Disclaimer: I'm an Indian and I've worked on offshore developments years ago. I work in product development now and there's no way in hell I'm going back to outsourced projects but I'll admit to some bias.
Outsourcing works. Most reputable outsourcing companies are looking for long-term relationships and they are usually very serious about feedback and customer satisfaction. However, like anything else, it takes work and research. You wouldn't hire a single developer without rigorous interviews and even so it would take a while before he/she was up to speed. How then can you expect there to be no overhead with the 5-10 outsourced developers?
I've personally worked on projects for Yahoo (US and UK) and Merck and my experience was that the biggest problems were always at the beginning - and they usually were communication problems.
What I'm saying is successful outsourced projects tend to work things out as the teams adjust to each other. In no case will it work if you just throw some documents at each other - there needs to be face-to-face and key people need to know each other in any team. The payoff is, of course, the work then continues at a much lower cost.
Outsourcing may not be for everyone, but it does work well in a lot of cases. Just my two cents.
I've heard nothing but negative experiences from those who outsourced to India, or anywhere remote for that matter. Could it be that the author is so gun-ho about it because he has a consulting outfit in New Delhi?
I think it mostly boils down to bad hiring decisions. People -- mostly nonprogramming managers -- said "hey, these guys are just as good and only cost a third as much" and hired more or less indiscriminately. (As with any other group of people, some Indians are smart, some less so. Some are passionate about what they do; others view it as just a job.) So they ended up with some guys who were indeed as good at one third of the price, and many more who weren't. They got the same result, of course, as what they'd get hiring indiscriminately in wealthier countries. I don't think companies with good hiring cultures have these problems, in India or anywhere else.
There are other problems with outsourcing to somewhere far away (which India is for many people): people will have stay up late or start very early to attend meetings, language can be an issue, and matters that are simple but take a few rounds of communication can end up taking days to resolve.
Those problems can all be managed with smart leadership, but when that's lacking, it's easier to point fingers at people with little chance of fairly defending themselves.
Here's how it works: Joe Manager goes to Silicon Valley and sees all these incredibly talented Indian engineers and thinks wow, I gots to hire me some of them.
What he fails to understand is that an Indian entrepreneur in Silicon Valley is not really representative of Indians as a whole, any more than a pro athlete is representative of the fitness of the average American, or a Hollywood A-lister is representative of how the average person looks.
In fact the quality of engineer you will get outsourcing to India is lower than the average, because the most talented engineers (in any country) aren't really interested in working for huge companies on other people's dull corporate applications. They're busy working on their own startups or at least for small companies doing interesting work, whether in India or in the US.
I worked for a company that was given a free 1 month trial of 1 dev guy out of india. We had a custom PHP framework (http://www.sinapse.org/), and he jumped in and started pumping out code pretty quick. When you looked at his pages, he was "accomplishing" his tasks, but his code was so inconsistent and completely outside the framework (it was all one-off-spaghetti-code-hell), so at the end of the month we had to throw everything out. We didn't outsource again.
I would add the only reason we tried it in the first place is that it was a free offer. So we weren't really "serious shoppers", and did not shop around.
Indeed. There are bad coders everywhere. The problem with outsourcing is generic: you're hiring consultants, not hackers. There are "good" consultants out there, but even the best of them are basically bean counters.
And no doubt there are plenty of great Indian hackers out there, but they're not on the consulting market.
In many respects the developer masses in India go through more rigour than the developer masses in many countries like UK/US, essentially due to the increased competition and overall standards in general in technical education there.
However, even in India developers/firms have been gearing towards work in markets other than enterprise Java and .NET for a long time now, as the competition becomes stronger all around in those fields and first mover advantage has generally been lost/taken by the early players.
Many realise the niches (such as bespoke RoR/Python/other work for smaller clients and individuals) remains largely untapped and profitable. We will soon see more impact by smaller independant Indian firms in these areas and as tech startups in general.
It's hard to get good data, but I've read the opposite - that educational standards for engineers may be much higher in the US than in India and China.
Here's a link to a business week article by Vivek Wadhwa, a prof at Duke:
We've had... I'd call it a moderate success with Ukraine and now things are getting better and better.
The reason why it works for us is because we've pretty much trained those guys from the beginning to work a certain way: they didn't do outsourcing before (servicing local market) and they haven't worked with anyone else but us.
So yes, it can be done and it does help enormously, especially if founder(s) cannot afford to code full time.
This article has some good advice on the 'how' part of working with Indian software developers. Regarding the 'whom' part, really it is the same as everywhere, find good developers.
I happen to know the Indian folks whom the author got on board his firm. They actually are 'very' good, and i guess that is the primary reason why the author is so happy at having it all figured out.
The point that is a bit off the mark in the article is:
Because of the fairly fluid and competitive labor market in the IT world in India right now,
many Indian employers don’t allow their developers to blog or make contributions to open-source,
for fear of losing them to competitors who might contact them directly
I am yet to hear of any employers here not allowing open source contributions per se. I guess one of the reasons for not having as many open source contributions/technical blogs from India (sadly) is that most of the business is still 'services' and people are jammed up with unrealistic schedules (committed by their firms). Also majority of the work is Java/.NET enterprisey stuff that is not too encouraging for the people to put more efforts back in to 'that' ecosystem.
One report I heard is that the rate is about one third of an equivalent UK person now. It used to be a much greater difference, but the Indian rates are going up as the developers get paid more.
I wouldn't, becuase i was joking, and after the massive downmod i figured i'd rather not muddy up the conversation more by trying to explain myself.
this was a joke... about how people outsource their work to other countries because they are too cheap to pay for developers in their own countries. sorry i forgot that the internet was serious business all the time.
Outsourcing works. Most reputable outsourcing companies are looking for long-term relationships and they are usually very serious about feedback and customer satisfaction. However, like anything else, it takes work and research. You wouldn't hire a single developer without rigorous interviews and even so it would take a while before he/she was up to speed. How then can you expect there to be no overhead with the 5-10 outsourced developers?
I've personally worked on projects for Yahoo (US and UK) and Merck and my experience was that the biggest problems were always at the beginning - and they usually were communication problems.
What I'm saying is successful outsourced projects tend to work things out as the teams adjust to each other. In no case will it work if you just throw some documents at each other - there needs to be face-to-face and key people need to know each other in any team. The payoff is, of course, the work then continues at a much lower cost.
Outsourcing may not be for everyone, but it does work well in a lot of cases. Just my two cents.