akshayxyz Posted June 27, 2012 Posted June 27, 2012 Now, I can't stop laughing... None other than, D.E.Knuth..one of the Bhagvans of CS...started as Physics Major Knuth had a difficult time choosing physics over music as his major at Case Institute of Technology (now part of Case Western Reserve University). He also joined Beta Nu Chapter of the Theta Chi fraternity. While studying physics at the Case Institute of Technology, Knuth was introduced to the IBM 650, one of the early mainframes. After reading the computer's manual, Knuth decided to rewrite the assembly and compiler code for the machine used in his school, because he believed he could do it better.[6] In 1958, Knuth constructed a program based on the value of each player that could help his school basketball team win the league. This was so novel a proposition at the time that it got picked up and published by Newsweek and also covered by Walter Cronkite on the CBS Evening News.[6] Knuth was one of the founding editors of the Engineering and Science Review which won a national award as best technical magazine in 1959.[7] He then switched from physics to mathematics, and in 1960 he received his bachelor of science degree, simultaneously receiving his master of science degree by a special award of the faculty who considered his work outstanding.[6] So, before switching to CS/Maths, he was able to better a program in assembly code (while he was studying Physics)....something must be wrong during that time. Ab bas chemsistry/bio ka example mil jaaye...
akshayxyz Posted June 27, 2012 Posted June 27, 2012 :hysterical: Even Djikstras started with Theoretical Physics..Looks like there is whole bunch of them, conspiring to refute mcenley's life's work on CS-UG requirements.
akshayxyz Posted June 27, 2012 Posted June 27, 2012 OK.. it did not take too long.... Here is the last nail in the coffin... One of the Turing award winner, made the mistake of doing Bachelors in Chemistry (not even Chem Engineering) http://en.wikipedia.org/wiki/Alan_Perlis :hysterical::hysterical::hysterical::hysterical:
akshayxyz Posted June 27, 2012 Posted June 27, 2012 Two more interesting entries in Turing awardees list. http://en.wikipedia.org/wiki/Herbert_A._Simon (BS in Political Science) http://en.wikipedia.org/wiki/Robert_Floyd (BS in liberal Arts and Physics) Floyd became a staff member of the Armour Research Foundation (now IIT Research Institute) at Illinois Institute of Technology in the 1950s. Becoming a computer operator in the early 1960s, he began publishing many noteworthy papers and was appointed an associate professor at Carnegie Mellon University by the time he was 27 and became a full professor at Stanford University six years later. He obtained this position without a Ph.D.
Crookbond Posted June 28, 2012 Posted June 28, 2012 Akshay - I hope you realize with the Turing Award winners you have posted didn't have Computer Science as a department. Donald Knuth and John Hopcroft (whom I met and spoke with recently) have a very interesting "modern" take on how CS should be taught. Anyways, if you want to discuss this seriously I can make the effort or you can be "hysterical" and spam with links (which are very rare examples) as much as you want and believe that you have won this argument. Cheers!
seedhi Posted June 28, 2012 Posted June 28, 2012 Akshay - I hope you realize with the Turing Award winners you have posted didn't have Computer Science as a department.Firstly, all of them had a Math department, yet several did Physics, Chemistry or Humanities majors. Donald Knuth and John Hopcroft (whom I met and spoke with recently) have a very interesting "modern" take on how CS should be taught. This is not about how CS should be taught, rather how to evaluate the prospective CS undergrad. Expertise in Math, Physics and Chemistry is very natural way to do so.This is especially so in the Indian context where poor students who have no access to internet or computers in high school should not be placed at an obvious disadvantage; testing on the basis of language would be unfair to students in the rural areas of India. Math, Physics, Chemistry is as close as one can get to a level playing field in India while testing a broad class of analytical skills. Anyways, if you want to discuss this seriously I can make the effort or you can be "hysterical" and spam with links (which are very rare examples) as much as you want and believe that you have won this argument. Cheers!Atleast he is posting links to support his argument, which you dont seem to do. Let me add that there are many, many excellent CS researchers who have emerged from the Indian higher education system which has placed emphasis on Math, Physics and Chemistry at the entry level. More broadly, the issue is whether you can compartmentalize scientific education, and I would say no. Computer Science in its different specializations has diffusion from several other disciplines, for example Vision and Graphics (physics), bioinformatics (biology, medicine), social computing (Humanities), Algorithmic Game Theory (economics) etc.
Crookbond Posted June 28, 2012 Posted June 28, 2012 I see how you have jumped in with "both feet" now. [1] I never said Math wasn't a pre-requisite. I bet you wouldn't find any post which talk about Math not being important. [2] Hopcroft is one of the first CS scientists and a PhD in EE. Not only does he say one doesn't require a physics/chemistry background, he even goes on to challenge the notion of advanced mathematics as a pre-requisite for CS. Hopcroft said - "CS has evolved and so should the curriculum and the prerequisites. Let us not be afraid to challenge notions we have held across decades. I have spent most of my life proving that Computers can *do* things. It's time to stop and show the things Computers can *do*. You on't require math or physics or chemistry or electrical to do this." So, when I said "changing of curriculum" I meant the whole process but of course IIT's would never understand that because all they worry is about a highly selective stupid entrance test. [3] Donald Knuth has some interesting thoughts on math theory for CS, read about it. [4] Posting 3-4 links who had have majored in one and have a successful career would be the most naive argument one can come up with. You can't generalize winners of Turing awards (FFS, CS Nobel prizes?) to a Tom Dick and Harry. How about this? Raj Reddy (India's only Turing Award winner) completed his bachelor in civil engineering and won a Turing award. So what next, pre-requisite and application of "construction" knowledge ? Here's what you are doing - http://en.wikipedia.org/wiki/Hasty_generalization Just because Turing award winners were Bachelors or Masters in Chemistry and Physics is no reason to suggest that Physics and Chemistry are good indicators for a CS degree. As Akshay himself has pointed out "Political Science" and "Arts" winners. So, does it mean you require application of Arts and Political Science for an admission test? This is getting ridiculous now! Also, why adopt just one thing from MIT? If you want to adopt MIT's admission process, adopt it completely. Don't do a selective amnesia. Have a common test consisting of - critical reading, mathematics and writing much like they do. There's a reason why they don't use physics and chemistry you see. This obsession with Physics and Chemistry (and it's applications or whatever) has to stop.
Crookbond Posted June 28, 2012 Posted June 28, 2012 http://www.cmu.edu/admission/forms/2012/apply_freshman12.pdf Looks like CMU also has lost it. Its high time Mcenley publishes his research on "how to select prospectivs CS UG - 101" Standardized test scores add to our knowledge of your ability, but we donÃÕ make decisions simply on the basis of test scores. Your high school record and standardized test scores (all SAT Reasoning Test or ACT with Writing and SAT Subject Tests) work together to make up the academic portion of your evaluation It's high time you read what you post. Just 1 year of Physics is a "must" that too high school physics. It isn't different than what I was proposing earlier .It also requires 2 years of Foreign Languages. The ones highlighted are "optional". So not a pre-requisite in my books. I have many friends doing their bachelors at US universities without taking or requiring even a single physics/chemistry examination. Oh dare I say CMU has worded it as - "preferred high school preparation". Not "mandatory", not "prerequisite" but "preferred". It means students who haven't done some of the above can get through too.
Crookbond Posted June 28, 2012 Posted June 28, 2012 So, in terms of logical problem solving.. I would say Maths~CS~Physics ~~~Chemistry..........Biology/Medicine, even though also logic/facts/rules based science, goes a bit too far from Maths/Physics/CS style of problem solving.. Yea - This is more like it. Now, the only question remains can we test on all of them? NO. Math is sufficient enough to test in my books.
Crookbond Posted June 28, 2012 Posted June 28, 2012 I am not asserting any causation. I am not introducing any fallacy. You are failing to see the abstraction/connection between PCM. As PK also said below. Once again I repeat - when you are selecting candidates for Bachelors level studies - lets say for CS. They are not testing your experience, extent of knowledge in CS/Maths. PCM ( and all forms of sciences) has this common abstraction layer - you have given set of established facts, rules and language/symbols/grammar - you have to work within them to explore new stuff, solve problems - using logic, backtracking, perseverance. Physics is almost same as Maths, and Chemistry is also not too different. I would go as far as saying that - a XII level student who has done something significant at international level in Chemistry - or lets say is capable of solving chemistry problems beyond his age (lets say Bachelors Final year) - will do well in CS, if he chooses to. And even stronger case for someone talented in Physics. The bold part is the most awesome thing. I am archiving this post. :hysterical:
seedhi Posted June 28, 2012 Posted June 28, 2012 Also' date=' why adopt just one thing from MIT? If you want to adopt MIT's admission process, adopt it completely. Don't do a selective amnesia. Have a common test consisting of - critical reading, mathematics and writing much like they do. There's a reason why they don't use physics and chemistry you see. This obsession with Physics and Chemistry (and it's applications or whatever) has to stop.[/quote']I have no idea in which world are you living in. Certainly not India as you are simply not aware of the constraints in India. 1. India, unlike the US, has a severe shortage of college seats. Therefore, for any institute there has to be a common exam for all engineering/technology disciplines, otherwise each high school student will have to apply to every discipline. And it will be a massive waste of resource to have separate exams for each discipline. 2. Secondly, at the undergrad entry level, there cannot be a fuzzy subjective evaluation like in Harvard or MIT since this will no doubt lead to corruption/nepotism. This is doubly important since India has a huge shortage, unlike the US. There has to be a standard test which eliminates prejudices. 3. Regarding the actual entrance test, anything other than basic sciences (Math, Physics, Chemistry) is terribly unfair to students from rural areas who are lacking in language/communication skills. Conclusion: Since there has to be a common test for every engineering discipline, Math-Physics-Chemistry is the obvious choice for the test. Also, it is the most fair criteria for the students in rural India. By the way, John Hopcroft is probably not aware of the constraints in India - shortage of seats, tendency of corruption, rural disadvantage etc, so lets not hype his views in the Indian context. The existing entrance system in India has managed to produce many many excellent CS graduates who did excellent research abroad. The entrance system for undergrads is more or less in good shape. What is lacking is a graduate research environment which needs to be fixed.
Crookbond Posted June 28, 2012 Posted June 28, 2012 I have no idea in which world are you living in. Certainly not India as you are simply not aware of the constraints in India. 1. India, unlike the US, has a severe shortage of college seats. Therefore, for any institute there has to be a common exam for all engineering/technology disciplines, otherwise each high school student will have to apply to every discipline. And it will be a massive waste of resource to have separate exams for each discipline. 2. Secondly, at the undergrad entry level, there cannot be a fuzzy subjective evaluation like in Harvard or MIT since this will no doubt lead to corruption/nepotism. This is doubly important since India has a huge shortage, unlike the US. There has to be a standard test which eliminates prejudices. 3. Regarding the actual entrance test, anything other than basic sciences (Math, Physics, Chemistry) is terribly unfair to students from rural areas who are lacking in language/communication skills. Conclusion: Since there has to be a common test for every engineering discipline, Math-Physics-Chemistry is the obvious choice for the test. Also, it is the most fair criteria for the students in rural India. By the way, John Hopcroft is probably not aware of the constraints in India - shortage of seats, tendency of corruption, rural disadvantage etc, so lets not hype his views in the Indian context. The existing entrance system in India has managed to produce many many excellent CS graduates who did excellent research abroad. The entrance system for undergrads is more or less in good shape. What is lacking is a graduate research environment which needs to be fixed. WRONG - Even under the given constraints we can have a better choice. There's no need to fixate on Physics/Chemistry as there's no correlation. The reason the IITs get good students is due to extremely high selectivity so you might as well drop some acid. Math is enough and we need to be more creative in the entrance examination. An exam like SAT conducted by an independent (preferably non-profit organization) would be the right step forward. The system for UG's is not the best in shape. The fact that Indian students are performing well in research abroad is testimony of the fact of the environments in countries like US. It also shows that to be the best in the world you can't just be a teaching institute, you need to be a teaching + research institute. The roots for research like at the ground level and hence, it important to take the right kind of people in an exam. FWIW, I never cited links and swore by them you did. Now, I counter your argument you go back to saying those links are not for India but for US. Why post or even argue based on those links and then say - "These links were just for reference.". Also, please note when speaking on any subject you need to have an eye for someone who understands and been in the system versus someone who hasn't. Just clearing an highly selective entrance exam doesn't mean you are anyone has earned any bragging rights in the world.
seedhi Posted June 28, 2012 Posted June 28, 2012 WRONG - Even under the given constraints we can have a better choice. There's no need to fixate on Physics/Chemistry as there's no correlation.Do you agree or not that all engineering streams must have a common entrance test in India? And if this is so then Math-Physics-Chemistry is the obvious choice for the test. Also' date=' clearly there is a correlation, since for the last 50 years the students who did well in the entrance tests have gone on in general to do well in life after undergrad.The reason the IITs get good students is due to extremely high selectivity so you might as well drop some acid. Math is enough and we need to be more creative in the entrance examination. An exam like SAT conducted by an independent (preferably non-profit organization) would be the right step forward.How is this not biased against rural students who have no access to computers/internet and who are clearly lacking in language skills because of lack of good schools ? Math, Physics and Chemistry are the most accessible skills to poorer students. The system for UG's is not the best in shape. You have just admitted that IITs select good students. If this is so, then what is the need for changing the undergrad entrance system?The fact that Indian students are performing well in research abroad is testimony of the fact of the environments in countries like US. It also shows that to be the best in the world you can't just be a teaching institute, you need to be a teaching + research institute. The roots for research like at the ground level and hence, it important to take the right kind of people in an exam.You have just contradicted yourself, if graduates from India are able to do well in research even if in the US, it is also because the undergrad tests in India do select the good students. I agree that the problem in India is in creating the right research environment but that is not a problem at the undergrad entry test level. FWIW, I never cited links and swore by them you did. Now, I counter your argument you go back to saying those links are not for India but for US. Why post or even argue based on those links and then say - "These links were just for reference.". I did not post those links, I just pointed out that you are making claims about your expertise without any evidence. Also, please note when speaking on any subject you need to have an eye for someone who understands and been in the system versus someone who hasn't. Just clearing an highly selective entrance exam doesn't mean you are anyone has earned any bragging rights in the world.Why dont you let us know what is your expertise, instead of making judgment calls on who is qualified to speak on this subject. I dont know who you are and I am not passing any judgment on your qualifications, but your arguments are flawed. Before anything you need to answer this question: Do the undergrad entrance tests in India select good students in a fair and competitive way? If the answer is Yes, then there is no need to fundamentally alter the entrance tests. If no, then you need to explain why you think so.
asterix Posted June 28, 2012 Posted June 28, 2012 Physics & Chemistry is part of the 1st year/semester syllabus in many Engg institutes/colleges in India.
akshayxyz Posted June 28, 2012 Posted June 28, 2012 Akshay - I hope you realize with the Turing Award winners you have posted didn't have Computer Science as a department. Donald Knuth and John Hopcroft (whom I met and spoke with recently) have a very interesting "modern" take on how CS should be taught. Anyways, if you want to discuss this seriously I can make the effort or you can be "hysterical" and spam with links (which are very rare examples) as much as you want and believe that you have won this argument. Cheers! And who is talking about teaching? Your main concern is/was, P/C can't be a criteria of evaluating eligibility for CS UG. In other words, if someone is very good at Physics/Chemistry - he should not be eligible for choosing CS-UG at IIT/MIT/CMU/...... I just showed you examples of even higher (highest) than UG level. You may argue that they were exceptions.. but I can imagine, someone like you 'not allowing' them pursue CS, since they were not from Maths. (I know very well CS did not exist during those days) Here is a question in simpler English - if there is a bright student - who has done extremely well in Physics/Chemistry - and now chooses to pursue CS at IITs. Would you deny him the place?
akshayxyz Posted June 28, 2012 Posted June 28, 2012 I see how you have jumped in with "both feet" now. [1] I never said Math wasn't a pre-requisite. I bet you wouldn't find any post which talk about Math not being important. [2] Hopcroft is one of the first CS scientists and a PhD in EE. Not only does he say one doesn't require a physics/chemistry background, he even goes on to challenge the notion of advanced mathematics as a pre-requisite for CS. Hopcroft said - "CS has evolved and so should the curriculum and the prerequisites. Let us not be afraid to challenge notions we have held across decades. I have spent most of my life proving that Computers can *do* things. It's time to stop and show the things Computers can *do*. You on't require math or physics or chemistry or electrical to do this." So, when I said "changing of curriculum" I meant the whole process but of course IIT's would never understand that because all they worry is about a highly selective stupid entrance test. [3] Donald Knuth has some interesting thoughts on math theory for CS, read about it. [4] Posting 3-4 links who had have majored in one and have a successful career would be the most naive argument one can come up with. You can't generalize winners of Turing awards (FFS, CS Nobel prizes?) to a Tom Dick and Harry. How about this? Raj Reddy (India's only Turing Award winner) completed his bachelor in civil engineering and won a Turing award. So what next, pre-requisite and application of "construction" knowledge ? Here's what you are doing - http://en.wikipedia.org/wiki/Hasty_generalization Just because Turing award winners were Bachelors or Masters in Chemistry and Physics is no reason to suggest that Physics and Chemistry are good indicators for a CS degree. As Akshay himself has pointed out "Political Science" and "Arts" winners. So, does it mean you require application of Arts and Political Science for an admission test? This is getting ridiculous now! Also, why adopt just one thing from MIT? If you want to adopt MIT's admission process, adopt it completely. Don't do a selective amnesia. Have a common test consisting of - critical reading, mathematics and writing much like they do. There's a reason why they don't use physics and chemistry you see. This obsession with Physics and Chemistry (and it's applications or whatever) has to stop. What a load of tangential nonsense in this post and others following. You are running in circles, but light years away from the key point. No one is arguing that Maths is sufficient to evaluate candidature for CS-UG. No one is saying that one *needs* Chemistry/Physics knowledge in pure CS. All - I have been saying, if one has done reasonably ok in Maths, but extremely well in Physics or even Chemistry, and now want to pursue CS at UG - there is no reason to deny him the place. Ability to solve complex problems in any field of science - be it Maths, Physics, Chemistry - is good enough to judge the ability.
akshayxyz Posted June 28, 2012 Posted June 28, 2012 The bold part is the most awesome thing. I am archiving this post. :hysterical: Its unfortunate to see that you claim to understand Maths and CS, but can't see the abstraction layer I am talking about.
Crookbond Posted June 29, 2012 Posted June 29, 2012 What a load of tangential nonsense in this post and others following. You are running in circles, but light years away from the key point. No one is arguing that Maths is sufficient to evaluate candidature for CS-UG. No one is saying that one *needs* Chemistry/Physics knowledge in pure CS. All - I have been saying, if one has done reasonably ok in Maths, but extremely well in Physics or even Chemistry, and now want to pursue CS at UG - there is no reason to deny him the place. Ability to solve complex problems in any field of science - be it Maths, Physics, Chemistry - is good enough to judge the ability. Dost - Now, look who's running in circles. I didn't argue anything against the highlighted, in fact you have finally come full circle now. My point was - Why fixate on Physics and Chemistry? All I have been saying that if one does reasonably ok in any field of science - includes Biology, Language Sciences etc. you can't discard the fact that the person can't be a good engineer. Physics and Chemistry is not cast and stone. If you want to go "abstract", you can include philosophy too.
Crookbond Posted June 29, 2012 Posted June 29, 2012 Do you agree or not that all engineering streams must have a common entrance test in India? And if this is so then Math-Physics-Chemistry is the obvious choice for the test. Also, clearly there is a correlation, since for the last 50 years the students who did well in the entrance tests have gone on in general to do well in life after undergrad.How is this not biased against rural students who have no access to computers/internet and who are clearly lacking in language skills because of lack of good schools ? Math, Physics and Chemistry are the most accessible skills to poorer students. Hadd chewtiyapa hai. If you select a cream of lesser than 0.5% on any entrance test in the world you are and will get outliers. Now, understand the preserving 'cult' statement I made earlier. You have just admitted that IITs select good students. If this is so, then what is the need for changing the undergrad entrance system?You have just contradicted yourself, if graduates from India are able to do well in research even if in the US, it is also because the undergrad tests in India do select the good students. I agree that the problem in India is in creating the right research environment but that is not a problem at the undergrad entry test level. I did not post those links, I just pointed out that you are making claims about your expertise without any evidence. :facepalm: Sexy! Typical Indian mentality. Wait on your *****, till it burns which plagues the IIT. Here's something for you to think upon - Stanford computer science rethinks core curriculum http://engineering.stanford.edu/news/period-transition-stanford-computer-science-rethinks-core-curriculum Why did Stanford do it? It was already graduating good students. Was there a need, NO?But, clearly they have the vision which people in the IITs lack. Strain the system until it breaks and when it breaks build again to form another system which will break soon. Forget admission tests, tell me the last time IITs did even this. Why dont you let us know what is your expertise, instead of making judgment calls on who is qualified to speak on this subject. I dont know who you are and I am not passing any judgment on your qualifications, but your arguments are flawed. Because if I did - you will pi$$ your pants because you have no understanding of this education business. But, that's not how the way an argument works. You still haven't answered what makes Physics and Chemistry a "natural choice" over other sciences like Biology, Arts and Languages which can also be good indicators of an engineering degree.
seedhi Posted June 29, 2012 Posted June 29, 2012 Hadd chewtiyapa hai. If you select a cream of lesser than 0.5% on any entrance test in the world you are and will get outliers. Now' date=' understand the preserving 'cult' statement I made earlier.[/quote']The reason that IITs get the cream of the students is NOT the entrance exam. It is because they have only a limited number of seats. To increase the student intake the imperative is to set up new institutes in a systematic way, not to change the entrance test. Are we talking of admission test or curriculum? The reason curriculum is updated is to match the changes in technology, and the changing nature of requirements for research, industry etc. An admission test of basic sciences does not require any major overhaul if it is performing well. If you dont want to reveal what your credentials are then what is the point of hyping them up ? Not sure if you have been reading my posts. Let me repeat what I wrote: 1. For a variety of reasons particular to the Indian situation, for any given set of institutes (say IITs), at the undergrad level there has to be a common entrance test for all engineering streams, which makes Math-Physics-Chemistry a very natural choice for test subjects. 2. Math-Physics-Chemistry are the most accessible subjects to poorer students from rural areas. They severely lack the language skills to compete with urban students. Urban, richer students will always have a natural advantage but including language skills will magnify it. At one point of time IITs (pre-1980 I believe) had English exam as well, but they dropped it. 3. Biology is completely ruled out the way our high school system is structured. It is impractical to expect all high school students wanting to become engineers to also study biology.
Recommended Posts