Wednesday, December 12, 2007

I inadvertently found a Antegooglewhackblatt

The query in question is whieghfield. And as a consequence of the Googlewhackblatt Paradox, this word shall no longer be an Antegooglewhackblatt, but a Googlewhackblatt.

Here are the screenshots

Wednesday, October 17, 2007

"Chaos" (The Movie) And The Pigeonhole Principle - An Interesting Goof

***Caution: Spoiler Ahead***

While reading the list of goofs in the movie Chaos, I noticed that nobody had pointed out this important mathematical goof in the movie. At around 1:13:00 time in the movie, note this set of dialogs

  1. "...about a billion dollars..."
  2. "Now. If you look at all the transactions, no 2 amounts are the same, and none of them are over 100 dollars"
  3. "...less than 100 dollars, from, say, 10 million accounts, no reflex"

According to the Pigeonhole principle, it's not possible to make about 10 million transactions, each less than 100 dollars, such that no two transactions have the same amount [assuming the least count is 1 cent]. It's only possible if transaction limit is 100,000 dollars.

However, if 10 million transactions [all withdrawals] are made such that no two amount to the same, then the minimum amount withdrawn would be ((10,000,000)*(10,000,001)/2) cents - about 500,000,000,000 dollars- far, far more than 1 billion dollars.

Pigeonhole Principle - Also known as Drichlet's Box / Drawer Principle

It's unfortunate that the people behind Chaos overlooked this important mistake -especially when they were trying to make a movie that apparently is so dependent on mathematics and Chaos Theory.

Finally, to withdraw 1 billion dollars such that no 2 withdrawals amount to the same, we need at least about 450,000 withdrawals [starting from 1 cent, 2 cent, 3 cent and so on till about 450,000 cents].

A Goof In The Movie "Crimson Tide"

Friday, October 12, 2007

Are the top web search engines correspondingly related to the top web browsers?

I should begin by admitting that this idea sprung up in my mind (probably) as a repercussion of my watching the movie "Chaos" today morning (there isn't much of a direct relation between the movie and this idea though).

This is pretty straightforward- the current top web search engines (Google Search, Yahoo Search, Live/MSN Search, Ask.com Search- in that order) bear resemblance to the current top web browsers (Windows Internet Explorer, Mozilla Firefox, Apple Safari, Opera Browser- in that order again) in some striking ways- apart from a broad similarity in their current market shares.

I'll list those here

1) Google Search and Windows Internet Explorer- Both these market leaders are zippy (a clean install of IE 7 runs and renders fast- and most of the complaints of IE 6 or IE 7 being slow or heavy are a result of bad add-ons or wrong settings). Both are lightweight- IE is the least resource consuming browser on Windows operating system, when stacked up against the other 3 in question here- while Google Search is the lightest web search engine out of the 4 being considered here.

2) Yahoo Search and Mozilla Firefox- Both these powerhouses pack large number of useful features, and both "get the work done", although in the process of achieving this, they tend to be heavier than their more popular rivals.

3) Live/MSN Search and Apple Safari- Slick looks mark their main resemblance, as is the tremendous promise both of them hold. Both are expected to gain market share at the expense of their larger rivals.


4) Ask.com Search and Opera Browser- "Innovation" and "Innovative features" are the hallmarks of both of them. Each one of them is an underdog, fighting for survival, constantly innovating in the process. Each of them provides innovative features and tools not to be found (yet) elsewhere.

I can even add AOL Search and Netscape Browser/Navigator here- with feature richness but undue bloat characterising them both.

The important and unresolved point, however, is- does this correspondence have something more than what meets the eye (something scientific, some patterns maybe)? Or is it merely a consequence of an unsettled mind watching a movie as confounding as "Chaos"?

If we are able to identify some common characteristics in the products and services talked about in this post (which caused them to attain their respective positions in the market), we shall be a step closer to identifying and listing "Rules for Success".

P.S.- Identification of these "Rules for Success" is an area of my interest, and I have put significant time and energy into it.

Monday, October 08, 2007

Microsoft Silverlight forces the web browser process to run at 'Above Normal' priority

I felt like puking after what I just noticed- I had downloaded and installed Microsoft's Silverlight runtime on my computer a day back. Today I was browsing the web on 2 browsers together- Opera and Mozilla Firefox. When I opened www.microsoft.com in Firefox the website threw up a Silverlight based animation [whereas it used JPEG or Flash/SWF while I had not installed Silverlight]. The animation slowed down my computer [Intel Celeron M 1.5 GHz based laptop] significantly and perplexed me.

On analyzing what was wrong, I noticed that firefox.exe process was running at 'Above Normal' priority, and manually trying to lower it to 'Normal' simply failed- firefox.exe would automatically [and immediately] switch back to 'Above Normal' priority [but that's only if Silverlight is active- as soon as the tab having www.microsoft.com was closed, Firefox could again be switched to 'Normal' priority- so I conclude that it was Silverlight which forced Firefox to run at 'Above Normal' priority].

However, to me there is something else that is far more important- a Flash animation running in parallel in Opera browser slowed down immediately [and significantly] when Silverlight got activated in Firefox, whereas the Silverlight animation running in Firefox appeared to run faster [relative to the Flash animation].

To me, nothing else is more important than what I wrote in bold above. With this [yet another] wicked tactic, Microsoft gives a user not 1 but 2 false impressions:-

1) Silverlight is made to look better than it is- by forcing the process to run at 'Above Normal' priority, Microsoft is essentially fooling users by the apparent performance of Silverlight

2) Flash based animations running in other browsers are made to look slower than they actually are- user gets an impression of poor performance of Flash, compared to a parallel running Silverlight animation

Also, this is going to impact performance of other browsers as well [which a user runs in parallel]. And because this has not been documented anywhere, nor is the user informed about this, users will unknowingly develop false impressions of speed and performance of other runtimes and other browsers. This is nothing short of pathetic!

Personally I feel that Microsoft deserves to be sued for such a lame and desperate act- this act is an admittance on Microsoft's part that it lacks the engineering talent required to produce a high performance runtime that can equal, let alone better, Adobe's Flash.

My 2 cents for Microsoft:-

"Real men triumph without cheating."

(Microsoft does possess the talent to decipher what I meant...)

Monday, September 10, 2007

The importance of using a high fidelity speaker system

I have frequently noticed that people give disproportionately large attention to having a good visual display subsystem, compared to the audio subsystem. I have seen many friends bothering less about the speaker system when getting a desktop computer assembled.

Personally, I feel that this tendency is not good. 2 reasons for this

1) I am an avid player of music on my Yamaha PSR-330 electronic keyboard cum synthesizer for over 9 years now. I believe I have a good sense and understanding of music and sound. And my experience with music/sound over years is that although sound plays a very important part, its contribution is more 'silent' as compared to the contribution of visuals. To put it more clearly, in many scenes in movies, it is the sound running in the background that creates the right mood/emotion, although this is hardly ever explicitly recognized by audience. Thus sound is more like a silent warrior that does play a significant role, but its importance is realized by only a handful of people [musicians and singers especially]. Although in real life, sound has this quality called 'prominence', in many scenes in movies [especially in the emotional ones], sound is not prominent, as larger proportion of attention of audience is towards the visuals. Visuals, in general, get a disproportionately large amount of credit for the overall effect a scene creates. This reason highlights the importance of the role played by sound.

2) I draw the following analogies between display and audio subsystems

Screen size (of a display) = Output power (volume of a speaker system)
Color gamut range (of a display) = Frequency range (of a speaker system)
Brightness, contrast etc (of a display) = quality of sound output (of a speaker system)

If any single component of a display subsystem is lame, it is outrightly turned down by people. (from point 1) A speaker system too must have high fidelity, lest the effect of sound intended by artist/creator never reach the audience.

One final word- Watching a movie on a small screen with a poor speaker system produces a fraction of effect that the same movie would have produced had it been seen on a large screen with good speaker system.

p.s. I recently bought 2 sets of speaker systems [both are creative sbs-370 systems- my own belief is that, as of now, this system gives highest value for money among all 2.1 speaker systems i know, in addition to reproducing sound quite faithfully]

Friday, August 31, 2007

i have upgraded to a new system of working

the key benefits that this new system imparts are
  • addition of ability to say 'no'
  • improved time management
  • increased mental concentration
  • reduced overheads
  • reduction in per-task time
  • removal of redundant elements from the system
expected long term benefits of this new system
  • time savings
  • reduced mental stress
  • increased leisure time [stems from 'time savings']
the key penalties of this system are
  • potential of reduced warmth in close relationships (in the short term)

Tuesday, August 28, 2007

qualms i have with the new user interface of google's orkut

the good things about this new interface
  1. user interface has been designed to mimic the look of 'widgets' on windows vista desktop or apple's mac os x desktop- this i believe in itself is not necessarily a good thing- the point is that this approach segregates [or 'segments'] the interface elements into distinct units, thus providing clarity
  2. addition of nifty little features [though no 'big' feature]
the bad things about this new interface
  1. more resource intensive, and thus noticeably slower on slower computers [and especially on web browsers like mozilla firefox]
  2. although 'segmentation' plays its part in providing cleanliness to the interface, the individual segments themselves are more 'filled' [or 'stuffed'], and thus the overall complexity of this new interface is [to me] higher than the old interface
  3. on 800x600 displays, scroll bars are forced, rather than a 'reflow', thus hiding the rightmost portions of page all the time
my belief is that though the new look does have many little improvements, some of which are nice too, it also has an increased overall complexity, and those users who have liked it have probably liked it because of it being different from the previous interface, rather than because of it being substantially (and absolutely) better.

finally, users with 1024x768 or larger displays and high speed computers will remain shielded from the more annoying ill-effects of this new interface, and so are more likely to like and appreciate this new interface, at least initially

Thursday, June 21, 2007

is instrumental music 'less distracting & disturbing' than music with words? i feel so

the background first...
i have been playing instrumental music on my yamaha psr-330 electronic keyboard cum synthesizer for over 9 years now, and before that i played it on my casio sa-11 electronic keyboard for about 5 years (maybe more).

i like to listen to instrumental music; of course not just any, but i like select ones, just like we all have our individual preferences when it comes to listening music with words.

now.
what i have noticed over years is that instrumental music seems to distract less and disturb less than music containing voices/words/lyrics. one day i started to think about it and first felt that maybe its because the total 'quantity' of sound contained in an instrumental song is probably less than the total quantity of sound in a song with words [i dont know how the 'quantity' of sound, or 'amount' of sound, contained in a sample is measured, but it looks like joule should be the right unit for this, just like joule can be used to express 'amount' of optical energy, or just like momentum can be used to express the total quantity of motion contained in a body].

then later i felt no, there are instrumentals having a lot of sound, and yet they still seem to distract less than music with words.

the only plausible reason i could think of was that brain's speech and language processing section(s) remain unengaged when listening to instrumental music, while they are engaged in both the recognition of, and semantic analysis of lyrics of music with words (even if sub-consciously).

i have noticed that when doing tasks that involve words- studying, reading a webpage or reading a newspaper- instrumental music does not distract or disturb, whereas music with words does (or at least it does more). this looks consistent with the undeniable fact that brain's speech and language processing section(s) are engaged when listening to music with words.

i dont know whether this idea i wrote here is true or not, but i thought about this, and felt like writing it, so did.

ah, yes, as i write this blog post (a task involving words), i am listening to faithless- drifting away (from "buddha bar"), an instrumental, and believe me it doesn't disturb an iota...

update
one new thought that came to my mind was whether listening to music with words, but of a language which we do not understand would cause less distraction or disturbance than music with words of a language that we understand (like english).
this question looks interesting to me, and to find out the answer, i fired O-Zone's Dragostea din tei [a romanian song i like but do not understand the meaning thereof] and indeed felt some difference, thought its nature remained unclear.
i feel that because i do not understand this language, so my brains's semantic analysis sections(s) should remain either free, or overbusy [trying to make out meanings from the jumble of words being thrown at it]. the language recognition sections(s), however, will remain busy even if i do not understand a language.

Thursday, April 19, 2007

capturing animations, images, music, poems, stories, textures and videos straight off the brain of an artist

in future, textures [for use in games] and special effects [for use in movies] are going to be increasingly complex. although developer tools too are going to improve, so as to enable development of more complex content faster and more easily, yet it still requires significant effort and time.

today i was thinking that in future, when we have the technology to precisely decipher the brain's thoughts, we should be able to copy it and utilise it as content. for example i know that i can 'think music'. i mean in my thoughts i can listen to music. i mean music/sounds/voices/audio is something that can be imagined. similarly, images and videos and animations too can be thought.

with technology to capture it live and save it as files, tremendous workload can be reduced. an artist can simply think of a video, or a special effect, and have his brain's thoughts captured live and saved as a ready-to-use video file. a musician can think of a sequence of notes in a particular instrument and have that audio captured.

i believe this is possible. our thoughts are a form of information, and it should be possible to capture them, just like we capture a scene using a digital camera. but certainly, this requires deep understanding of brain's working, and its signals, along with the technology to capture it.

with above technology available, it should also be possible to copy/edit existing information in human brain. it should be possible to copy the data already present in brain, and to artificially introduce new, or modify existing data. things similar to what is done on a computer. this can make truckloads of important information quickly available for analysis or use. also, there should be a decrease in need of traditional input devices like pens, mouse, keyboard etc. a poet should be able to think the words of a poem and have them automatically captured and saved in form of a text file. we should be able to think something, and cause it to be treated as a command/task by a computer.

and of course, this system will not develop its own content, and will not assist in its development. it will, however, capture whats going on.

i hope to see spiderman 20 built using this technology.

Wednesday, April 18, 2007

my maniac mind's maze

if i remember that when i was a child, i was saved from a seemingly fatal accident by an adult, and if i go back in time and save my childhood version from that fatal accident, and if this child who subsequently evolves into an adult does the same (and so on), does this situation qualify as an ontological paradox?

i think this is the smallest blog post i have ever written.

i think the above line is incorrect, because addition of above line probably caused this post to no longer remain smallest.

i think the above line is correct, and its intriguing too.

i think the above line is correct.

i think the above line is correct.

i think the above line is correct.

i think the above line is correct.
.
.
.
.
.
.
.
.
.
i think my mind is maniac right now, so i should sleep.

update- 18-4-07

just now i realised that sentences of the type "i think..." cannot possibly be correct or incorrect. they are all a description of what one anticipates/hopes/thinks. and hopes are not correct or incorrect. they can be stated honestly or dishonestly. but they being correct or incorrect is absurd. so i believe that if i write 3 sentences

i think that i am a good human.

i think the above line is correct.

i think the above line is correct.

then the first line may be an honest description of a thought in mind, while the second line is invalid. so is third.

i think the above is correct. honest.

finally, i think i am being a real maniac right now. honest. again.

update 2- 18-4-07

i read detailed meaning of the word 'correct' and one of the meanings is 'something which confirms to truth or fact'. this reads same as one meaning of honest, the meaning i used above. it means use of the word 'correct' is okay.

once again.

i think i am a good human.

i think the above line is correct.

i think the above line is correct.

the first line is honest. if i want to say that it (i.e. the first line) is 'correct', then in my opinion the necessary condition for this is that i be actually thinking that i am a good human, and not (necessarily) my being a good human in reality. the thought that i am a good human must have taken place, although i may not actually be a good human. now is this analysis correct? i think yes it is correct. the immediately previous line is correct (i do think that the previous line is correct- but it does not necessarily mean that it is really correct- only i actually think it is correct).




i am a maniac.

the previous line is correct.

version 3
18-4-07

Saturday, April 14, 2007

witty analogy to abstractly explain difference between various linux distros

today my friend came to me asking which linux distro i have with me. i replied. then he asked me whether there shall be any differences between shell operations of various distros. i told him that for learning purposes, all major/current linux distros behave almost identically. then he asked me what exactly is the difference between various distros like ubuntu, red hat, fedora etc. i wondered for a while thinking how to explain this to him, and then spontaneously this analogy came to my mind.

i told him that just like a 'small car' is a thing and like there are various embodiments of it from different vendors- hyundai atos/santro or daewoo matiz or chevrolet spark or hyundai getz or maruti suzuki alto, a linux distribution is an abstract concept, and the various distros by different vendors all are embodiments of it. thus no matter whether it is a hyundai car or a suzuki one, they all will behave identically for learning purposes, and each one of them is both different in numerous little things [depending upon what a particular company's engineers and designers deem appropriate], and yet identical broadly.

my friend both perfectly understood the difference, and appreciated this analogy. i dont know from where this nice analogy came to my mind, and am writing this here because i think this can nicely solve the same query of any other fellow too.

Saturday, April 07, 2007

a botch that i could have caused because of T9 predictive text messaging

i am an avid user of T9 predictive text messaging. i find the concept simply fabulous. some days back i was thinking about it, when i concluded some things about it. essentially, T9 is a function with one to many mapping, with an added condition such that for every element in the domain, the corresponding element or elements in the range are arranged in a fixed order. and this order is preset worldwide, based on analysis of frequency of usage of words in messages.

one thing that we can control while designing a system like T9 is the order in which different words with same combination appear for that key combination presses, but something we cannot change (fundamentally) is the unique key combination for a word (the key combinations are elements of domain set, while words are elements of range set). and it is because of this limitation (and by sheer coincidence too) that there are certain key combinations which result in mutually-conflicting words, and here lies the potential for problem.

what happened with me was that i got a message "what did you do about sony ericsson k750i?". what my friend was asking was whether i selected the phone or i rejected it (i was in the process of finding a suitable new phone for purchase). since i was in a hurry that time, i quickly hit the reply button and pressed keys in this order 7--> 3 --> 5 --> 3 --> 2 --> 8--> 3 --> 3. the word by default that comes for this combination is 'selected'. in reality i intended to write 'rejected', which, incidentally (and unfortunately) has same key combination as 'selected'. i had almost hit the send button when my finger spontaneously went numb (maybe my reaction time has lowered down) and the message did not get sent. i pressed the 'next word' key and then sent it. considering the situation, its certainly not a big deal even if i had sent 'selected' instead of 'rejected', since neither it mattered, nor it was non-amendable. yet it surprised me so much that here i am writing a blog about it. and i almost murmured to myself "oh my! what a coincidence!".

i believe such 'delicate' key combinations need to be found out by brute force coupled with dictionary lookup of valid words (so as to reject invalid words coming out of brute force) (with included logic to group words with similar key combination) and coupled with manual inspection to detect potential conflicts. conversly, a dictionary with all valid words coupled with a valid-word-to-T9-key-combination mapping program (again, with included logic to group words with similar key combination) and coupled with manual inspection too can serve this purpose. and for whichever key combinations there is potential of troublesome meaning, when a user forgets to change word for a key combination, there the phone should highlight the potentially-troublesome word(s) on the display as a reminder to user to have a re-look and make sure he has chosen the right word(s).

i really believe this should be done. not a necessity, but certainly a nice addition, and sometimes helpful too. especially when my beloved T9 can covert a casual sounding "hey i heard you have got many movies" to an inflammatory "hey i heard you have got many mother". creativity with such combinations has no end ;-)

Friday, March 30, 2007

a puzzle about minimum number of weights required, and how i by chance found its solution

it was a very boring class going on, and so my friend sitting next to me on the desk gave me a puzzle. what minimum number of weights does one need to be able to make measurements upto 100 kg [assuming a least count of 1 kg], and tell those weights too. i have heard/read this problem before, but i had not read or thought about its solution.

my strategy, which i by chance got somehow, was this.

-to measure any weight, we need to build it up
-to build up a weight, there are 3 methods
  1. a direct weight of its measurement. example to measure 2 kg, we have a weight that weighs 2 kg
  2. additive weight- i.e. a weight produced by adding 2 or more available direct weights. example to measure 2 kg, i add 2 available 1 kg weights
  3. subtractive weight- i.e. a weight produced by subtracting 2 [or more- in some combinations] available direct weights. example to produce 2 kg, subtract 3 from 5.
having formed this strategy, the simple task was now to implement it on the whole length of integers from 1 to 100.

here's the digest

-for 1 kg- 1st option is mandatory, unless we want to deliberately use a less efficient combination. hence 1 kg weight is mandatory

-for 2 kg, although all 3 methods can be used, subtractive approach looks like the most efficient, since for subtraction we would necessarily have to have a highest possible weight [from which an already present direct weight- 1 kg in this case- would be subtracted]. not only would this highest weight itself serve us till its own magnitude, but by adding it to already available weight rather than subtracting that from it, we would be able to achieve higher weights. a simple way to calculate the magnitude of this subtractive weight is to add the sum total of already available weights to the value we want to form. example 1 kg is available as direct weight. we want to form 2 kg value. so add 1 [already available] to 2 [value we want to form] and thus we need 3 kg weight. with 1 and 3, not only can we form 2 by subtraction, but we can also form 4 by addition. had we chosen to have 2 kg direct weight so as to be able to form 3 kg by addition, we would not have been able to form 4 kg, and also there would have been a redundant combination- subtracting 1 from 2 to achieve 1- which we already have in form of 1 kg direct weight. with 1 and 3 there is no redundant combination. and hence it is most efficient. so far till 4 kg.

-for 5, again we need to add 1+3+5= 9. we can achieve 5,6,7,8,9,10,11.....13 with this

-14- add 1+3+9+14= 27. values up to 40 possible.

-41- we need 81 as per above algorithm.

-with 81 available, we can calculate any value till 121. so there are 21 redundant values here.

i will someday try to find out a way to calculate precisely till 100, and not till 121, since any unnecessarily heavy weight will cause unnecessary cost increase- since metal is obviously used in creating these weights. anyways, i am happy that a nice strategy by chance came to my mind the moment he put his puzzle.

how i think one can cool a room in a college's hostel with no cooler or air conditioner

its getting hotter in india, and especially here in delhi. i am on the 3rd floor of my hostel, and naturally its hotter here than it would be at the ground floor.

my room is approximately 6 meter x 6 meter x 8 meter [l x b/w x h]. only a ceiling fan on top and a fluorescent tube light. the ambient temperature here should be about 34 degree celsius [i have no thermometer with me here, but to find out the ambient temperature, i put my laptop to standby mode, and then after 2 hours- to give it time to cool to ambient temperature- i resumed it from standby mode. immediately i checked the hard disk temperature in hddlife program, which showed me 35 degree celsius. i compensated for slight increase in temperature during resuming by 1 degree, and so 34 should be a very accurate calculation. here in india we call such a method an intelligent jugaad :-)].

anyways, this is uncomfortable. my friend too cribbed about it, and i suggested him something that came to my mind all of a sudden. why not fill a bucket with cold water [from the water cooler in my hostel]. the water's temperature should be at most 13-14 degrees celsius, and thus there is a good difference of about 20 degrees between water and room air.

put this big bucket in room, and let its water absorb heat. water has high specific heat, and according to newton's law of cooling, the temperature difference will ensure rapid heat absorption. i dont know whether this method will indeed be effective in making the room temperature just comfortable, or less uncomfortable at least. i think it should work, after all whatever heat is absorbed by water, its obviously reducing temperature of room. but any reduction in room temperature will also cause heat from outside room to flow in. nevertheless, there should be some benefit of this no-cost method.

if it doesnt work satisfactorily, i simply gonna take that bucket, turn it around over my head and drench myself with cold water to mar the heat.

Monday, March 26, 2007

windows xp's management of thumbs.db has a potential privacy threat

i found this out of curiosity. actually a thought came to my mind minutes back that does windows explorer delete thumbnails of those images which have been deleted from a folder, from thumbs.db too? it turns out that it doesn't.

i put 10 images in a folder

then i switched the folder to thumbnails view and this generated thumbs.db

its sha-1 hash is- E7CB4A8AA45C9CF0E64720AC670EEBFBE34FA6E4

then i removed 3 images from the folder and refreshed and re-entered the folder but thumbs.db's sha-1 hash continued to be- E7CB4A8AA45C9CF0E64720AC670EEBFBE34FA6E4

this confirms that windows xp does not delete redundant data in thumbs.db

i think this is not good engineering. a company which so vehemently emphasizes on its product being secure and responsible when it comes to privacy shouldn't take this issue lightly. its easy for image viewers like irfanview, xnview, or farstone viewer to implement reading thumbs.db and/or extracting thumbnails from it. remnants of deleted images can cause situations akin to those caused by browser's history, or by files that have been deleted, but shortcuts to whom remain in the 'recent' folder [in this case, the filename and file type is the only available data, along with date and time of the shortcut of course].

one good thing windows explorer does is to update thumbnails of those images which have same name as some previously present image in the folder, which had same filename as a newly added image. but in this case too, as i found out, if i modify the image's size and modification date time etc to match that of old image [assuming dimensions etc are all identical], windows explorer does not update the thumbnail. it is not programmed to check if a newly added image with filename and date/time same as a previously added image is different from it. this is not needed too since the probability of a newly added image being of same size and with same specifications and same size/date etc is zero for all practical purposes.

Wednesday, March 14, 2007

The magnitude of a Gigabyte (Gibibyte actually)

Today a friend of mine told me that good quality blank DVDs are now available for 11 Rupees (approximately 0.25 US $- down from about 15 rupees). I just smiled in surprise. Approximately 4.4 GiB of digital information can be stored for just 11 rupees now. That is about 4 Paisa per MiB. An intriguing fact is that standard floppy disks with 1.44 MB capacity are still available for 10 rupees.

As a reminder to myself of the huge amount of information that 1 GiB represents, I thought this. Consider a man with life span of 80 years. The number of seconds in his life are

80*365*24*3600 + 20*24*3600= 2522880000 + 1728000= 2524608000

1 GiB has 1073741824 bytes.

So if this man were to dictate the decoded (assume binary to decimal) form of each byte at the rate of 1 byte per second, he would manage to dictate only about 2.35 GiB in his entire lifetime, assuming he never sleeps, never eats and just keeps dictating 1 byte per second. And that is just about half of a DVD. An 11 rupee DVD.

Wednesday, February 28, 2007

percentage is not always the right indicator of magnitude of change of something


when i read this news, the one thing going on in my mind since many days finally got confirmed- that percentage is not always the right way of expressing change of a quantity. example this news makes one feel that reliance has reduced the rates to a lesser extent than others, whereas the reality is that they their old rates were already lower than others, and their new rates are same as rivals. mathematically, the change expressed may be correct. but from a marketing perspective, its incorrect to say that reliance reduced their rates to a lesser extent than rivals [thus implying that their final rates are higher than rivals- because initial rates of all companies were not told in the first place]. this makes me feel that if changes are to be told in percentage form, then initial values of the quantities involved must be mentioned. otherwise the inference one can draw from those percentages [or in simpler words the mental impact on reader] may not be correct, and can convey a wrong message. if i sell just 1 copy of my software today and sell 2 tomorrow, is it right for a news story to say that i achieved 100% growth? it must mention absolute numbers too. the funny thing is that this news story does mention actual numbers [although it may be seeming that i am writing something 'against' this story]. what i am writing is about many stories -some of them in respected publications like the new york times or cnet news.com- which talk only in percentages.

that said, percentage is a good indicator in many situations. example is a company sells 1 million copies of a software per month, and sells 1.1 million in a particular month, then its good to say that there was 10% growth in sales [although merely telling 10% growth still leaves something to be desired- and again this is the intial numbers]. in this case, user's inference [or impact] will not be different from what it should be. there wont be any incorrect inferences drawn unintentionally by anyone.

basically, when reporting some change, our ultimate aim shouldn't be to just mathematically

in general, i have found percentage to be good indicator when numbers involved are large compared to magnitude of change. this looks like manifestation of real world- where 'sudden huge growth or sudden huge declines' either do not happen, or happen rarely. and generally only slow growth or decline happens, which is nicely indicated by percentages.

i think one can decide on a case-by-case basis as to what to use

Sunday, February 18, 2007

time to nitpick on gmail:-)



ok i know what i'm gonna write sounds really stupid but i observed this and so will write it. gmail homepage [this one i captured on 18-2-07 near 1 am] says 'over xyz' megabytes. now i assume that the number shown shows currently available space. in this sense, calling it 'over xyz' is incorrect, because its not over xyz megabytes, but equal to xyz megabytes at that instant of time. little mistake. anything wrong with this? (or with me...)

firefox's 'session restore' feature can be annoying or a privacy threat

this is what i discovered last year in my college's library where there was a problem with the ups system. it would keep shutting down all of a sudden, turning off all computers instantly. i was using minefield nightly builds then, and current firefox 2.0.0.1 implements session restore in same manner, so my experience applies to it. actually what i observed was that when firefox restores a user's session, it also logs you back into any email/blog etc that you were logged into before the crash. while this is certainly very good thing for personal/private computers, it caused me problem back then. why? i was usually logged into my email account, and had to wait till power was restored so that before leaving, i could ensure that i power up the computer, cause minefield/firefox to restore my session, and then log-out and clear the cache. i couldn't leave the place simply when computers would get turned off as if i didnt wait and log-out, someone else later would be taken straight to my email account by minefield/firefox. this made me feel that this feature is not suitable for public computers. imagine, a user using firefox at public terminal is logged into his email account, and by chance firefox crashes. the annoyed and pissed-off user leaves things as it is [either assuming that like in internet explorer, browser crash means one has to re-login, or 'not knowing at all about existence of some feature like session restore']. this wont do him good. this issue needs to be solved.


related official webpage
http://wiki.mozilla.org/Session_Restore

Saturday, February 17, 2007

an interesting physics/mathematics problem on distances, speeds and time (and my solution to it)

minutes back when i logged in to my gmail account, i saw a very interesting problem sent to all his orkut friends by my junior cousin. i couldn't resist the temptation to get my hands on it. since i could not find a piece of paper nearby, i opened microsoft paint and tried solving it there (very stupid thing). i hurriedly declared victory, having thought that i had solved it, and mailed the png format image to my cousin, only to be struck by a thought that made me feel my solution was incorrect. i hastily replied again to him that previous solution is incorrect. and i set out to solve it again, this time finding a piece of torn paper. i solved it, and will soon tell him the solution too. here are the resources

below is part of mail/puzzle he sent to me
this is a real brain teaser..
WARNING:the answer may seem logically correct as the figures here have been arbitarilly chosen..

ok..so there's this girl who reaches the station at 6am everyday.Her driver comes and picks her up and they reach home by 7am.
Now,one day the girl reaches the station early..at 5am and decides to walk towards her house.She meets the driver on the way and that day they reach home by 6.40 am..
The ques is..for how long was the girl walking??..and NO the answer is NOT 20 mins....
Best of luck folks!



[red][b]please inform me the ans and the way u found it[/red][/b]

and here is original incorrect solution i sent to my cousin



here is final (hopefully correct) solution
and here is the logic
  • let distance from station to home be x km
  • since driver picks up girl at six (obviously he would travel from home such that he picks her immediately) and reaches at seven, x km traveled in 1 hr. driver speed= x km/hr
  • since it takes him 1 hr from station to home, it takes him 1 hr from home to station. so he starts from home at 5 am
  • girl too started at 5 am that day. let distance [from station] at which they meet be z km
  • let girl speed be y km/hr
  • girl travels z km at y km/hr, so time = z/y hr
  • driver travels x-z km at x km/hr, so time= (x-z)/x
  • equate time by girl to time by driver to get first equation
  • since girl started at 5 and reached at 6:40, total time [girl on foot+ girl in car] = 1 hour 40 minutes. 2nd equation obtained
  • since driver started at 5 and reached home at 6:40, his journey to meeting point took half of 1 hour 40 minutes i.e. 50 minutes. his distance till meeting place = x-z and time= 50 minutes and his speed as already established is x km/hr. 3rd equation obtained. 3 variables 3 equations the problem is solved!

Tuesday, February 13, 2007

I made a high score in Rocket Mania Deluxe :-)

I am happy today, because this is just 5th or 6th time I'm playing this game. Each time I had played it previously, I was casual. This was the first time I played it with seriousness and it paid off nicely. I played a 'Strategy' type round this time at Hard level- the topmost level available in this game.

Here are some of the screenshots

This one below is when I launched 6 rockets at once...

Finally, this one shows how I build complex paths to maximize the number of simultaneous rocket fires while including all the matches present on board...

Sunday, February 11, 2007

Which game consoles would I buy If I had USD 500-600, and why not to buy a PlayStation 3

ALSO SEE: A related post by me on Facebook [link].

Currently, the Sony PlayStation 3 starts at about $500 for the standard version, and goes up to $600 for the premium version. Although many people are buying the $600 PS3, I believe that they could have got much more value for their 600 dollars. If I were to spend $600 on video game consoles, what I would not have done is to buy a PS3, and what I would have done is:
  • Buy a Sony PlayStation 2 for $125, plus a Nintendo Wii [$250], as well as a Sony PSP [$200], for a total of $575. With the remaining $25, I would have bought a Wii classic controller [$20]. I would have spent the remaining 5 dollars feeding myself with a burger and a coke - celebrating a wiser buy! It is not hard to observe that this deal provides significantly higher value for money [VFM] than a PS3 alone. Specifically, this deal allows one to use the huge library of now-cheap PS2 games, the fun-filled modern gaming of Wii, and the portable entertainment and gaming experiences of the PSP. This deal is also more suitable for those households which have more than one kid. However, this package lacks the ultra-advanced graphics of PS3 games.
  • An alternative deal, suitable for those who don't want to sacrifice graphics quality is - an Xbox 360 Core [X360] for $300, a Wii [$250], 2 Wii classic controllers [$40], and double the amount of burgers and coke... This package provides two next-generation game consoles, without sacrificing on graphics quality. Undoubtedly, this deal is better than the $600 PS3.
  • Yet another alternative is to go for the X360 Core [$300], the PS2 [$125] and accessories/games worth $175. Once again, this package provides more VFM than the $600 PS3.
  • Yet another alternative is to buy the standard X360 [$400], a PS2 [$125] and accessories worth $75. Again, a better buy than the $600 PS3.
PS3's VFM increases if one goes for the $500 version. It's more difficult to produce an alternative package that provides as much VFM as the $500 PS3. Some individuals may find more value in the X360 Core + PS2 + accessories [total $500] package, while others may find more value in Wii + PS2 + accessories + some games [total $500] package.

Of course, one can work out a similar alternative deal for the $400 X360 - a $250 Wii + $125 PS2 + a Wii classic controller.

The whole idea behind these alternative deals is to maximize the VFM while minimizing loss of functionality. One must also remember that these alternative deals get the owner multiple devices, providing redundancy [decreased usage of each device will increase their life; since there are multiple devices, more people can use them simultaneously].

Two things to note here:
  1. The PS2 can play DVDs, as well as games from the first generation PlayStation [PS1].
  2. The Wii can play NES, SNES, Sega Genesis, Turbo-Grafx 16, Nintendo 64 and GameCube games, in addition to Wii games. It is 7 game consoles in one body.
As a side note, it must be remembered that the PS3 can play all PS3, PS2 and PS1 games. One must carefully weigh one's gaming requirements before choosing a particular package. It must also be remembered that the PS3 includes an inbuilt Blu-ray player, which may be provide more value to select individuals.

Update (April 2009): It is now possible to buy an Xbox 360 Pro ($299) along with a PS2 ($99) for the price of one PS3 ($399 - and this PS3 does not even have backward compatibility with PS2 games). It is also possible to buy an Xbox 360 Arcade system along with a Blu-ray player for $399 total, a more sensible deal in my opinion than a PS3.

Update [Jul'16]: Following article is directly related to this idea - "Why The US Navy Should Build Smaller Aircraft Carriers".

Update [Mar'21]: Other things I could have spent that $1,000 on while waiting to order an iPhone X at 3am [link].

Update [May'21]: One of my relatives bought a Kit Seltos HTX and also got accessories installed from the showroom for a total of about INR 15.5 lakh [current Seltos prices are about 50K higher than when he bought]. This is his only car and he really needs two cars. He could've bought two cars while spending the same amount of money - a Skoda Rapid and a Nissan Magnite. I don't know about others, but to me this dual car deal looks much, much better than the single Seltos.




***

Update [Jul'21]: The general idea of choosing more "value" applies even to picking a single product. It isn't just about splitting a costlier product into two, higher-value-received products. For example, there's no comparison between the MG Hector / Hector Plus and the Skoda Kushaq - size, grandeur, body weight, road presence, solidity, etc. The screenshot below shows that only a fool would buy the Kushaq instead of Hector / Hector Plus.


***

Update [Jul'21]: We booked Hector Plus 7-seater diesel Super variant for approximately INR 18.5 lakh rupees on-road price. Did we violate my own principles and ideas outlined in this post? Likely yes. Could we have bought two good cars for this price instead of one great car - theoretically yes, practically no, since the reason why we bought the Hector Plus was specifically related to the current situation and requirement.





***

Friday, February 09, 2007

Yes! Yes! Yes! I Cleared Number Madness :-)


I'm all smiles. Especially because this is the first time ever I played this game :-)
Although the large number - 260 - is a little worrying...

P.S. This is an extension for Mozilla Firefox, and can be download at the official AMO website

My Photography On Flickr

Monday, January 22, 2007

What To Do When The English Language Has Some Seemingly Unsolvable Problems

I was reading about an online magazine, when I came across a statement "...is published bimonthly...". I got confused - because I knew that bimonthly means two things - "twice a month" and "once every two months". How is a reader expected to guess, which particular meaning of bimonthly is being used (especially when both the meanings are equally probably in this situation).

It is fair to say that the use of this word will almost always create confusion, and so a standard rule needs to be created to resolve this issue. One solution is to always specify the intended meaning in braces

For example

...bimonthly (once every 2 months)...

OR

...bimonthly (twice every month)...

That is, if instead of simply writing bimonthly, we always start writing bimonthly (once every 2 months) OR bimonthly (twice every month), then there shall be no confusion. Of course, one could then argue that why use the word at all? Why not just write once every 2 months OR twice every month. This is a reasonable and sensible argument.

My Questions On Yahoo Answers

Saturday, January 20, 2007

Apple's Efficient And Sensible Strategy - Extend Third-Party Technologies

I observed this interesting strategy of Apple, one that appears to be working for them. On one hand is Microsoft, which undertakes (costly and slow) original research to develop technologies from the ground up (such as WMA, WMV, the Windows kernel, ActiveX, and so on), and on the other hand is Apple - which chooses an appropriate open source or open standard technology, optimizes and extends it, and fully embraces it. This strategy allows Apple to equal or surpass Microsoft while making smaller investments in technology.

A few examples

Microsoft - Trident (MSHTML), Direct3D, Windows Kernel, WMV, WMA, ActiveX

Apple - WebKit (KHTML), OpenGL, FreeBSD, H.264, AAC, Java


Apple's strategy is good for the world too - instead of spending millions of dollars in creating new and different solutions for the same problem, Apple picks the 'good enough' existing solutions to those problems and improves them. Microsoft, on the other hand, prefers reinventing the wheel.

My Detailed Profile On Google

Thursday, January 11, 2007

Convenient Way To Keep Using Trialware Software Repeatedly

This idea came to my mind circa June 2005, and I don't know whether other people already use this trick.

Software companies typically release demo or trial versions of their programs that work for a limited number of days and then either stop working, or disable many features - requiring the user to purchase the software in order to access its full functionality.

If one could quickly reinstall Windows somehow (a fresh install), one would be able to repeatedly install the trial software and use it in trial mode (in which it provides full functionality).

Create a c: partition that is less than 4475 MB, and install Windows on it. Next install other useful programs that you use. Now install the trial programs that you want to use infinitely. Finally, make a 'disk image' of c: partition on a DVD, using a program such as Symantec's Norton Ghost. Every 30 days (assuming each of the trial programs expires after 30 days), simply restore back the c: partition from the DVD - the trial programs shall never realise that they are being used over and over again.


Remember to never keep any user data on c: else it will be wiped out during the restore operation. The restore operation should take less than 30 minutes, and 30 minutes every 30 days is definitely worth this benefit.

About Me - My Profile On LinkedIn