B.Sc (Hons) Computer Science Programs in Orissa Colleges a proposal

Chitta Baral, Associate Professor
Computer Science, University of Texas at El Paso
El Paso, TX 79902
homepage: http://cs.utep.edu/chitta/chitta.html
Preamble All of us know how Orissa has a vast mineral resources. To be exact in terms of all-India reserves, it has 98% of chromite, 70% of bauxite, 38% of graphite, 26% of iron ore and 23% of coal. It has adequate water, surplus power, and a large coast line of 480 kms. One of the main reasons, there are not that many manufacturing industries, particularly the ones that depend on minerals, in Orissa (for that matter around the mineral Belt of Orissa-Bihar-W. Bengal) is because of the old freight policy of Indian Railways (which falls under the central govt.) which charged transportation cost based on the weight of the material transported, not on the distance. Thus, the closeness of Orissa was not an important factor. Since these laws were changed in 1991, Orissa has been able to attract several mining and manufacturing projects. (According to a presentation by Mr. Muthukrishnan of TISCO who directs the project in Gopalpur Orissa is number ONE in attracting new projects in mining and manufacturing in last few years, in terms of the total investment. )

Now that is all good. How about other other industries? One of the other main focus of Orissa govt has been software industries. The main resource for a software industry is trained human resources. With very few B.Engg (Computer Sc) and MCA programs Orissa does not have critical mass of this resource
The OSEDC (Orissa State Electronics Development Corp) is aware of this and has a separate proposal to set up an institute called Indian Institute of Information Technology - IIIT -- in Bhubaneswar, for many similar reasons. This proposal is complementary to the IIIT idea and will be positively affected by its creation. Besides, NASSCOM says, Companies have projected a demand for appx 41,000 software professionals each year between 1996 and 1998. With some 55,00 (total) new IT professionals entering the work force during this period, the job market will remain tight. at this point. It does not have or produce a large enough pool of software professionals to be attractive to software companies. (Bangalore had this resource because of the many Electronic related public sectors -- such as ITI, BEL, ISRO, etc., and many many -- more than 20 -- private and govt Engineering colleges there.) To not miss the boat we have to act fast. But in these days of liberalization we can not have more public sectors in Orissa, and because of the huge cost factor we can not have too many new engineering colleges. Thus we can not just copy Bangalore. We need to come up with an innovative, quick and cost conscious approach. The rest of this document provides such a model which I believe will result in large scale availability of software trained human capital in Orissa, thus making Orissa an extremely attractive place for software related investment.

The main idea The main idea is to develop B.Sc (Hons) computer Science programs (perhaps in stages as dictated by the financial limitations of the state) at colleges all over Orissa where there are B.Sc (Hons) in Mathematics and/or Physics programs with future goals}to develop B.Com (Hons) in Information Systems programs across Orissa.

The modus operandi would be to limit costs by using existing infrastructure, such as buildings, class rooms, etc to the largest extent possible, with additional cost involved only in training Maths/Physics lecturers in Computer Science, and providing computer access, and hiring only one new faculty with computer science background (perhaps with an MCA or a M.Tech degree in computer science) for each college.

The innovativeness of developing B.Sc(Hons) computer science programs is that it minimizes additional cost and it is broad based in the sense that it can be, in stages, but fairly quickly, introduced in large number of colleges across Orissa. Unlike the currently existing MCA, and B.Engg (Computer Sc) which can only be offered by universities and the PG granting colleges, and Engineering colleges, thus limiting the number of institutions that can offer these degrees, the B.Sc (Hons) degree in Computer Science can be offered by a much larger number of colleges all across Orissa that already offer B.Sc (Hons) degrees in Physics and/or Mathematics.

A pilot/experimental program in a few (3-8) colleges should be introduced as soon as possible. This will make it easy to fund and start the program and will allow time for the idea to sink in and correction of any oversights before it is introduced in a larger pool of colleges.

Disclaimer The rest of the document outlines a possible implementation of the above idea. It is not our intention to say that this is the only way to implement the main idea and implementors in Orissa are perfectly entitled to take ground realities into account in making implementation decisions; but we request them that they take the arguments and main points in this document into account.

Main Issues
Who will teach? The main idea is to train 3-4 Physics and Mathematics lecturers from each college, in Computer Science, so that they can teach the B.Sc(Hons) courses in that college.

This will be much cheaper than hiring new computer science faculty. Moreover, the average salary of computer scientists in companies are so high that it would be difficult to get them to be lecturers in colleges, where they will get paid much less, because of the fixed salary slabs in colleges.

Also, private institution (such as NIIT) affiliates can not in general be hired to do the teaching as the instructors there are more trade oriented. I.e., they may know the usability of software packages such as LOTUS or EXCEL, they are not usually good at the basics of computer science, such as `data structures, and `finite automata'. Also, from a long term perspective it is better that we develop a pool of well qualified lecturers in computer science.

Nevertheless, the core of 3-4 trained lecturers could be supplemented with 1 new lecturer with a computer science degree and 1-2 application oriented courses taught by private institutions (or their affiliates).

subsection Ensuring quality of the trained teachers The success of this proposal critically depends on the quality of the students that will graduate with the B.Sc (Hons) in Computer Science degree. This in turn critically hinges on how good their teachers are, and this will depend on how well they are trained.

The main idea here is that the selected Mathematics and Physics lecturers are trained in Bhubaneswar, Rourkela, Berhampur, Burla, and Sarang where there are existing institutions with computer science faculty, who will be in the trainer pool. (CET, OEC, XIM, Dept of Computer Sc and Application -- Utkal and OCAC in Bhubaneswar, REC in Rourkela, NIT in Berhampur, UCE in Burla, and the Engineering college in Sarang.) These instructors will be helped by a advisory board of Computer Sc faculty from reputed institutions in India and abroad. Also, in case that well qualified instructors are not available locally, the government should hire well qualified trainers from outside. This investment will pay off in terms of better trained lecturers which will result in well qualified graduating students.

To ensure the quality of the trained lecturers, they will have to pass examination in each of the courses, that they will be teaching. (Just attending will not make them eligible to teach.) Special efforts should be made to ensure that the exam process is corruption free so as to ensure the quality.

Facilities Initial facilities in terms of computers and softwares will be provided by the Orissa govt, and later software companies and well-off alumni or well wishers (particularly NRIs) will be approached to donate individual computers (with their name as the domain name, if they would like that) or even to set up software labs bearing their names by donating most computers and softwares in that lab. (In the USA, a major source of income for colleges and universities is through fund-raising from alumni, well-wishers, and local businesses. This needs to be pursued with vigor.)

Initial physical facilities such as class rooms, laboratory space etc., for the new B.Sc (Hons) program will be provided by the corresponding college.

The aim should be to have 1 computer for each five Computer Science hons students and 1 computer for each 10 computer science pass students. (The government may consider charging Rs 5000 per year for each hons students and Rs 2500 a year for each pass student, as a computer usage fee, with waivers and/or loans for students from economically disadvantage background. Here also, scholarships could be funded through fund-raising, and people funding a particular scholarship will have the scholarship named after them.)

Affiliation and authorization issues For speedy implementation of this program, the program will be designed so as not to depend on the Central Govt or the All India Council of Technical Education.

The program will stress on the B.Sc (Hons) science aspect avoiding engineering so as not to require a certification from the All India Council of Technical education.

The program will be formally certified from the Universities in Orissa and classified as a B.Sc (Hons) program. Since, the program will follow the current pattern used in other B.Sc (Hons) programs and will not be proposing any radical idea %(except the 4-yr with thesis %option which should be pursued separately at a non-emergency %pace so as not to delay the implementation %of the 3-yr program), it should not have problem in getting approved by the University authorizing bodies.

(I just found out from the Silicon India news magazine% footnote{The exact news item is as follows: ``The Tamil Nadu Institute of Information Technology (Tanitec),a project which was announced in the 1997 Tamil Nadu budget, will be active in education as well as research and development. It is also expected to play a catalytic role in making Tamil Nadu ' an intelligent state.' The committee headed by Mr. N. Vittal, former secretary, Department of Electronics (DoE), which was set up to prepare a blue print for the project, recently submitted its recommendations to the Chief Minister. The institution will have a high degree of autonomy since it is likely allocated Rs 100 million for the project in his Budget speech. The committee has recommended starting the B.Sc. (Information Science) course by June, 1998 and B.Tech. course by June 1999. The institute is likely to be located in the Anna University premises. ''} that Tanitec -- Tamilnadu institute of information technology -- is also proposing a B.Sc degree in Information Science, in addition to B.Engg degree in Information Technology.)

The training program The training program has to be carefully designed with the main goal being that the training produce competent/qualified lecturers. (I was told that the 6 months training on Computers that the Orissa govt has been current providing to lecturers has not been very effective because of not so good trainers, no examination for the trainees, and no evaluation process for the trainers' teaching ability.) Because of the important role it plays in the success of the whole plan, the training process should be handled by an autonomous body without any outside interference.

Selection of the lecturers for training Mostly, lecturers in Mathematics and Physics will be trained. This is because the mathematical training received by them during their B.Sc and M.Sc makes them better prepared for computer science. In general a fair criteria, with little room for corruption, should be used to decide on the lecturers that will be trained. Some of the preference criteria that make sense are:

  1. Preferring lecturers with a Ph.D, as that indicates their ability to self learn, and
  2. Preferring lecturers in colleges that currently teach electronics and/or computer science as 4th optional subjects in +2 science and that teach the computer science paper in the Math (Hons) programs.

Incentive for the lecturers Various incentives should be provided to get the best lecturers. Such incentives include salary increments and promotions. But some of the other incentives that are also integral in the training process may include, a computer for each lecturer (that way he/she can do programming at home and would make the training place flexible -- no need to have a compute lab at the training center) and internet/email access (to be in touch with computer science faculty all over the world, particularly with the advisory board). An alternative to individual computers to trainee lecturers could be giving 2-3 computers to the parent college of each trainee that will be used by the 3-4 trainees from that college. (This alternative may be more practical if computers must be located in an A/C room.) Also, the fact that being trained in computer science gives the lecturers opportunity to consult in computer companies, is an incentive factor. Finally, they will be provided opportunity to do Ph.D in computer science at a later point.

Training details The basic training will be part-time for the first 3 semester and full-time in the 4th semester. A suggested training break-up is as follows:

  1. First Six months: (A1) Digital Circuits and (B1) Structured programming (in Pascal).
  2. Second six months: (A2) Automata and formal languages and (B2) Data Structures and algorithms (in C).
  3. Third six months: (C1) Design and analysis of algorithms and (C2) numerical analysis and (G) Object oriented programming with Java.
  4. Fourth six months: One of the following combinations.

(i)(D1) Databases and (D2) Artificial Intelligence and ($H_1$) Agents, workflows, and client-server computing using Java.
(ii) (E1) Operating Systems and (E2) Computer Architecture and ($H_2$) Introduction to networking and distributed systems.
(iii) (F) Compiler theory, system programming, LEX and YACC and ($H_3$) Graphics and multi-media.

The only restriction would be that lecturers from each college will be equally distributed between the three options in the last semester. To ensure the quality of the training some of the procedures that should be followed are:

(i) The training should be conducted by an independent body that will get the best faculty from local institutions. (For example, in Bhubaneswar the instructors could be from OEC, CET, XIM or Utkal University. The process of trainer selection should be judicious and fair so as to get the best (in terms of knowledge and teaching ability) trainers, and if qualified trainers are not available locally, then they should be brought from outside. (Note that the training location need not be fixed, and since the lecturers -- trainees -- will have their own computer a full-fledged computer lab in the training location may not be necessary.)

(ii) To attract the best trainers, they should be paid well.

(iii)The trainers will be helped by the advisory board and will be given real-time internet access (email, web browsing and web server access).

(iv) Trainers will be evaluated by the trainees after every course they teach. Trainers with bad evaluations will not be have their contracts renewed.

The trainees must pass the exam in each of the courses to be allowed to teach in their colleges. During the training their teaching load will be reduced (say halved) in their parent college. Reputed faculty from abroad and India will be invited to teach during the summer, and software engineers/scientists from industries will be invited once in a while to give short courses on novel softwares.

The 3 yr B.Sc (Hons) Computer Science program This will be a trend setting program where we demystify computer science and make it widely available, and treat it as a science degree not a technological degree. As mentioned before, the main advantage of this degree will be that being a B. Sc (Hons) program it can be offered in a much larger number of colleges as compared to MCA and B.Engg (Comp Sc) programs which can only be offered in few colleges -- that are authorized to have PG depts -- and universities and engineering colleges respectively.

To be able to get speedy approval from the universities, one of our goal would be that the B.Sc (Hons) Computer Science program should be structurally as close as possible to the other B.Sc (hons) programs that are currently available. So first we review the structure of the current B.Sc (Hons) programs.

The current B.Sc (Hons) programs Currently, B.Sc (Hons) programs have one Hons topic, two pass topics and ancillary and foundation topics. Also, the program is divided into three parts, with one exam at the second year involving part 1 and 2 and another exam at the end of the third year involving part 3.

In the Hons topic, there are 8 total papers, 2 paper each in part 1 and 2 and 4 in part 3. There are 4 papers in each pass topic, 1 paper each in part 1 and 2 and 2 papers in part 3.

The regular 3-yr program As in case of the other B.Sc (Hons) programs this program will be for three years after +2. Thus, to avoid possible delay in implementation, initially no authorization from the All India Council of Technical Education will be sought.% footnote{Note that many universities in US -- such as the prestigious University of North Carolina -- offer B.S in Computer science, even though, they do not offer any engineering degree. Similarly, in the prestigious University of Texas at Austin -- Computer Science is in the College of Science and not in the College of Engineering. The college of engineering there has a separate program in Computer Engineering.}

The three year honors degree will be like any other existing B.Sc Hons degree in its structure. For the Computer Science Hons, one of the pass subject must be Mathematics and the other may be Physics or Chemistry. The details of the pass subjects, and the ancillary and foundation subject will remain unchanged. Hence, we now only discuss the possible content of the eight Hons papers. They are as follows:

  1. Digital Circuits (A1) and Automata and formal languages (A2). (50 marks each with a total of 100 marks. no practicals.)
  2. Structured programming -- in Pascal (B1), Data structures and Algorithms in C. (B2) (30 marks each in B1 and B2 and 20 marks each for programming practicals in B1 and B2.)
  3. Algorithm analysis and design (C1) and Numerical analysis (C2) (50 marks each with a total of 100 marks. no practicals.)
  4. Databases (D1) and Artificial Intelligence (D2). (30 marks each in D1 and D2 and 20 marks each for programming practicals in D1 and D2.)
  5. Operating Systems (E1) and Computer Architecture (E2) (30 marks for the theory of E1, 20 marks for the practical of E1, and 50 marks for the theory of E2.)
  6. Compiler theory, system programming, LEX and YACC (F) (60 marks theory and 40 marks practical.)
  7. Object-oriented programming and Java (G) (40 marks theory and 60 marks practical.)
item Electives (H). One of the following:
  1. Agents, workflows, and client server computing using Java.($H_1$) (40 marks theory and 60 marks practical)
  2. Introduction to networking and distributed systems. ($H_2$) (75 marks theory and 25 marks practical)
  3. Graphics and multi-media. ($H_3$) (50 marks theory and 50 marks practical.)
Among these courses, Part I will consist of A (A1 and A2) and B (B1 and B2), Part II will consist of C (C1 and C2) and D (D1 and D2) and Part III will consist of the rest. But to decrease the load of teaching all of Part III in the 3rd year, G will actually be taught in the second year.

(It is hoped that the B.Sc program will start 6 months after the start of the training of the lecturers. The training program is designed accordingly, except that for the first batch outside lecturers will be needed for teaching (D1), as the lecturers will be getting trained in (D1) at the same time. For all the other courses, the training of the lecturers will precede the teaching in the B.Sc (Hons) class.)

The program quality will be ensured through university conducted exams as in the other B.Sc (Hons) programs. For the practicals, program portfolios will be submitted to external evaluators who will have responsibility of checking copying of programs and giving have the grade.

The 4-yr B.Sc (Hons) with thesis program Although the B.Sc (Hons) Computer Science program will be of three years, students will have the option to do a thesis and two independent study courses in the fourth year to obtain an augmented degree which will be called B.Sc (Hons) Computer Science (with thesis). (An authorization by the All India Council of Technical education for this 4-yr degree may be necessary.) This degree will make it easier for students to do higher studied abroad, particularly in the US, where B.S degrees are of four years.

For the 4-yr with thesis option students need to pick two topics (from the list in the previous section) which were not electives for independent study (but it will involve exams and projects) and do a thesis. They may do this while working in a company.

To not delay the implementation of the core 3-yr program, this option should be pursued separately at a non-emergency pace. Hence, it should not be a part of the initial proposal. It should be only proposed after the 3-yr program is approved and installed in a few colleges.

Selecting the students The students for the B.Sc (Hons) program may be picked based on their PCM marks in I.Sc. They must have at least 55\% in these subjects though. (This will avoid the extra hassle of an entrance exam and current policies will be followed for transfer between depts and hons.) Another possibility may be to use the current Orissa JEE for engineering colleges and pick students from them.

Relation with other programs The other related programs that are university approved are the B.E and B.Tech programs in Computer Science (and Engineering), MCA programs and M.Sc in Computer Science programs.

This program differs from the B.E and B.Tech programs by staying away from Engineering. It differs from MCA by being a science program with substantial theory in comparison to the MCA program which focuses on applications.

This program will be like any other B.Sc program with similar rules and policies, with minimal change in case of the 4-yr with thesis option.

In comparison to the programs offered by private institutions such as NIIT, ACE etc., this will be a legitimate program. The 2-3 year PGDCA programs offered by these institutions (particularly, in Orissa) are mostly money making ventures with little regard to teaching. The lack of good teachers footnote{Some affiliates in metros such as New Delhi are pretty good. A friend of mine after taking NIIT course in Delhi got a job in HCL. But I have not heard any such success story in Orissa. The best I have heard is that students after finishing their course work at NIIT have got a teaching job at the same or a similar private institution or a data entry job.} and the absence of legitimate exams are the main reason for the poor quality of the program offered by them. Besides, the amount of money charged by them makes it unaffordable to most.

But, they sometime offer good short courses in particular software packages. As mentioned before, a college may contract with them to teach such a course as an elective.

Relation with other existing and proposed institutions in Orissa The existing institutions offering B.E/B.Tech and MCA will initially help in providing instructors for training and in the process will be helped through the contact with the advisory board members and additional infrastructure.

The proposed IIIT can take over the training role once it is established and can serve as a central node between the various B.Sc (Hons) granting colleges and the advisory board.

The competition from the B.Sc (Hons) programs will provide impetus for the institutions granting B.E (Computer Sc) and MCA to improve their standard, so as to maintain their superiority over the B.Sc (Hons) program. The M.Sc (Computer science) program in the universities should then be upgraded to be next step for students with B.Sc (Hons) Computer Science degrees

Advisory Board The advisory board will have two parts; one consisting of reputed faculty in India and abroad and the other consisting of leading computer scientists and computer entrepreneurs in Industry in India and abroad. (Some of the possible candidates from the academia in the Computer Science and/or Engineering field are:

Computer Science faculty at REC, UCE, OEC, CET, Utkal, etc. Chitta Baral, Associate Professor, University of Texas at El Paso Laxmi Bhuyan, Professor, Texas A & M University Chita Das, Associate Professor, Pennsylvania State University Gautam Das, Associate Professor, Memphis State University Rabi Mahapatra, Assistant Professor, IIT Kharagpur (visiting Texas A & M Univ) Prasant Mohapatra, Assistant Professor, Iowa State University Bhubaneswar Mishra, Associate Professor, New York University Durga Mishra, Associate Professor, New Jersey Institute of Technology Jaydev Mishra, Professor and Chairman of Computer Science, University of Texas at Austin Brajendra Panda, Assistant Professor, University of North Dakota Dhabaleswar Panda, Associate Professor, Ohio State University L. M. Patnaik, Professor, IISc Bangalore Dheeraj Pradhan, Endowed Chair Professor, Texas A & M University Amiya Pujari, Professor, University of Hyderabad Kamalesh Rath, Assistant Professor, University of Texas at Dallas Ashok Samal, Associate Professor, University of Nebraska, Lincoln )

Visiting Faculty Reputed visiting faculty and professionals (some from the advisory board) will conduct some of the training and will also give short term courses and seminars.

Real time Internet access will have a major impact on this. (For example, in US professors are paid by the university only for the non-summer months. In the summer months, they are free to pursue other activities on their own, and many grant agencies that fund summer research do not require the professor to be physically located in the US. This means many of the Oriya computer science professors can come to Orissa in the summer and do research and teaching there. But, they would do that only if they have real-time internet access through which they will be in touch with the day to day activity of their students and research colleagues in US. I personally will be very happy to spend 2 months in the summer teaching and doing research in Orissa every 2-3 years; but again, only if I am given real-time internet access from Bhubaneswar to my computers in the US. Similarly, Oriya computer scientists working in the US would also spend more time in Orissa and would help out through short -- 1 week -- courses if they are provided internet access.)

Internet As mentioned in the previous section, having real-time internet access from Orissa is extremely important for the success of this program. It will also enable fast communication between the instructors, the lecturers and the advisory board members which is critical to the success of this program. Moreover, the World Wide Web accessible through the internet is like a virtual library and access to it would mean less expense in terms of setting up full blown computer science sections at libraries in the different colleges.}

If there is a delay in expanding the govt networks such as the VSNL, NIC etc. to colleges and institutions, then private internet providers should be enticed, perhaps with a 2 year contract for real-time internet connections at several colleges and institutions.

Time Line: A best case scenario

  1. Form the advisory board by Sept 1, 1997.
  2. Course details formed by Oct 1, 1997.
  3. Identify the training instructors (by Oct1, 1997) and put them in touch with the advisory board which helps them with course materials, books, course content development, softwares, etc. Provide email access to the training instructors by Oct1, 97 and if that is not feasible then at least provide international fax access so that they can have fast communication with the advisory board members.
  4. Identify lecturers to be trained by Dec 1, 1997 and start training by January 15, 1998.
  5. Start B.Sc (Hons) classes by July 1, 1998. (Obtain university senate approvals before that.)

Where money needs to be spent?

Pilot program around Bhubaneswar -- first alternative Perhaps we start with a pilot program in Bhubaneswar involving about 8 colleges around Bhubaneswar. A rough estimate of the cost for the training and the infrastructure at these 8 colleges would be as following.

  1. Training 32 lecturers\footnote{Besides the computers, since a good class size for a computer science class is 30 students, the cost of training 32 people will be the same as training fewer people. That is the rational behind suggesting 32 lecturers from eight colleges.} of eight colleges (say from BJB, Ravenshaw, RD Womens, Basic Sc at OUAT, SB Womens, SCS, Maharshi, PN, Stewart Science)
  1. Infrastructure for the trainers -- 2 trainers in the first year, 4 trainers in the second year (Computers, books, fax/internet access, renumeration)
  2. For the 32 lecturers (32 computers, all books, transportation cost, allowance
item Infrastructure at each college.
  1. Internet Access at the eight colleges and at the training institutions (Utkal Univ, OCE, CET).
  2. For 32 honors student in each college in each batch and additional pass students. (10-15 computers in each college with OCAC or a similar body in charge with the maintenance of these computers.)

Future training programs Future training programs may be held in Rourkela, Berhampur, Sarang and Burla, or may be done centrally at the IIIT. In the later case lecturers from other parts of Orissa may be transferred temporarily to Bhubaneswar/Cuttack area colleges to take the training with the understanding that they return to their original college after the training.

Second Alternative If setting up the training requires time, and we need to start right away, then another alternative would be immediately start the B.Sc (Hons) Computer science program in few (3-4) colleges by hiring MCA graduates as lecturers for that college, and then following-up with the training plan for expanding the program to other colleges.

Future plans -- short term and long term

Computer Science Pass Students doing B.A (Hons) in mathematics or B.Sc (Hons) in non-computer science subjects could take computer science as a pass subject, where they will be required to take four computer science papers (A, B, D and E). This will lead to combinations such as:

  1. Mathematics (Hons), Physics and Computer Science pass.
  2. Physics (Hons), Mathematics and Computer Science pass.
  3. Chemistry (Hons), Mathematics and Computer Science pass.
  4. Botany (Hons), Zoology and Computer Science pass.
  5. Zoology (Hons), Botany and Computer Science pass.

Such combinations are important because more and more physical scientists and biologists have to know how to use computers. For example, biologists working on gene sequencing need to know the basics of computer science. In fact many of the Oriya physical scientists/biologists in the US do a considerable amount of programming.

B.A. and B.Com programs The idea of B.Sc (Hons) in Computer Science will be extended to have B.A and B.Com (Hons) in Computer Application and Information Systems by training Commerce and Arts lecturers with suitable background. These two programs will be very similar and focus more on application programming and less on the underlying theory (where Mathematics plays a critical role).

I urge Oriya faculty in Information System departments to suggest the details of such a program

M.Sc programs in Computer Science The M.Sc Computer Science program at the Universities, such as the one in Utkal University, will be impacted by the B.Sc (Hons) program. They would have to be upgraded, with better and larger faculty and additional steps may be taken to make them really good programs. (While students with the 3-yr B.Sc Hons Computer science degree will be eligible only for the M.Sc Computer science program, the students with the 4-yr B.Sc (Hons) with thesis degree will also be eligible for the M.Tech in Computer Science programs.)

Possible Criticisms/Questions and Responses to them

  1. Q: Will it be easy to train Math/Physics lecturers in Computer Science? Also, from the students perspective is it easy to learn computer science?

    Ans: It is not easy, but it is doable with enthusiastic participants and a well-designed program. Most lecturers I encountered during my ISc in 81-83 were pretty good, and in the US and many other places most early Computer Science professors were converted Mathematicians/Physicists/Electrical Engineers.

    From the students point of view, I can say that Computer Science is not some elite subject which only a chosen few can do. In fact it is sometimes easier than Maths and Physics. (I have been the head of the graduate program in Computer Sc at the University of Texas at El Paso for last 5 years, and during those years I have had students from all kinds of backgrounds -- from civil engg to Physics to Maths, who have gone through leveling courses and then done their MS in Computer Sc, and then joined prestigious Software firms in the US.)

  2. Q. Are 3-4 lecturers in each college enough?

    Ans. Enough to start. Initially, each lecturer would have to learn the equivalent of the leveling and basic courses -- there are 4 such courses at our university, and learn 4 of the later courses. That means in each college each of these 3-4 lecturers can teach the basic courses and they can teach about 8-10 advanced courses. That would be enough for a B.Sc (Hons) in Computer Science.

  3. Q: Won't the lecturers leave and join some company.

    Ans. Many of them are settled and probably won't leave. %Perhaps they can sign a bond to not leave for certain number %of years. Also, after the training they will have good conceptual knowledge with some programming experience; they won't be expert programmers. It is the later who are sought after by the companies. (In general people who have been teachers do not normally like to join a company.)

    Some lecturers may leave for higher study though. In fact this should be encouraged by the govt, as this will not only produce more qualified computer scientists with ties to Orissa, but also serve as an attraction to go through the training. %with %appropriate measures to ensure they come back for a %certain number of years.

  4. Q: Where will the students find the job? (I was told about 50\% of the Computer Science graduates of Orissa Engineering College in BBSR, have difficulty in finding jobs.)

    Ans. Recall the NASSCOM statement about current and anticipated demand of IT professionals. If some graduates have not found a job, it is probably because they have limited their search area. They could easily find jobs if they went to metros or Bangalore and stayed there and looked for jobs. This is because many companies prefer local talent because of the cost/hassle of interviewing out of town applicants. (CS graduates in US know this very well.)

    We can not wait till enough companies are there in BBSR/Orissa to start new CS programs. This is because the companies will come only after there are enough CS professionals around. So, even if there is not enough demand for IT professionals in BBSR now, we still need to go ahead with this proposal. Initially, the CS graduates who could not find job in BBSR will go to the metros and Bangalore. But soon more and more companies will come to Orissa because of this availability of human capital. Also, hopefully some of the graduates will return and set-up their own companies.

  5. Q. What about the private institutions such as NIIT etc that have opened shop in urban centers of Orissa.

    Ans. From my experience, because of the lack of good teachers they are pretty bad in Orissa. (In Metro towns with educated IT professionals teaching part-time some of them are pretty good.)

    But, many of these institutions, in their attempt to protect their turf may put political pressure against the idea of B.Sc (Hons) Computer Sc in colleges across Orissa, and create problems.

    Our political leadership has to overcome this.

    Some kind of partnership between them and individual colleges in terms of using their facilities (particularly computers) for the lab classes may be formed for short durations. But this may trigger corruption in terms of lucrative contracts; so this has to be done properly.

  6. Q: But the trend now is towards privatization, not more govt spending in secondary education.

    Ans. I sincerely hope the well-off Oriyas and Software companies come to the rescue in reducing the investment by the government. The government may also decide to charge computer usage fees for Computer science honors and pass students.

  7. Q: How about the quality of these programs?

    Ans: The B.Sc exams conducted by the university will ensure some degree of quality. But overall, the quality of these programs will critically depend on the training imparted to the lecturers and the facilities made available at the colleges. This should be taken very seriously, as it does not serve any purpose to graduate unqualified students who are overlooked by the software industry, thus giving a bad name to the whole program.

  8. Q: How about focusing on a really good institution instead of many many not so good institution.

    Ans: The idea of having an IIT quality Information Technology institute, such as the proposed IIIT in Bhubaneswar is an excellent goal. Bhubaneswar and Orissa does need a top quality information technology institute graduating to quality computer scientists. But together with the IIIT, we also need to have broad based programs that can be easily duplicated in all nooks and corners of Orissa. The proposed B.Sc (Hons) in Computer Science program is such a program.

    The goal of this proposal is to provide opportunity to a broad segment of the population; not just for a chosen few. Also, the vision is to not just to make Bhubaneswar a software city, but to make Orissa a software state.

*Acknowledgment The suggestions given by Prasant Mahapatra was extremely valuable. Comments from Manoj Sahu, Surendra Ray, Manaswi Patnaik, Purna Mohanty, Gautam Patnaik, Laxminarayan Das, Aniruddha Sahu, Laxmi Bhuyan, Saroj Behera, Suniti Behera, Prabhu Mohapatra, Srikant Nayak, many members of ornet@cs.columbia.edu, and many people that I talked to at the OSA convention and the Invest Orissa Symposium in Houston was also very helpful in improving this proposal. end{document}