CSEMS Research in Indian Academia

[UlanBatori]

I had a long discussion with a very experienced friend (EyeEyeTee Peechdee, high-up industry position in India, many saal going around whole Duniya as part of Yoo Enn, dealing with militaries.. and now a brofejar MrsDr also eyeeyetea Peechdee and Brofejar) during a trip to the Yoo Ess where I was the chaffeur. So I do have a good bit of inside info.

So the Indian system does give out money! I mean, how many desis come to Yoo Ess for week-long or 2-week conpherenj, staying in rejort hotels like the best in the duniya! The trouble is, it is more Soviet system - you have to APPLY for every trip and have it stamped by Babus, rush around getting this abbroval that abbroval, abbly phor vija to Paco's relatives, look for tickets... run around some more...

IOW, they don't control the budget, everything is microcontrolled by some Babu with no stake in achieving project objectives. Like Soviet system - you spend like money is going out of style, because there is no reward for saving, and no pangs of conscience for spending. OTOH, in my case, even when I have $$ (Cinderalla Phase) every penny spent causes me pain and I squeeze out everything I can get out of it. Always looking for innovation as in "cheap". No way can I afford to fly to and from desh, go stay in 4-star hotels in desh for 2 weeks, flying around conference to conference, all on my project funds - even if I have the budget I would rather spend it on paying a hungry undergrad or squirrelling it away to do something else more useful.

Same for all project purchases - everything has to be Approved in India. So time spent on actual technical work and thinking may be 5%, time spent on bureaucracy is 80%, vs, in my case, 25% and 60%. And I curse that!!!

Maybe Indian system needs to give more autonomy to the profs - meaning, u mess up, its ur problem.

[shiv]

I have a different take on this. Good academic research is SUPPOSED to be done **with no application in sight***, because the best breakthroughs will generate much bigger payoffs in something far down the line that you cannot even imagine now. This is part of the issue: ppl declare that BECAUSE they have no clue of any application, their research is good.

But certainly research in engineering ****IMO!!!!**** must be done with some intention of solving a problem that has not been solved before, or finding a better or different route to the solution. Is it the RIGHT problem to solve, is a matter for much agonizing. You may only have a glimmer of hope that there IS a solution, not how to find it. That's what Peechdee PIGS are for, after all.

The above tries to capture the reasons for friction between the researcher and funding agency and its reviewers. More on that later.

So how does one make the connection between the nation/guvrmand's NEEDS, and the academic researcher's abilities/ideas/skills? There has to be an intermediary, or more than one level of intermediary. In the US, since it was set up due to Da Fire In Da Belly circa WW2, there is the 6.x classification:
6.1: Basic (academic) research. Totally UNCLASSIFIED. Open literature. Success criteria are not Patents or Products, but peer-reviewed publications, & students educated. OK, they like to hear of Invention Disclosures and Technology Transfer, but really it is all about Papers and PhDs.

6.2: Applied. Some done in university academic schools, some in university Applied Labs (full-time researchers), some in industry, Small Business and govt. labs. Since they have to pay at DOO-hours not PIGS-hours, the budget is about 3 times as big as that of a 6.1 project. May also be much shorter term, usually incompatible with PeeChDee.

6.3: Prototype/System development. Usually outside university scope. Industry/govt. lab projects. Budget about 10x of 6.1.

6.4 System field tests/deployment etc. Budgets 1000x 6.1
**************************************************

Research aims and funding is something people have agonized over for decades. In another life I wrote the folllowing piece which I cross post here simply to keep the thread moving

How do people discover things? How do you fund people to "discover" things?

There was a paper in the British Medical Journal over a decade ago that
discussed this topic - and all you freaks on this list reminded me of
that paper.

The author (can't remember his name) classified methods of discovery into 4
categories:

1) Pure chance: - some guy happens to come across 4 instances of an extremely
rare disease that most people don't see even once in a lifetime, and he
works on these and finds out something new.

2) Hard work: - based on the theory that you can't trip over anything
unless you get up and walk. That was how the first cure for syphilis was
discovered - someone tested hundreds of chemicals on patients and the 612th
(or 512th??) compound worked, and it was called Salvarsan.

3) Serendipity: (a word allegedly named after a story about the 3 princes
of Serendip or Sri Lanka - if anyone knows this story I would love to hear it).
Serendipity is the act of noticing a hair sticking out behind a leaf in a
bush, and finding a furry animal in the bush.

The discovery of penicillin was Serendipity. Fleming noticed that a mould
had actually killed bacteria and paid attention to this fact, unlike
generations of researchers who had simply thrown fungus bearing culture
plates away as "contaminated"

4) "Altamirage" - a word coined by the author of the paper. There was this
Spanish nobleman who one day went searching on his estate for a cave into
which his dog had fallen. He found the cave, went in and was surprised to
find that the cave was full of prehistoric tools. Being interested in this
sort of stuff, he would go down the cave with his little daughter and he
would sit there examining making drawings and theories about the original
inhabitants of the cave and tool makers. His daughter did not have much to
hold her attention for long, so she would pick up this and that and one day
she is said to have exclaimed "Look Papa!" - pointing at some paintings on
the walls of the cave that her father had not noticed.

As it turned out, all the theories put forward by this Spanish nobleman
were trashed, but the paintings, discovered by his daughter are still
famous - the cave paintings of Altamira.

The point the author is trying to make is that discoveries are frequently
made by people who play with this and that, never spending too much time on
anything definite. Indeed, many important discoveries are apparently made
by people whose vocations are quite unconnected with the subject of the
discovery. Some of these people are called geniuses, but could be called
loonies under other circumstances.

How do you approve a research grant to a person who does not allow any
single topic to hold his attention for very long? But the evidence seems to
suggest that it is just such characters who make important discoveries.

[Vayutuvan]

Shiv: Yes, that will work for discoveries. But engineering requires enormous attention as well as a little bit - OK a lot - of jack-of-allness as well. There is one article where Stanford president says that today he would want all his researchers to be T people - one deep area and a lot of breadth (as suggested by the horizontal line of T). I think even that view is a little dated already. I think what we need today for cutting edge research in natural sciences/engineering is Π π (Greek letter pi - capital and small respepectively) people or M people or even Pi people with three or even four vertical strokes of differing lengths.

[Indranil]

I am a researcher in the field of high performance computing. My major professor is counted in the who's who of this field. About 5 years back he wanted to return to India. He started teaching at an IIT for one semester every year. I traveled with him for a couple of years. Here are my observations.

1. India is still not welcoming of professors of eminence who want to go back. Whereas China is setting up labs for them and actively scouting for them globally.
My professor first wanted to teach at IISc. IISc said okay great why don't you join as Asst. Professor. My professor said, please have a look at my history. To this he was told. Yeah, we know, but if you are good, you will soon climb the ranks! Obviously, he did not take up on the offer. He later took up an endowed chair position at an IIT. But he was always treated as the teacher who comes for 1 semester. Everything moved so slowly for him. Masters students did not care to look up who he is. Worse, they thought this guy wants a true thesis, which was a hindrance to get the degree and get out as fast as you can.

Your advisor's experience at IISc (w.r.t. the joining as Asst. Prof. nonsense) is not typical, but rather an outlier! I know of profs at IISc who returned and joined at senior levels, and similarly at other IITs.

Not as atypical as you think.

NRI scientist AJ Paulraj wins Marconi Prize for 2014

The story goes that in 1970, Stanford Prof. Thomas Kailath, a brilliant and influential systems theorist who is himself a Pune-native, visited IIT Delhi to lecture on non-linear estimation. Inspired by Kailath's lectures, Paul went on to make fundamental advances in the area much to the Indian Navy's benefit. In 1971, after the war with Pakistan exposed shortcomings of the Navy's (British origin) sonars leading to the loss of a Naval ship, Paul led a successful project to redesign the sonar adding many new signal processing concepts. Three years later the new technology was widely deployed in the fleet.
...
In fact, Paul's work enabled India to overcome the military export restrictions imposed by the west. In an ironic twist, the Navy allowed him to go to Stanford on a two-year sabbatical, joining his mentor Tom Kailath. He returned to India in 1986 and served as the founding director for three major labs - CAIR (Center for Artificial Intelligence and Robotics), CDAC (Center for Development of Advanced Computing) and CRL (Central Research Labs of Bharat Electronics).

But by 1991, according to the now familiar narrative, bureaucratic battles began to take their toll, and with the consent of the Indian Navy, he returned to the US and Stanford University. ''His departure for Stanford University was a major loss for our country and the circumstances that led to his move may explain why we have so few Nobel Laureates from India,'' Admiral Tahiliani said.

[Vayutuvan]

But another who worked with Prof. Kailath is still at IISc (Prof. Umapathi Reddy).

[UlanBatori]

FYI, Prof. Kailath is from Bangalore, Keralastan onlee. Met him once in those parts do-chaar saal pehle.

[UlanBatori]

Hope this does not become OT, but I posted this elsewhere and want to add a small discussion here
I am not a fan of these SEZs etc to privatize national-level research priorities - they invariably lead to something with the word "scam" at the end. The question I have is whether there are any examples of real, basic and applied RESEARCH advances made in Indian ENGINEERING over the past many years by private enterprise, i.e., outside of the defense and nuclear domains. I know TIFR was well-reputed, but did not last as a well-reputed org for many decades. Gone now from public view, AFAIK. Hindustan Motors of Ambassador fame? Reliance? Tata/Birla? Mahindra? Larsen-Toubro? Kirloskar? Please think on the other thread.
This is my basis for being pessimistic about Indian engg. R&D coming out of private enterprise. Opening this door in Defense research will be just like Kabir ke dohe as paraphrased by IIT denizens:

jisko jitna chahiye,
kaat-kaat le jayiye

BTW, this is not limited to India. In the Yoo Ess Aye, there are stellar EXCEPTIONS to the general rule that private enterprise is terrible at real national-level, long-term research breakthroughs.

Bell Labs, yes, has been VERY innovative, but most of their BASIC RESEARCH I bet was funded by guvrmand. They did invent long-term things, so they are one of the stellar exceptions.

All the rest (yes I know this is aerospace arrogance/ ignorance) came from NASA or ARPA/DARPA or NRL or AFOSR/AFRL or ARL. JPL and Johns Hopkins APL are famous, but again they were govt-funded, they just ran the day-to-day administration and pocketed the overhead. Lockheed Skunk Works, General Dynamics Electric Boat Division, Boeing, all worked on govt. contracts, but the basic research usually came from NASA/ DoD and universities. Westinghouse and General Electric and Pratt&Whitney, same way. Still I treat all these as the stellar exceptions. In the huuuuuge US economy and its history, these are just way too few. If you have experienced, tough "sumb1tches" like Admiral Hyman Rickover and Chief Designer Kelly Johnson and NASA chief Werner von Braun running the show, yes, the taxpayer can get spectacular results through "private" industry contracts from the guvrmand. But India has, say, Admiral Golf Nad***** instead, shrilly demanding with false "statistics" that any successful indigenous development program be scrapped before going into production

[SaiK]

pooch: how did ge, p&w et al made it to big leagues? don't they share some storyline for us? what i am driving at is to use private funding space into national projects, and finally, the private space gets to productionize their investments. this would be a win-win for all, especially when there is a lack of synergies among the edus and r&d labs, and there exists little interest with GoI labs towards getting this goal materialized.

but then, i am going by the driver: GTRE calls for international participation for tech sharing? i hate this.. they blundered on this one. i'll take back all my questions, if they withdraw that, and look into pure r&d all home grown.

[UlanBatori]

GE, P&W, Boeing, Lockheed all made it on the immense funding of the US WW2 war effort. Note that the US lost over 50,000 (mind-boggling but I believe correct) airplanes, hajaar-hajaar tanks, hundreds if not thousands of ships, and how many - 1 million? human lives in that war. Losing would have meant far more. So that is the sort of production volume on which these companies learned their corporate knowledge. Most fields of research came up in that trial by fire and blood. You know that Operations Research got its first real boost, when it was used to plan the flight path and pattern of escort fighters to provide the most air cover and minimize losses of Liberty ships in Atlantic convoys headed for Britain. "minimize" didn't mean anywhere near zero, but each ship lost to submarines was another 50 or 100 men lost, and another few thousand who would die in Oirope because their cargoes did not reach.

IOW, the experience base behind the US defense companies, was built on the bodies of hundreds of thousands of American servicemen. Not just their tax dollars.

For all the takleef that India has faced, faar more Indian soldiers, sailors etc have died fighting FOR THE BRITISH than for free India. (MAY THAT NEVER CHANGE, INSHALLA*!)
Otherwise there would be a different kind of seriousness to Indian defence research.

So we cannot emulate the GE and other stories in India by just giving peacetime lucrative contracts to fatcats. The performance and pressure came long before the money in those cases. The question is how to advance research without that real blood pressure (pressure of minimizing blood loss).

[SaiK]

slighly dated info:
http://articles.timesofindia.indiatimes ... arch-tejas

He believes that projects undertaken by DRDO may not have got delayed had Indian industry and universities provided necessary support earlier. "Unlike what people believe, there has been no delay on DRDO's part. Our 7,600 scientists conceptualised, designed and then built products as industry had not developed sufficiently then. In other countries like USA, industry takes over soon after a project has been conceptualised. Even Universities provide necessary assistance in research. This is slowly happening in India now.

kinda reverberates similar thoughts however... i agree it does not need to be that way.

The DRDO won't give up on the Kaveri engine just as yet. One reason that Selvamurthy gives for not having succeeded in getting the Kaveri engines fitted to the Tejas is the absence of high-powered engine manufacturers in the country. Elsewhere, companies like GE and Rolls Royce manufacture engines for aircraft. Even the Tejas has had to fall back on GE engines. "We have managed to get 70 kilo Newton thrust out of the Kaveri engine. It has been tested for 55 hours on an IL-76. The Tejas requires a thrust of about 80 KNewtons. We shall try the engine on a Tejas and we believe that it will cover upto 80% flight envelop. We may use it for the trainers or even unmanned aerial vehicles (UAVs). There is also great possibility of use of the marine Kaveri engine," the DRDO R&D chief said. According to him, several foreign firms have expressed eagerness to partner with DRDO for an engine that can provide 130 KNewton thrust required for the AMCA. "In the last five years, the armed forces have ordered Rs 140,000 crore worth of equipment from us.

if the above gives any input, it should be the high-level requirements - [less firang eagerness, of course!!] - 130kN + correct on the failures to reach 100% of the designed thrusts.

highlights from chacha cache:
http://webcache.googleusercontent.com/s ... clnk&gl=us
---> centre of excellence in Computational Fluid Dynamics (CFD)at the IISc, Bangalore
---> centre of excellence inaerospace system design and engineering is beingset up at IIT, Mumbai
---> hyper-media digital library in IIT, Kharagpur,
---> NAL, BLR
---> singlecrystal super alloy and directionally solidified superalloy for use in high performance aero-engines

but these still lack focused areas information available on public domain... points to the need that we are not investing, focusing in the core areas of r&d.

another of those dated articles that you may like to bang at:
http://forbesindia.com/blog/economy-pol ... ence-suck/

Prof Jayant Murthy of Indian Institute of Astrophysics has commented: “The institutions that are being set up have the same old people in charge. These people, by and large with exceptions, are hidebound and hierarchical. In many cases, they are self-perpetuating because they pick their successors in their own mold.”

[SaiK]

http://nal-ir.nal.res.in/view/subjects/07.html
good links for you to look at.. and thereof!

[Rahul M]

I had a long discussion with a very experienced friend (EyeEyeTee Peechdee, high-up industry position in India, many saal going around whole Duniya as part of Yoo Enn, dealing with militaries.. and now a brofejar MrsDr also eyeeyetea Peechdee and Brofejar) during a trip to the Yoo Ess where I was the chaffeur. So I do have a good bit of inside info.

So the Indian system does give out money! I mean, how many desis come to Yoo Ess for week-long or 2-week conpherenj, staying in rejort hotels like the best in the duniya! The trouble is, it is more Soviet system - you have to APPLY for every trip and have it stamped by Babus, rush around getting this abbroval that abbroval, abbly phor vija to Paco's relatives, look for tickets... run around some more...

IOW, they don't control the budget, everything is microcontrolled by some Babu with no stake in achieving project objectives. Like Soviet system - you spend like money is going out of style, because there is no reward for saving, and no pangs of conscience for spending. OTOH, in my case, even when I have $$ (Cinderalla Phase) every penny spent causes me pain and I squeeze out everything I can get out of it. Always looking for innovation as in "cheap". No way can I afford to fly to and from desh, go stay in 4-star hotels in desh for 2 weeks, flying around conference to conference, all on my project funds - even if I have the budget I would rather spend it on paying a hungry undergrad or squirrelling it away to do something else more useful.

Same for all project purchases - everything has to be Approved in India. So time spent on actual technical work and thinking may be 5%, time spent on bureaucracy is 80%, vs, in my case, 25% and 60%. And I curse that!!!

Maybe Indian system needs to give more autonomy to the profs - meaning, u mess up, its ur problem.

saarji, some of the points may be cleared up from my own exposure to basic science research in India.

#the pee-chaddi students (usually) don't need to be paid out of project funds because they directly receive a fixed fellowship from CSIR/UGC when they clear the qualifying exam. others become institute fellows and are paid out of DST funds the institute receives annually. only in the rare cases are students paid out of projects funds and even these try to qualify the NET exam afterwards in order to wrangle the more assured CSIR/UGC fellowship. funds for paying the students is almost never a constraint.
the DST or CSIR would often fund trips abroad for students and profs so that project funds need not be utilised for such.

# the profs have near complete control of the grants once it has been sanctioned (which is done by senior scientists, not babus)

#in addition to grants to individual profs (source might range from DRDO to CSIR to videsi univ) the institutes gets its own funds annually from the DST which is shared by the profs subject only to approval by honchos like the director and the governing body (which usually means the honchos from similar institutes, other eminent scientists and a couple of babus from DST).

IMHO, the single biggest problem is the lack of performance based evaluation of faculty. once a new faculty joins (after pee chaddi and a couple of post-docs) he/she is permanent staff and can't be fired. many promising researchers develop a tendency to go to seed after making faculty.
and I would stick my neck out and say that at least 50-60% of basic science profs in India are deadwood, place-holders who do little more than grace faculty lists and attend seminars(sometimes, for the complementary chai biskoot). of the rest, the overwhelming majority have discovered the joys of publishing papers, however irrelevant, only because that gets you the recognition needed for 'young scientist of the year' type of awards from the gobarmand.
long ago in nukkad I had recounted the fascinating presentation by a student of one such prof :
http://forums.bharat-rakshak.com/viewto ... 27#p767027

thus, only a minority actually do research for the sake of doing new science, rather than just publish papers with high IF. (one worthy resorted to referring to his own papers numerous times so that his IF would go up)

also, while they don't have final say on overall funding, the big shot scientists (like CNR Rao, TVR, AK Raychaudhuri etc) enjoy tremendous say and have near full autonomy from the babus regarding how those funds are used. given that, I would hesitate to call the system soviet. these people, who actually implement policy almost invariably have spent part of their career in US, continue to have strong ties at top tier univs in US and europe and are strongly influenced by its academic system.

[Vayutuvan]

Rahul M: Referring own papers is discounted by half.

[Rahul M]

seeing that no one else was interested in his papers that was all he could do.