[amitk26]

Originally Posted by anujmishra

OT: Only few system might be using long long for clock ticks, but many including linux kernel does not use it. That is why we are expecting Y2038 issue. One fine day in Y2038 we will move to past.

Not Linux Kernel but IIRC other RTOS based system and the internal platform with which you are more familiar then me used long long but I may be wrong.

[msdivy]

Originally Posted by DerAlte

The problem is, we are so starry-eyed with the 'billion dollar' success story of the "IT services" that we fail to realize the size of the other market: trillions of dollars. And we haven't even scratched the surface of the market with our short-sightedness

Well, still the IT services is billion dollar market. Glad that, that share of market is captured and gets us revenues. So I wouldn't say short-sighted.

Asking services company to do products is like asking trucking company to run cab services, since both involve driving. But both have different focus & approaches. Product development requires more insight into the market. USA is still the primary market and they have 20 year head start. Hence they rule when it comes to products.

Note: Eastern european countries produce excellent programmers. But those countries hardly produce any products. It is more to do with markets & marketing and USA excel in this aspect.

[tsk1979]

Originally Posted by msdivy

Well, still the IT services is billion dollar market. Glad that, that share of market is captured and gets us revenues. So I wouldn't say short-sighted.

Asking services company to do products is like asking trucking company to run cab services, since both involve driving. But both have different focus & approaches. Product development requires more insight into the market. USA is still the primary market and they have 20 year head start. Hence they rule when it comes to products.

Note: Eastern european countries produce excellent programmers. But those countries hardly produce any products. It is more to do with markets & marketing and USA excel in this aspect.

Well if you are in it for 20 years, its great, but if you are in for the long haul, services is something which will go to the lowest bidder. What happened in BPO/KPO? Philippines is currently whipping India. How long before the level of English in China improves.
Service is a dangerous game to bet on esp if you got the contracts in the first place due to lower costs.

[Samurai]

Originally Posted by anujmishra

Later part, I was discussing about whole performance not about long long.

I was just trying to say that, renting a big wearhouse is not necessary for parking Maruti 800. Hope you got my point.

So you are referring to the part where I said I let compiler do such optimization. Somewhere in the 90s I was convinced by some compiler writers (in some tech forums) that such manual optimization is completely unnecessary. A commercial grade compiler can do more much efficient job at that. After that I focused on design optimization, and didn't worry about manual code optimization.

Originally Posted by anujmishra

Well, I did not write any code here, to prove my programming ability.

I was not planning to write until I saw a non-programmer throw down the gauntlet.

Originally Posted by msdivy

Note: Checking the assembly of either of code blocks (for instance here in http://assembly.ynh.io/ in GCC), the code without ternary operator produces 8 less assemble instructions !
So the moral is give preference to code readability & extensibility before other things.

I find your ternary version code easier to read, probably because of messed up indentation.

But this brings up an important aspect. When we say readable code, who is our audience? Shouldn't we set some lower limit? If we try to make all the code readable by Forrest Gump, then we can't use some of the strongest features of the language.

I don't try to simplify it so that a newbie can read it. Not if it comes at the cost of losing the ability to use advance features of a language. For example, anybody who wants to read my code, has to be familiar with Boost C++ and some standard design patterns used in the company and the technology domain like IP Telephony. Then he/she can read my code like a book.

However, even the strongest C++ programmers I have hired right out of college, were not familiar with STL, let alone design pattern or Boost C++. [See, we are back on topic] Therefore, they take a year of training before they can look at our company C++ code and start making sense out of it.

[DerAlte]

Originally Posted by msdivy

Well, still the IT services is billion dollar market. Glad that, that share of market is captured and gets us revenues. So I wouldn't say short-sighted. ...

That's precisely the self-satisfied self-serving short-sightedness that I was talking about. Perhaps you fail to visualize that NOTHING of the achievement you are talking about was planned. It is a phenomenon, a HAPPENING that everyone tom-toms as achievement. Much like iron ore exports (both the +ives and -ives) or the brain drain.

Nothing has been done FOR it (planning, facilitation) the way China, Philippines etc. have done. Nothing has been done to stop the things going AGAINST it. Again, quite unlike what China has done. And China is already the WORLD'S vendor for outsourced manufacturing.

Originally Posted by msdivy

... Product development requires more insight into the market. ...

Outsourced Product Development (developing someone else's product), not Product Development (making your own products). India hasn't made many software products either, despite all the hullaballoo about IT services.

Originally Posted by msdivy

... Note: Eastern european countries produce excellent programmers. But those countries hardly produce any products. ...

You haven't SEEN any, right? Understandable why - you are not living there.

Originally Posted by tsk1979

... Service is a dangerous game to bet on esp if you got the contracts in the first place due to lower costs.

Correct. One can't survive on Labor Arbitrage alone. China doesn't get contracts based on low rates. China delivers completely finished products, including taking care of logistics. It is incidental that their rates are low and there is no competition.

Indian IT services companies run to the original software maker (Oracle, SAP, whoever) to get it installed and configured, and THEY are the ones who were given the responsibility to commission it. Why? Because there is a multi-year Maintenance contract to be won, even if you lose money on the original contract. We are the world's janitors in IT services, not architects or anywhere higher on the value chain. We get paid commensurate. And we should be happy with that?

[msdivy]

Originally Posted by tsk1979

Well if you are in it for 20 years, its great, but if you are in for the long haul, services is something which will go to the lowest bidder.

Thats true and the respective IT services company must worry or plan their future.

Originally Posted by DerAlte

Outsourced Product Development (developing someone else's product), not Product Development (making your own products).

There are plenty of companies in india which do outsourced product development. According to http://www.nasscom.in/, in 2012

• IT services exports = USD 40 billion
• Offshore Software Product Development (OSPD) = USD 1.2 billion
• Software product exports = USD 1.5 billion

So there is some product development business in India.

You haven't SEEN any, right? Understandable why - you are not living there.

Any example give many folks here an idea to pursue.

We are the world's janitors in IT services, not architects or anywhere higher on the value chain.

The money is huge to be ignored in this janitor business. There are many architects & domain experts here and their numbers will only increase. I am hopeful that they make the products business share bigger.

[vivekiny2k]

Originally Posted by Samurai

I hope you are not saying my code is not readable. Any good C/C++ programmer should be able to read it effortlessly. The simple fact that amitk26 and I came up with the same logic indicates that it is neither complex nor rare.

No that was not in response to your post, but to say that a lot of times most efficient code does not pass the first line of review.

Originally Posted by anujmishra

I absolutely not agreeing with this. In most of the business performance matters to extent that performance would be the main criteria for marketing. One thing is just to make it work and another is make it work better than benchmarking data.

Also, regarding readability, if you hop through some opensource projects code (e.g linux kernel code), you will not find any of their code readable. But this is also skill, which comes with experience, what I feel.

I don't either, but it's difficult to argue for a system that will be maintained by who knows years later. However, the hardcoding may have actually improved the efficiency in my case as long as one can be fairly confident that there won't be more than 5 warehouses in future (there were actually only 4 at the time).

Code like a kernel is a different story because you don't have everyday programmers needing to making changes because of a FCC regulation for example.

[amitk26]

Originally Posted by DerAlte

Outsourced Product Development (developing someone else's product), not Product Development (making your own products).
<snip>
India hasn't made many software products either, despite all the hullaballoo about IT services?

Well let me assure you there are plenty of outsourced product development here but when some one outsources the product development it is actually a service rendered to the entity who outsources the task.
Lots of MNC centres here are developing products for parent entities but most of them are cost centres the profit centre is either at the HQ or in the region where the major customer lies. True value addition is never reflected for cost centre as billing is time and materials based even in case of fixed billing the estimate is arrived based on effort and infrastructure/support overheads and not on the market value of the product

In my books if product development is done here , marketed , branded here then it qualifies to be called Indian product and the examples are very few and I already eloberated examples.

[Samurai]

It is hard to talk about employability of technical graduates without alluding to the technologies or industry they will be working at. As anyone who has followed this thread since 4 years is aware, here we discuss all that. Therefore, the thread title is slightly modified to reflect that fact. - The Support Team

[msdivy]

Originally Posted by Samurai

So I wrote another tool to generate source code from a definition (meat 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.

Thats is a great process innovation story. Thanks for sharing !

When you undergo campus recruitment, you generally have no control over which company you join.

Similar thoughts are echoed by Paul Graham in his essay - A Student's Guide to Startups. Every fresh grad oriented towards programming must consider this option.

Originally Posted by Samurai

However, even the strongest C++ programmers I have hired right out of college, were not familiar with STL, let alone design pattern or Boost C++. [See, we are back on topic] Therefore, they take a year of training before they can look at our company C++ code and start making sense out of it.

The idea is if the code is simple to understand (like reading English text), then an intern can be productive much earlier. But again it depends on the type of the project.

[ampere]

Another thing I noticed is in most cases (not all), learning stops once you start working. Some how many differentiate learning and job. On the job learning is mostly restricted to learning of tools which would facilitate the work. (Of course that is also needed)

I rarely come across those who are OK at their advanced stages in career to get down to soil their hands. By this I mean not learning new tools, but working on updating basic concepts and follow up on latest research happening in their fields. May be it is not needed in their field of work and may be jobs may not encourage it as well. But what I wanted to convey was; thanks to environment, the appetite to learn goes down slowly. One needs to be conscious of this slow down and work on it constantly.

[DerAlte]

Originally Posted by msdivy

... There are plenty of companies in india which do outsourced product development. ... So there is some product development business in India. ...

What was the ratio of exports? Insignificant, right? And, mind you, these figures include Captive Centers - Indian-registered companies who are subsidiaries of large product companies like Cisco, GE, etc. Even there the main work is maintenance!

Originally Posted by msdivy

... There are many architects & domain experts here and their numbers will only increase. I am hopeful that they make the products business share bigger.

I am hopeful too, as you are. Except that the relevant faculties are atrophied in the same architects and domain experts after doing low-value assignments day in and day out. The ignorance of basic issues in product design is horrifying. Not that there are no world-class designers / architects in India - the main problem is "too few", not enough to go around!

Originally Posted by amitk26

Well let me assure you there are plenty of ... actually a service rendered to the entity who outsources the task. ...

Originally Posted by amitk26

... most of them are cost centres the profit centre is either at the HQ or ...

From the p-o-v of Indian resources, the difference doesn't at all matter. Profit Center matters a bit - it increases the competitiveness, and hence the ability of the resources - if they know what they are doing.

My allusion was to (too few or none at all) true-blue Indian companies who deliver Product Development Services to clients in the outsourced model, as the main business. It doesn't matter to which country the Product is identified with. OPDS is a specialty generating higher value, and commanding higher rates. Ergo - higher business volume in this segment. Do we have the inherent abilities? I believe yes. We are just taking the easier route.

[amitk26]

Originally Posted by DerAlte

From the p-o-v of Indian resources, the difference doesn't at all matter. Profit Center matters a bit - it increases the competitiveness, and hence the ability of the resources - if they know what they are doing.

My allusion was to (too few or none at all) true-blue Indian companies who deliver Product Development Services to clients in the outsourced model, as the main business. It doesn't matter to which country the Product is identified with. OPDS is a specialty generating higher value, and commanding higher rates. Ergo - higher business volume in this segment. Do we have the inherent abilities? I believe yes. We are just taking the easier route.

It matters a lot Mr.DerAlte to the junior most engineer in the food chain.
Engineers come-up with really good ideas all the time and what we call as idea funnel is created but in cost centre model if a idea will be converted to a product or a feature in product depends on whims and mercy of person responsible in profit centre going to fund it. If they are myopic you stand no chance.

If you are a cost centre then you can create idea PPTs create basic prototype by moonlighting and need to market /beg for someone to fund the development. Once a prototype is thrown in dustbin it is really heartbreaking for the guy/gal who moonlighted to create it. The same idea may be purchased by same people from market at 2X price 2 years later.

In profit centre model you put your cost as per the market value of the product created and thus have more leverage to create the products, buy products from start-ups or simply pay people more and get better people depending on your priority.

It is not that India does not have profit centres just that general marketing demand is so low that these profit centres are not pushed to fund/buy ideas.
Look at Maruti-Suzuki it is biggest profit centre of Suzuki but what it does just pays inflated royalty to parent and it is does not fund internal R&D or buy technology locally.
Do you think a Suzuki subsidiary in US or UK could have survived like this ?
Answer is no they are already bankrupt in USA.

[DerAlte]

Originally Posted by amitk26

It matters a lot ... they are already bankrupt in USA.

I was avoiding discussing Cost Centers in the current context. It is known that Cost Centers are rife with internal politics and LLGT cock-eyed policies and implementation, but then that is human nature - even Profit Centers are affected by the same!

Negativity that comes out of such expressions and discussion vitiates the main discussion here, and is best avoided. It has nothing to do with Indian technical graduates. Fortunately, in international companies such things are nationality-agnostic stupidities.

Once again, I was talking of Indian companies providing Development Services specifically in the area of Product Development to other clients. A Service, not even a Profit Center. A 100 such small companies will benefit Indian resources and business more than 1 large Captive Center run in Profit Center mode. Large Chinese companies providing Manufacturing Services have come out of 100s of small entrepreneur driven Manufacturing Services providers.

There is a lot to learn there, provided we intellectually rise above petty issues whether one small idea was accepted by management or not. If the genuine merit of the idea is proven (and usually we arrogantly avoid doing it, like "Hey, I am holding a diamond, how in heavens are you not able to see that?"), the same management moves it's high-inertia posterior to monetize it.

[tsk1979]

I look at it from a POV of chicken and egg problem.
When a product company (cost center, profit center, left center, right center up center low center or whatever) which requires an engineer with 5 years of experience, or 2 years of experience, goes hunting for engineers it finds only "drones" which require a pilot.
So recruitment is minimal, and as a talent pool, currently Indian talent pool is good enough only for maintenance and repetitive jobs.
The IT industry has done a fine job of dumbing down the brightest minds.

Add to that an education system that works on "how you game the system".

So you get engg graduates good enough only for maintenance roles and simple repetitive tasks easily done by a non tech graduate.
In more mature economies, you get around 10-12% of technical graduates fit for "creation". Here its not even 1%.

If more product companies come to go to Campus it will slowly increase this ratio.
But when a new product company comes, first they look for experienced people, you cannot start a center with fresh graduates "only".

Due to the polluting of the Engg graduate ecosystem by the biggies, it is difficult.

Think of India has a large pond with very big fish.
The main fishermen supply to a burger making company and require fat, so they feed fat in the pond, and all fish become fatty.

When the protein shake company comes and tries high protein fish, there is no high protein fish left because they are all corrupted.

Only some very intelligent fish who can afford to stay hungary and know what type of food to eat stay high protein, and its difficult to find such fish.

Reminds me of supernatural season 7 or so where the leviathians started feeding fast food to the nation which made people fat and dumb so they could be eaten. I would call the scenario now the same.