by Akarsh Simha.

Photos of my trip to Dibrugarh are finally up, after I’ve overcome eclipse-depression. A few processed ones are on my Flickr Album at http://www.flickr.com/photos/asimha while the bulk, raw photos are on my Picasa Album at http://picasaweb.google.com/akarshsimha

This trip to Dibrugarh was the most miseventful and depressing trip so far in my life, IMO. Never have I felt so disappointed before. Dibrugarh neither turned out to have a good lot of scenic travel options around, nor did we get to see the eclipse of the century – the 3.5 minute totality predicted at Dibrugarh. All we got to see was the Brahmaputra, and the darkening of the cloudy, overcast skies during the totality (and the random behaviour of birds).

The only “perks” of the trip were insignificantly few – that of having a brief look at Kolkata, some good photographs involving dragon flies, the Brahmaputra, spiders and tea plantations, and some time spent away from the computer, with friends. Not enough and not worth it – just like several of those recent Hosahalli trips under overcast skies.

Besides, I hate travelling without good company – sitting doing nothing in the flight, for example. All I can do is to crib and compose this post, so that I can vent out some frustration of an untriumphant waste of a journey at the least, and ease myself a bit.

Over all, it was really disappointing to go all the way for absolutely nothing. This was much worse than that pathetic, most pointless visit to Kavalur that happened last year around the same time (which I did not blog about), which was much better in that I did not miss any significant event. These are the times when you start becoming agnostic… Wish I were in Varanasi with Amar and Vivek instead.

I amn’t alone – all of us were really disappointed. Particularly the two Pavans who calculatedly (and I say their arguments were logical!) changed the venue from Varanasi, which we had initially planned, to Dibrugarh. I don’t blame them for anything, because their reasons to favour Dibrugarh were absolutely sensible. Many folks pointed out that Dibrugarh was at a cloud-cover maximum as per predictions, but then Patna, which was at a predicted=cloud-cover minimum, had only overcast skies. It’s just our bad lack… really bad luck.

I guess I will be making a lot of trips to Argentina just to compensate… at the least I will end up visiting and exploring a foreign country instead of a boring, hot, and monotonously Indian town.

I’m currently at an enthusiasm minimum, so don’t expect any trip photos in the near future.

KStars now gives you an option to correct for General Relativistic effects near the sun!

According to Einstein’s Theory of General Relativity, light rays bend around the sun because of the sun’s gravity, and this correction is predicted to be about 0.0005 of a degree for light rays that just graze the sun’s surface. During the Solar Eclipse of 29th May 1919, Arthur Eddington verified this theory experimentally.

KStars now lets you simulate this experiment, which essentially measured the apparent positions of stars near the edge of the sun and showed that General Relativity accounted for the difference in the observed and expected positions of the stars.

Fire up KStars, hit ‘0′ to center the sun, and zoom in quite a bit to see the stars at the edge of the sun. Now hit ‘r’ to toggle the relativistic corrections!

As you all know if you’ve been reading my blog, I spent my summer at the Tata Institute of Fundamental Research (TIFR) at Mumbai, as a visiting student under their Visiting Students’ Research Programme (VSRP).

I’m now back to Bangalore, with lots of time to spend blogging the proceedings of the last one month.

I somehow felt I didn’t do much work during the last one month. I computed the Jacobian of the operator that I talked about in my earlier post, and it turned out to be unity for all that hard work . But well, it was a good thing it was unity because it didn’t complicate the equations all the more! So we were able to write down the kernel for the Navier-Stokes equation. Then came the task of converting symmetries into constraints on the two-point correlation functions of the velocity field. I wrote down Ward identities, but I couldn’t solve them. Prof. Spenta gave me an ansatz to try and I didn’t have much luck with it. That apart, we ran into some interesting issues regarding anamolous scaling dimensions, which still need resolution.

But here I’m, back in Bangalore, hardly spending time on Physics. It’s just that I’m lazy to start…

Overall, I must rate VSRP as a good experience. My sleep requirement seems to have reduced drastically and I sleep deeper these days, thanks to the fact that we were baccommodated off-campus at Wadala (1 hour away from TIFR / Colaba) – which was, according to me, one of the major downsides of VSRP. The other downside, IMO, is that TIFR is a research institute, and professors don’t know what to expect from undergrad students. However, in my case, Prof. Spenta very quickly adapted to my ignorance and I patiently taught me fundamental concepts! But then, it is good exposure to the scientific community.

Life in TIFR is really comfortable. The west canteen’s awesome food is pretty well-known. Plus, work timings are very free. There are colloquia every now and then and it is really interesting to see biologists attending Physics seminars and vice versa – promoting interdisciplinary exchange of ideas. I think TIFR is the place to be in India for research, and VSRP is an ideal summer internship programme.

This blog post might have some use for freshers who have the aforementioned question in mind – to bring or not to bring a laptop in the first semester. This was a reply that I sent to a student’s parent, and thought it might be useful to others. This post will expose the pros and cons of having a laptop, but will leave the ultimate decision to you. As a side note, in my case, I took my desktop in the first semester because I was (and still am trying not to be) a computer “addict” (thankfully, in a better sense than a game addict). This post might have a bias towards physics department students once in a way. So here goes.

Pros of bringing a laptop / buying a desktop:

• All students will do a course on Computational Methods in their first or second semester. The Engineering Physics students will probably do it in the second semester, whereas the dual degree MS in Physics students will do this in the first semester. A laptop is useful for this, as you might have assignments where you will need to write and submit a program.
• Although there are places to access a shared computer (which I will outline later), you will not enjoy the comfort of your room while studying / working. So it’s a good idea to have a computer in your room.
• Much of the teaching in the first few semesters is done via computer-based presentations (commonly dubbed “power-point”) rather than on blackboard, so a personal laptop / desktop computer will be useful to study for the examinations.
• From the second semester onwards, internet / intranet connectivity will be provided in hostel rooms for restricted timings. Thus, personal laptops / computers will be useful from the second semester onwards, since it will make it easy to communicate or get information.

Disadvantages of bringing a laptop / buying a desktop in the first year:

• The internet facility is not provided during the first semester in the hostels
• There are some very good common computer facilities, particularly if you are a Physics student. The Physics dept. computing facility is a really well-maintained facility that is almost always accessible. A list of other common places to access a computer figures at the end.
• Most students bring a desktop / laptop only in their second year. Thus, any laptop / desktop available in the wing will be in high-demand by your wingmates! Sometimes, this can get irritating – but it isn’t a serious problem if you don’t mind helping your wing-mates. I remember that although I used to try to be liberal with my desktop, it sometimes used to get me frustrated when I find it booted back into Winblows from Linux. I remember that I formatted the Winblows disk some day suddenly in frustration.
• It’s a good thing to avoid internet / game addiction and explore exciting hobbies that you can pursue in the institute in the first year. Internet can be addictive if you are seeing it for the first time at IITM. If you’ve already been a heavy internet user – then this point doesn’t concern you, as you’ve probably outgrown the initial hype now.

Places where you can access a computer:

• The Hostel Computer Room: This is a common computer room in the hostels. Usually, it is badly maintained – but then, there’s an administrator whom you can “pain” and get stuff working if it isn’t. This is the nearest computer-place near your room. However, by Murphy’s Law, all the computers here are occupied when you most need them – so it is better to know other alternatives!
• A Senior’s Room: Most seniors understand how important internet / a computer is for your survival (at least I do!) and will let you use their computer after you are subject to some basic “interaction”. I remember that a very friendly 4th year made a very open offer for us to use his computer whenever we wanted, and we actually used it for preparing some presentation.
• The Department Computing Facility: This is a facility that exists (well, rather is supposed to exist) in every department. Although I do not know, I’ve heard cribs that these aren’t well maintained in most of the departments. However, the Physics and Computer Science departments have very well maintained DCFs that are open for most of the time when one is awake. The administrators are very approachable too. This will be about 8 ~ 10 minutes on cycle from your hostel
• The Digital Knowledge Center: This is a place in the library which is typically crowded and has no place left for you to access a computer . It has mostly Winblows machines, which is why I’ve never been there after my first semester. If you do have a laptop, you have WiFi connectivity in the library, and the DKC has reasonable signal strengths as well. There are some Linux machines, but they are probably servers. The DKC will also be about 8 ~ 10 minutes on cycle from your hostel.

And a final point: Don’t get too worried. Things will be fine.