[DerAlte]

Originally Posted by Eddy

... A candidate does not know what a century is, when the last millennium started and does not know which century he lives in etc ...

The interview should end then and there. He is unlikely to know anything else either. That person would have quite likely spent his time inside a smoke haze.

Originally Posted by amitk26

... Usually for a fresher interview revolves around these achievements. If a candidate is able to explain the objectives and how he realized in sufficient details including both theory and his realization then he/she should be selected. ...

Correct, but a person has more facets than just the project. That would be only a dip stick check.

Originally Posted by zoombiee

... I think for comp sci guys, its better to find a good professor and do a project that will actually let them learn something in detail.

IF the professor knew anything practical in detail, he wouldn't be a professor - he would be working in the industry.

As Samurai succinctly mentioned, "The current professors with M.tech and Ph.D have no clue about industry".

[zoombiee]

Originally Posted by DerAlte

IF the professor knew anything practical in detail, he wouldn't be a professor - he would be working in the industry.

As Samurai succinctly mentioned, "The current professors with M.tech and Ph.D have no clue about industry".

I dont agree with the first statement. People do get into teaching because they have a genuine passion for it and can pursue their lines of interest (this is a smaller percentage I agree). It is not necessary for every one to pursue graduation to work in the industry. Quite a large amount of innovation in IT has come about from colleges and not industry especially on the OS and DB fronts. I have spent about a decade in the industry and find that by and large the Indian IT industry dumbs out most people as time goes by. Most of the work is plain ridiculous and does not need technical graduates. There are a few genuine places where work is generally really good (again a small percentage).

[sgiitk]

@zoombiee; and @DerAlte; As a lifelong academic let me add. Till about two decades ago people did opt for teaching but today the rewards ratio is too poor to think about it.

[DerAlte]

Originally Posted by zoombiee

I dont agree with the first statement. People do get into teaching because they have a genuine passion for it and can pursue their lines of interest ...

You don't have to agree. Instead, please do interact with academia to get the correct picture. Go to an HoD, no less, and see if he knows as much as you. The only things he would know for sure would be the curriculum and the calendar of classes. Even I wish that they are into teaching because they have genuine passion, and not because of economic realities.

There definitely exists a small set of people who after having worked 15-20 years in the industry have now become teachers. Do interact with them to know the current state they are in. The ones who predate them in the institutions keep them locked up in lectures and don't allow them any leeway or progress - for fear of getting superseded by superior knowledge.

Originally Posted by zoombiee

... I have spent about a decade in the industry and find that by and large the Indian IT industry dumbs out most people as time goes by. ...

I have spent 35, and I have never found any reason to believe that. The industry doesn't dumb out people. It is the people who chose to languidly spend their time dumbing themselved down. What is required first is the desire and courage to LEAD ONESELF, and not wait for instructions from others. If one really wants to excel at what they do, the industry doesn't come in the way, nor does it disincentivize it. If one couldn't, the industry is not at fault.

Even in the worst of the industry environments, there is enough scope to innovate and invent, *provided* one is involved intimately enough to recognize it. Most people spend their time only peripherally involved, and also waiting for others to tell them what to to.

I have also spent 25 of those 35 interacting with academicians, from Level 1 institutions to Level n, on various occasions. I have seen enough to be disillusioned. Your idealism is refreshing.

PS: Dr. Gupta, you are old school!

[zoombiee]

Originally Posted by DerAlte

I have spent 35, and I have never found any reason to believe that. The industry doesn't dumb out people. It is the people who chose to languidly spend their time dumbing themselved down. What is required first is the desire and courage to LEAD ONESELF, and not wait for instructions from others. If one really wants to excel at what they do, the industry doesn't come in the way, nor does it disincentivize it. If one couldn't, the industry is not at fault.

I regularly keep in touch with my HOD and i can confirm that he is indeed more knowledgeable than me. I do know a few more technologies (so called) than him, but he is definitely more knowledgeable. Quality of Technical institutions - that is another debate .
All my questions are in relation to the field in which I work on (comp sci).
I can also confidently tell you that I have more working technical knowledge than my engineering director who has about 20 years of industry experience.

Q1 -> When one is in the industrry (in INDIA), one really does not have the liberty of choosing what you work on. There is a exception to this, but in most service companies you just cannot. You have to work on what is available
Q2 -> 7 times out of 10 dirty and quick approaches are taken. Pain-staking perfectionist solutions are neither appreciated nor encouraged in general. Everything comes down to dollar savings
Q3 -> Most projects use off the shelf products. Very less room for innovation and usage of skill. For ex - Most ORM specialists have little or no knowledge of underlying database knowledge. A college pass out will know more than these industry veterans

All in all - there are incompetent people in the industry and in colleges. The percentages usually would be the same. The same incompetent person in college will remain incompetent even after 10 years in the industry. Passion/Pride/Hunger/Responsibility are not taught in colleges. They are taught much earlier and such people should not even get into colleges.

[DerAlte]

Originally Posted by zoombiee

I regularly keep in touch with my HOD and i can confirm that he is indeed more knowledgeable than me. ...

Nice to see someone retain respect for their HoD.

Originally Posted by zoombiee

... I can also confidently tell you that I have more working technical knowledge than my engineering director who has about 20 years of industry experience. ...

Very nice. How?

Originally Posted by zoombiee

... one really does not have the liberty of choosing what you work on. ...

Why should one? Isn't the company paying money have the primacy in deciding 'what'? Expecting that independence is rather selfish, don't you think?

Originally Posted by zoombiee

... 7 times out of 10 dirty and quick approaches are taken. Pain-staking perfectionist solutions are neither appreciated nor encouraged in general. Everything comes down to dollar savings ...

Strange sentiment! Obviously an organization's interest is in the money / economics, and not idealism and perfectionism - which are an anti-thesis. Funnily, whether or not to use Q & D approaches are always in our control, no?

"Dollar savings" is an emotional expression; the universal truth is investor's interest / returns. Can / should / must you deny them that?

Originally Posted by zoombiee

... Most projects use off the shelf products. Very less room for innovation and usage of skill. ...

The objective is construction efficiency and much smaller turnaround time, Start to Roll Out. And I see you don't look at industry from "Investment Efficiency" and "RoI" p-o-v at all. What is the difference between "innovation and usage of skill" and "trying to reinvent the wheel"?

Innovation cannot be necessarily found by going back to first principles and "assumed personal skill" (which is another way of saying "it gives me kicks to do it"). And yet, despite all these 'perceived' constraints, creative people find sufficient avenues to satisfy their desire to 'innovate'. Everyone benefits, and no one says "No, don't do it". Just needs the right objectives, no?

Originally Posted by zoombiee

All in all - there are incompetent people in the industry and in colleges. ...

The sordid state is not due to "incompetent people" (yes I agree; they are present everywhere in the same %age). It is a set of people choosing an easier way of life, a life devoid of any sense of excellence. Corrupted principles, not lack of innate abilities. Extreme self-serving subservience. Absolute reliance on the paradigm "No pain, a little gain, I can lead a hassle free life. Who wants headaches.".

[zoombiee]

Originally Posted by DerAlte

Nice to see someone retain respect for their HoD.

Very nice. How?

That is quite difficult to explain here. If we meet over a drink sometime, I will let you know, but trust me what I said is true.

Why should one? Isn't the company paying money have the primacy in deciding 'what'? Expecting that independence is rather selfish, don't you think?

I did not comment on what is right and what is wrong. It was a statement to substantiate how people are made dumb over a period of time. Institutionalization is a very big factor and most people fall prey to it. A small percentage who defy it do not represent the fact that industry makes people dumb over time and infact rewards them for it.

Strange sentiment! Obviously an organization's interest is in the money / economics, and not idealism and perfectionism - which are an anti-thesis. Funnily, whether or not to use Q & D approaches are always in our control, no?

I agree with that. There is a short term goal and a long term goal. Most service companies have limited shelf life contracts with their customers. Something that benefits in the short term may prove costly in the long term.

"Dollar savings" is an emotional expression; the universal truth is investor's interest / returns. Can / should / must you deny them that?

Yup I do. But that decision should come from the customers typically having understood all the pro's and cons. I dont see this being done in an ethical way.

Innovation cannot be necessarily found by going back to first principles and "assumed personal skill" (which is another way of saying "it gives me kicks to do it"). And yet, despite all these 'perceived' constraints, creative people find sufficient avenues to satisfy their desire to 'innovate'. Everyone benefits, and no one says "No, don't do it". Just needs the right objectives, no?

Again I agree. Some one (not every one) who is well versed with say SAP will not be able to answer basic algorithm questions or GK questions when asked in an interview. What I meant here, is that there is a tendency to become slaves of a particular technology instead of understanding the basics.

I think we are on the same page. What I wanted to say that industry does not mould people or make them more professional than they are when they come out of college. A person who is aggressive and wants to learn will do so both at a crappy college and a crappy company. A person who is not so aggressive and likes inertia will do so both at the best college as well as in the best company.

There will always be a small set of people who defy all odds. They do not represent the general truth. There are a few honest politicians, yet the generally held thought is that politicians are corrupt.

[Samurai]

Originally Posted by zoombiee

Q1 -> When one is in the industrry (in INDIA), one really does not have the liberty of choosing what you work on. There is a exception to this, but in most service companies you just cannot. You have to work on what is available

Well, it is possible to choose your work. And here is one way to do it.

I gave a speech couple years back at an engg college function, where I was the chief guest. Following is an excerpt from that speech.

Almost all of you are obviously hoping that if you ever make it into large companies like TCS, Infosys, etc., your career like made. Well, it is not really that simple.

When you undergo campus recruitment, you generally have no control over which company you join. The first company that hires you gets you. Large companies hire 10,000 to 30,000 people every year. And they have no way to ask each new employee what they want to do. Instead they are assigned to various tasks irrespective of their interest or talent. So your career becomes a lottery. In fact, there is no difference between a distinction student and a just pass student once they enter a corporate. Nobody looks at your marks sheet, academic accomplishments or extracurricular activities.

Large companies mainly focus on developing you into a specialist. That means you will rarely have an overall understanding of the system. If they put you into a specialization, they will keep you there forever, whether you like it or not. You will not know about anything else. This can be very frustrating after a while. Ideally, one should start off as a generalist, before choosing an area to specialize on. Imagine becoming a heart specialist before becoming a general purpose MBBS doctor. It doesn’t make sense, in fact it makes you rather dumb. Yet, this is what happens most of the time in IT field.

Now, you might be wondering how you can escape this fate. Well, it is very much possible, if you have the will. Don’t become part of this grand lottery. Instead, choose where you want to join. In the beginning of your career, avoid joining large companies. Instead, look at companies that have less than 500 employees. Small companies can’t avoid generalists. They need employees who can do everything, so they will train you in everything, make you work on everything. You will get to work on every aspect of project or product development. After working 4-5 years as generalist, you will be capable for creating solutions or products all by yourselves. Now is the time to become a specialist, in a field you enjoy most. With your seniority, you can pretty much dictate which specialism you want to pursue. This is how you can get to enjoy your career the most.

Therefore, don’t just look at employers by their name recognition. Look at what they will bring to your career. Take control of your career from the get go.

This actually worked for me personally. I started with a company with 30 people, so we learned to do everything. Even as programmer, I used to do client visits along with sales guys to support them.

So... when I encountered my first software services project at HP in 1995, I already had 5 years of product development background. This project had nearly 150 people including 120 C++ programmers across 10 teams. The development work itself was very boring and repetitive. The design was done by Andersen consulting, and we had no access to it. We were supposed to read pseudo code and write C++ code. I didn't want to do it. So, I looked towards creating new work. I was a tools guy by then. My mantra was "If you need to repeat, automate it". There was already a CASE tool that created some of the code. So I wrote another tool to generate source code from a definition (meta data) file. I brought it to a level where 80% of the source code could be generated and only 20% need to be hand coded. Whenever bugs were found, I fixed the tool instead of source code. Showed it to the boss (team leader), after she was suitably impressed, I asked whether the whole team of 10 can use it. She right away saw the benefit and green signaled it.

Now my personal project attained semi-official status within the team. I could work on it openly. Since the whole team needed access, I had to add many new features. Since everybody in the team needed access to each other's code, I had to maintain a common definition file which can be edited by all without overwriting each other's changes. So I kept the definition file in my HP-UX workstation, and turned my standalone tool into a socket based client-server application. The client tool would talk to my server app and retrieve the definition file. Until the user returns the file, nobody else can take out the file for editing. But disaster struck on the second day when a user returned a blank file and wiped out the combined definition file. Now I realised the importance of a version control system, and added that feature using the RCS utility (Revision Control System of Unix) on the backend. Now locking and version control was managed by RCS, allowing people to access older definitions if required.

Now I hit the next problem. When I was alone, it was easy to ensure I was using the latest version. Everytime I fix a bug, how do I ensure others use the latest client instead of the outdated one? I didn’t want to move code generation to the server side since it would steal my CPU power. Also, I had no superuser power to maintain it in a central place. FTP was the only way to share files. So, I wrote a socket based version control service. Every time the client tool would run, it would check with my server whether an update exists. That way I could send a warning to download the latest or even stop the execution if the update was critical. Now I had complete control over the client tool no matter where it ran. It also had counters to keep track of number of times the tools were invoked. The productivity of my team members suddenly went up, and two other teams in the same floor noticed it, and soon adopted the tool. There were 7 more teams (nearly 100 people) in Madras, headed by much senior team leaders. They didn’t show any interest in our un-official tool and continued to do it the hard way.

Meanwhile, the entire project had hit a snag. After 4 months of coding, we still couldn’t compile. Our source code had dependency on a template library, which was specifically designed for this project. That template library didn’t compile on HP-C++ compiler because of some compiler bug. Until HP C++ compiler team fixed the bug, we had no way to compile any code. One fine day I started looking at the template library and discovered a shocking fact. That code didn’t require template feature at all. It was not generic code. The entire 150 people team were held hostage to a feature that was not needed in the first place. All because of a stupid designer who didn’t know how to use templates. But who is going to listen to a 5 year experienced programmer, when the library was designed by the venerable Andersen Consulting? When I explained the problem to the 3 team leaders in Bangalore, they couldn’t understand it. They were technical, but they had never done template programming, so they couldn’t get it. Therefore, there was no question of escalating it to the higher ups.

Then I made an offer they couldn't refuse. I said I can modify the code generation tool to generate non-template classes to replace the template class library. That means we can start compiling. That was too irresistible, and I got the go ahead. In couple of days I modified the tool to generate non-template classes to take the place of template classes. Then we had all the Bangalore programmers re-generate the source code, and now we could start compiling. It was a bloody coup! Madras team continued to ignore us, while being very skeptical about our compilation status. That was mainly because they were in-charge of coding the template library, and didn’t want to accept any suggestion from us. In the next couple month, we compiled the code, unit tested the code, fixed all the bugs.

Meanwhile, the top boss (my manager’s manager’s manager) who was located in Bangalore was confused. She knew the Bangalore team was compiling using some non-template version, while Madras team was waiting for the compiler bug fix. While she suggested the Madras team to follow our lead, inter-branch rivalry was enough to resist the recommendation.

Then there was a funny situation. Those days HP C++ compiler used to get patches very frequently. One day the non-template class code generated by my tool ran afoul of the C++ compiler. It compiled before update, but failed after that. It took me more than a week to find the issue and put a workaround. But the secrecy around the un-official tool caused a funny situation. When nobody could compile in Bangalore for a week, the top boss was all up in arms. It was another matter that Madras team hadn’t compiled even once. So she called all the key people in Bangalore to discuss the matter. Since I was 2nd in command of my team, I too was invited.

She: This is a serious problem, why is this happening?
<couple team leaders gave a brief summary and turned it over to me for details>
Me: It is because of blah, blah, blah...
She: How can we fix it?
Me: It will take time, we are looking into it...
She: Why don’t we consult somebody about this?
Me: Who?
She: Madras team.
Me: They don’t use this tool.
She: Why not bring in Mr.X from Madras team?
Me: How can he help?
<I am being very evasive because she doesn’t know the author of the tool>
She: Isn’t he in-charge of the template library?
Me: Yes
She: So call him.
Me: But this is non-template version
<I just don’t want to admit I am the author, especially since the tool isn’t working>
She: How different is it?
Me: Very different.
<All the other seniors were studying the walls, ceiling and table surface with great interest>
She: What if it is different, didn’t he develop the non-template version too?
Me: No
She: Bloody hell, who then developed the non-template version of the library?
Me: <sigh> I developed it... so I have to fix it.
She: <speechless>

The meeting ended with that. It was probably the only time she was so speechless and ended the meeting that way. After that she must have dug deep to find out all about it. However, I didn’t get into any trouble. She knew about the productivity that was achieved due to the tool, but didn’t know it was developed right under her nose.

A month later, the compiler team fixed the problem with the templates. After that the template team took some more time to compile and test their code before releasing it. Finally Madras team could compile, after 6-7 months of coding. They actually had hopes of beating us using the template version. They knew Bangalore teams had compiled and tested using non-template version. But what about migration to template version? Many jokes were made about us having to redo everything all over again.

But they just didn’t know the power of our tools. The day before the template library was released, I reverted my code generation to the earlier version. The one that generated template version of the source code. By morning every team member in Bangalore had the latest tool and re-generated the header files. Since the changes were only needed to header files, the hand written source files were left untouched. By 11AM, everybody had recompiled and fired the unit tests. By noon, we were done. Madras team was still looking at their makefiles.

After this second coup, many Madras teams showed interest in my tools and started using it. I had few other tools, but I don’t remember the details. By the time the project ended after 16 months, my various tools were invoked nearly 200,000 times, as reported by my version control tool. In this project, most of my coding was related to my unofficial tools. I set the requirements, did the design, and developed using the language and tools of my choice. I spent another two years in HP, and I continued the same habit. When I was put in-charge of a testing team, I created a development project within the team to create a distributed testing tool using C++ and CORBA. So I got to do C++ programming even while running a testing team. By then all the managers had stopped questioning me on such matters.

Can this be replicated? Yes.

Last year, one of my best programmers quit the company. He is one of the few who understands Boost C++ library and could use it well, apart from Java and Python. And after 4 years of working with me, he had become a very good generalist. But his new employer uses only C. Asking a Boost C++ programmer to use C is akin to chinese torture. When he met me couple months later and cried about it, I revealed my old bag of tricks. So he started looking around and found a failed testing tool project within the group, that was badly needed done. He offered to do it using C++ and they accepted it. Thus he went ahead and did that very complex project using C++. So everybody was very happy and impressed. Next he demanded to use C++ in his main project, and they approved it. Now they hold him in very high esteem, all because he decided to change his situation.

Moral of the Story: You can get your way, provided you can demonstrate irresistible value proposition.

Originally Posted by Eddy

Samu san : A candidate does not know what a century is, when the last millennium started and does not know which century he lives in etc; but has done a fantastic project. Does he become more eligible for your kind of engg roles ?

That is like saying a guy who is bad in English grammar has written a fantastic book. It doesn't compute. It is highly likely that he will get caught when we start questioning him on the project.

Originally Posted by amitk26

If a student has written something as final year project in CV it becomes starting point of conversation. Usually for a fresher interview revolves around these achievements. If a candidate is able to explain the objectives and how he realized in sufficient details including both theory and his realization then he/she should be selected.

Absolutely. This is the most important part of fresher interview. People who didn't really do their project can be caught almost instantly no matter how much they prepare.

[zoombiee]

@ Samurai - I agree with you on this. I agree that if one wants to do something, he/she can do it by being creative. However, this is a minority population. Most people in the industry are not like you or your programmer. The industry in general views all of us with the same experience with the same eye (excluding certain startups). So in general what I want to say is that people who are incompetent in their college mostly remain incompetent in the industry. The ratio of competent:incompetent remains the same. I will again iterate, that finding work in the industry that one really likes to do is extremely hard to come by and also requires an element of luck. Some of the questions that I have been asked when interviewed in the industry have been plain ridiculous.

[Samurai]

Originally Posted by zoombiee

However, this is a minority population. Most people in the industry are not like you or your programmer.

I could have just said Be the change you want to see, but it has become a cliche. Besides, it doesn't give any ideas. Hence the long story.

Originally Posted by zoombiee

The industry in general views all of us with the same experience with the same eye (excluding certain startups).

It is up to you to change it. If somebody tries to look at you with the same eye, poke it, lightly Make them jump back and see you with a new eye.

Originally Posted by zoombiee

So in general what I want to say is that people who are incompetent in their college mostly remain incompetent in the industry. The ratio of competent:incompetent remains the same.

Well, we can't do much about them. If they are in your team, you can try to raise their level, if possible.

Originally Posted by zoombiee

I will again iterate, that finding work in the industry that one really likes to do is extremely hard to come by and also requires an element of luck.

Start looking to add value, you may get lucky.

[DerAlte]

Originally Posted by zoombiee

... in general what I want to say is that people who are incompetent in their college mostly remain incompetent in the industry. ...

Untrue / incorrect. Competence in college is only about being able to regurgitate the answers to the questions asked, since the questions never really give the latitude of finding your answer.

Some brains work differently - they are bored very soon with canned information and and prejudiced answers. They are thoroughly dissatisfied with the lack of challenge (learning by rote and regurgitation is not challenging for them). They are hardly incompetent. Most of them really excel in the real world as they get a wide variety of challenges which gives them ample scope to apply their intellect / power of analysis.

The reverse happens more than the above case. Students who learn by rote reach their limit of competence by the time they get a job.

[srishiva]

Man, the replies to this post keeps running...

How about employability of economists, businessmen, doctors and any other field. Isn't it prevalent in every filed in India?

[zoombiee]

Originally Posted by DerAlte

Untrue / incorrect. Competence in college is only about being able to regurgitate the answers to the questions asked, since the questions never really give the latitude of finding your answer.

Some brains work differently - they are bored very soon with canned information and and prejudiced answers. They are thoroughly dissatisfied with the lack of challenge (learning by rote and regurgitation is not challenging for them). They are hardly incompetent. Most of them really excel in the real world as they get a wide variety of challenges which gives them ample scope to apply their intellect / power of analysis.

The reverse happens more than the above case. Students who learn by rote reach their limit of competence by the time they get a job.

Competence is not about scoring marks or answering questions. Its about ones ability to solve problems logically. People who score good grades in college may not necessarily be competent.

[DerAlte]

Originally Posted by zoombiee

Competence is not about scoring marks or answering questions. Its about ones ability to solve problems logically. People who score good grades in college may not necessarily be competent.

Err... aren't you contradicting yourself here? ( happens to everyone)

* You earlier said "people who are incompetent in their college mostly remain incompetent in the industry". How did you define 'competence' then?

* Solving problems 'logically' is out of scope / 'out of syllabus' in colleges in India, right? Or did you mean that competent people solve problems logically, even if they arrive at the wrong answer / solution?

Originally Posted by srishiva

.. How about employability of economists, businessmen, doctors and any other field. Isn't it prevalent in every filed in India?

What exactly did you want to say / know with that question? Yes, in many ways it is true in all other fields too.

BTW, employability is not an issue with neither 'Businessmen' nor 'Doctors' - and only Govt. needs Economists. Fresh-out-of-college Doctors somehow manage not to shorten the life of their patients when treating them. Comparatively, fresh-out-of-college technical graduates would either
- freeze to inaction (simile: the patient would die from doctor's inaction),
- demand that they be trained (simile: "Please teach me how to treat a patient") or
- manage to destroy whatever they were told to create (simile: the patient dies from their action)!

[zoombiee]

Originally Posted by DerAlte

Err... aren't you contradicting yourself here? ( happens to everyone)

* You earlier said "people who are incompetent in their college mostly remain incompetent in the industry". How did you define 'competence' then?

* Solving problems 'logically' is out of scope / 'out of syllabus' in colleges in India, right? Or did you mean that competent people solve problems logically, even if they arrive at the wrong answer / solution?

What I meant was this -
1 -> People are inherently incompetent. Their problem solving ability is not good
2 -> People institutionalize incompetence over a period of time. Its either in their nature or due to external constraints. Reasons for this are complacency, lure of low hanging fruits, sheer laziness, inertia ....

Most of such people will continue to remain incompetent for the rest of their lives. Solving problems is not out of scope. Its a part of the curriculum, but there isnt much weightage given to it in the grand scheme of things (grades). You can choose to ignore it or take it on.
You may sometimes arrive at wrong answers or solutions by using some basic logic, if the understanding of the core subject is less. Understanding the subject should eradicate that problem. If there is inherent lack of problem solving skills, nothing can be done.

Not sure where the contradiction lies. I think its more of a misunderstanding. I think we could continue endlessly over this. Lets agree to disagree (not sure if there is a genuine disagreement as we seem to have a misunderstanding).