Chennai first week impressions

I have been in Chennai for a week now, and still have trouble getting used to the high humidity along with the high temperature .Ofcourse, I have seen high humidity in Allahabad and mumbai, but then Allahabad and mumbai were still slightly different, while mumbai had a consistent breeze to counter the problem, allahabad had a less salty humidity. I guess these are all just minor changes my body should adapt to and the only reason, it hasn’t done so yet is me being in A/C most of the time. In any case, i can’t change it.

And ofcourse with the language(tamil) come back all the problems i had with it. i.e problems of vagueness.

But, what has blindsided me is my reaction to the tamil i hear and the conversations people try to draw me into and sometimes succeed. Damn it… Infact, i am overwhelmed by the amount of random ambient conversations/noise i understand. So much so that, i want to put up FM radio and tune in to a English/Hindi(akashvani??) channels.
Looks like there’s a breeze going on today, reportedly the start of North east monsoon in Chennai. Work has so far been dull. i.e to say most of the time has been spent setting up my system or fighting the internal IT team for their inefficiency. Or sometimes getting outright pissed off with Windows 7.. Got the code base + dependencies running on the development box.
Infact my target for the week ahead is to getting the basic VIP migration tool implemented in the basic form by the end of the week. From what i know of it, it seems like a cakewalk, but there are a couple of calls with John Sage, that might throw up a curve bal. Let’s see how things work out. Am sure if i put in extra effort i can pull it through in any case(i.e running a full test scenario in the test/dev boxes) we now have required routers to do the same. The week so far has been frustrating enough(since it has had very little coding so far.) I would like to get this done the next week, if only for relieving that frustration.

And God.. the roads are bad ..especially during the rains…. Guess being a port, roads require a lot more maintenance, but that’s no excuse for not maintaining it.
And god, i hate the unpredictability of chances of rain. I guess it’s part of being a port or a seaside. In the plains or hills, you can always say it’s gonna rain in a couple of hours, but here i have to check with the meterology department…

Windows 7– fuckups

Just trying out Windows 7 at my new company as it’s the supported default OS here. Changing the kb layout for an user should change the layout right after the username is selected for password. Damn windows doesn’t do that… Bloody thoughtless design… I mean, what’s the point of changing kb layout if i can’t use the new layout to type in my password and login???? ***#$@@#$@

And an even bigger fuck up perhaps is that it seems to pick up the right kb layout in case of locking the system ..Damn am never shutting down the system anymore..

Update 07-Nov-2011:
Other fuckups.
1. The authentication window assumes a default domain and displays it right below the username input box. But it’s non-editable. If you want to use a different domain you have to enter the in the username input field. Damn…. Me having never had used windows domains spent quite a lot of time wasting….how hard can it be to provide another input box for Domain name?? IE 8 on Windows xp does something different. it populates the domain extracted from the url. Interesting choice..
2. And no, don’t even get me started on authenticating for use in privileged mode. I have to type in my credentials every time i am trying to install stuff?? Damn it.. is it configurable somewhere to remember my credentials for a given amount of time… AFAIK no…

Lotus Notes Domino– web interface:
This one’s a classic. If you mistype the password in the mail page it fails obviously to say not authorized. The fuck-up is that the page doesn’t seem to clear the cookie. so whenever you visit the mail page again you get the same error message..

Laptop observations

About a week ago,My Laptop started becoming less responsive suddenly and spuriously(i.e: At unexpected,unpredictable occasions and passingly) , even though i hadn’t changed/touched it i thought it might be Xmonad and switched to Gnome Shell 2 days ago. Damn, what a fool i was.. No difference, except perhaps, me getting used to mousing and getting annoyed with Gnome3’s habit of bunching together similar windows and forcing me use mouse to switch focus across them. Anyways, the problem was before my eyes, and i had been blind. There had been this other issue of my LVM (non-system, i.e: not /home or /boot)disc failing on occasions for quite some time about a month or so. The error being read-only filesystem or unmountable partitions. And sometimes zero size partition while trying to mount or doing a fsck. I initially dismissed it as just an issue with the chord of the ext. hd being a little shakily connected and my typing on laptop upsetting it. I could not have been more wrong. It was a failing old hard disc reaching the end of life. with the help of google + serverfault search, i ran the smart monitor tools. and realized it’s failing. Now, am more careful about my data and backing it up. But i have to wait till i get a new hard disc. With my moving town and stuff, i it will be a couple of weeks.

Haskell –my observations

Just trying out haskell. For all the coolness and power of infinite lists, actual performance looks absymal. Just implemented the fibonacci series as a recursively defined infinite list. And running it with naive basic options takes a visibly long time. Clearly under-utilizing the CPU resources i have. But i guess, it’s not very surprising and i ought to try it with options for memoization and other stuff.
Am beginning to doubt in my obsession with haskell’s pattern-matching i have run into a bad case of tail-call recursion.
Update: I was wrong. it turns out tail-call recursion is actually good for performance. My prog was horrible because it repeated every fibonacci call twice…. Thanks to stackoverflow for pointing out my ignorance.

Nagios Monitoring service

Nagios monitoring service is a brilliant example of open-source high quality software. Not to say, all open-source software are high quality, but one look at nagios and how it gives so many default monitoring, out of the box, and you see why it’s worth learning. Ofcourse, there are a million possible variations in the requirement of a Server Management and a million other services,hardware,metrics, but the beauty is they are all mostly configurable. I doubt any proprietary s/w team could support as much variety, by virtue of being a limited team size. (Pretty similar situation with Linux). On the downside, and yeah, i do think there’s a downside, this necessity of being able to support a wide variety, forces quite a lot of abstractions. the downside being, abstractions leak, slow down performance/speed(Look at Synthesis, Alexa Massilini). In the end, i would say that’s probably a dilemma any software made in the current world faces and makes some point of tradeoff. Which is better/worse, we can all speculate, but will be tested by time.


The first things that strikes one while watching a Quentin Tarantino movie is the music. How the bgm volume varies up and down, and in general is louder than other movies, and perhaps, how it’s an eclectic, surprising variety of music.. And that they know where to cut out bgm and just let the natural sounds play.. Kill Bill is brilliant for the very same reason. Not to mention how similar the Japanese sounds to my Tamil,Malayalam,kannada,Hindi trained ears and auditory cortex

Xmonad usage and effects on mouse skill

Ever since i have moved to XMonad, i have been using the keyboard a lot more often. I almost didn’t notice it,(given i preferred keyboard even before) but recently noticed that i seem to make more mistakes in moving the mouse to a location. I mean there are quite a few times i have clicked the mouse on the wrong menu. I initially i dismissed it as to being tired and late at night. But noticed in the day too. Perhaps, the most important change maybe that, i don’t use the mouse except when moving around on a video stream(youtube,vlc etc) or browsing around the web idly reading stuff.

Update: 14-Oct-2011:
Been on the gnome shell for the couple of days and realized one more thing. Xmonad+mouse is problematic in the case of windows that are a lot smaller than the others. Like a skype call window. the moment you move out of that small window area, your focus is in a different window.. so if you use the mouse to move between windows and then type, you have a jarring experience and if you also have a touchpad, god help you….

T20 phenomena/effect..

I look at T20 scores and wonder whether they can be translated to 50-overs cricket. Take for example the case of RCB vs South Australia today, there were 419 runs scored in total in 40 overs for the loss of 10 wickets. Makes me wonder if 10-20 years down the line, there will be teams scoring 500 runs in 50 overs. or will 400 become the on par score. with 500 a possible highest score. We know 434 is not safe enough and might still be chased. This reminds me of my childhood. My cricket interest was slightly different from others in thesense that, throughout the match i will be calculating runs per ball/ runs required per ball. Especially during the chase. i.e: to say even when it was unlikely, it seemed possible to me because i was calculating rather than considering historical scores/ lost wickets/ experience of batsmen at crease etc.(the stuff the commentators usually go on about). Like fastest possible 50/100 will win the match and stuff.There are enough balls for both.

Cut back to the current day, there are T20s where that level of run-rate happens and goes on..Yuvraj hits six sixes in an over. One would expect i should be excited by these T20 matches and line up to watch them.. Funnily enough, i don’t even bother to watch the online streaming videos, which is the easiest form i can watch. I am still content looking at cricinfo scorecard and sometimes reading the commentary. Even more dampening is the fact, am not even following the schedule/every match. Merely, checking the scorecard when i feel bored.

Human cognition and Newton’s inertia

Have you ever noticed how the autowallas standing in the auto stand tend to ask extra bucks over the meter, compared to the ones already on the road? Makes me wonder if there’s an equivalent of Newton’s law of inertia in human reasoning/thinking too. Ofcourse this is just hand-wavy theory and needs to be justified with experimental data that conform to a partial differential equation. But may be it’s just the embodied cognition at work. i.e: perceived effort(of the autowalla) is higher because the extra force to overcome the initial physical inertia. In either case, i am about 80% confident that the p.d.e’s will model the phenomena reasonably well.

Moral: So what is the upshot of all this? Well, i definitely don’t want to put in the effort required to do all the data collection, analysis and modeling. But, wonder if i can see similar trends in my behaviour. Take the case of my running routine, when i look at my calendar, as long as i have given 1 day gap between the running days, i seem to keep the routine, but once i take a 2-day break, that’s it it becomes 4-days/ a week very easily.