Java and .Net both a disaster: research

[Piyush Daiya]

Hi,
I just got this article ( http://www.zdnetindia.com/techzone/trends/stories/70597.html ) thro some friend.
Interesting thoughts.
Wondering how many enterprise development architectures are based on Java currently and how many will be?
Piyush

[John Fontana]

I thought that this article was a bit strange in that it did not explain the statistic. Why did it stress that outsourcing was the answer? Was it only internally developed systems that failed? It did not mention this, yet is was implied...

[Michael Bronshteyn]

B]"The only practical way to mitigate the risk [of a failed implementation] is to outsource development." [/B]
Since this artical is written on "zdnetINDIA", makes perfect sense why outsourcing is a solution.
What is not clear if both Java and .Net failed and outsourcing is solution, what platform it is going to be written on? But I guess it does not matter, just ship the project overseas.

[Mark Fletcher]

My take on that article,
It was absolute rubbish.
Ok first of all the statistic of "70%" of Java programs being failures. I refer you to an article by W. Wayt Gibbs, "Softwares Chronic Crisis", Scientific American (International Edition) pp 72-81, September 1994. Some statistics from the article:
1) On average, large projects take 50% longer than they were planned to do.
2) Three quarters of large projects are operational failures.
3) One quarter of large projects are canceled.
These three statistics highlight the risks prevalent in developing large complex applications. As the article was written before Java was even out, Id make the conclusion then that the figure quoted in the zdnetindia article isnt much of a surprise. But at the same time its clear that the failure rate is largely language independent. Im sure that other languages/platforms will suffer similar failure rates. Singling out Java is rather unfair.
The phrase "The only practical way to mitigate the risk [of a failed implementation] is to outsource development." Is total crap.
It doesnt matter if you develop in house or if you completely outsource your development, your still opening yourself to the same risk.
And Id argue that outsourcing would be even riskier in the longterm.
Consider the following scenario. Youre a CTO for a Dot COM thats clinging on by its bare teeth. The promise of lower costs due to outsourcing is too good to resist, so you fire a portion (or all of them, hey its increasing shareholder value were talking about here!) of your inhouse developers and contract out a critical portion of your product (say a shopping cart application, or an TCP/IP stack for an embedded device that youre planning on selling).
And then, the worst happens. You get the code back from the offshore developers. It doesnt work to your specifications. Worse still, its buggy as hell.
Ok, so send it back to the contractor to get it fixed. Ah, but the contracter has a few other contracts on hand from a bunch of other companies and cant spare the manpower to fix it *straight away*. Maybe next month.
So what do you do? Pass it out to another outsourcing firm? Doesnt matter how talented those developers are or how long you can make them work for as little as possible, as long as possible. Its still going to take them time to ramp up, to get used to the application and its source.
All the while the CTO's watching the savings he made disappear in lost time, in failure to market. If he had kept those developers in house, with their familiarity of the application in hand, he might have avoided it.
If you ask me, I wouldnt be surprised if our "correspondent" on zdnetindia has some stock in Wipro.
Cheers,
Mark

[Mark Herschberg]

I was going to post, but Mark Fletcher took the words right out of my mouth (or keyboard as the case may be :-). Well said, Mark.
--Mark

[Harpreet Singh]

Wow, that is some logic!
Since the project will fail regardless of the language you should outsource it. Let's see what's a good analogy ... since most fresh produce can go bad, vegetables should be grown in Russia.
I must say some of the most illogical people work in the logic industry!

[Piyush Daiya]

HAHA Harpreet,
By any chance r u Harry from SAKEC?
Its interesting to get above comments regarding article.
I would like to point out, few things:-
1) The article does appear in zdnetindia but it is not written by an Indian, if you would check it is written by Angus Kidman, ZDNet Australia.
2) Everyone has their own way for looking at things.So, it wouldnt be right to jump to conclusions.Like, it is mentioned outsourcing the development, in the sense that move it to people who are experienced in that particular domain and field and not necessarily for cheap implementation.
3) As both technologies have their shortcomings, so none of the technologies will be leaders.Thats what the article says.--->Despite those problems for early adopters, Gartner is predicting that by 2005, the battle for enterprise development supremacy will be a neck-and-neck two horse race, with Java and .NET each commanding around 40 percent of the market.

Arent we going overboard with outsourcing fear.It is not just IT work that is getting outsourced, there are many other works too.What about medical transcription?It has been going on for years.WEll, i would like to point to another article and yes it is on zdnetindia and is written by an Indian.
http://www.zdnetindia.com/news/national/stories/366,70849.html

Piyush

[Mark Herschberg]

Originally posted by Harpreet Singh:
Wow, that is some logic!
Since the project will fail regardless of the language you should outsource it. Let's see what's a good analogy ... since most fresh produce can go bad, vegetables should be grown in Russia.

I believe the break in logic is in your argument.
Empirical studies have shown a high failure rate in software projects, and larger projects have a greater chance of failing. We continue to undertake projects because the economic benefits make them viable despite the risk.
Mark did not argue that because projects tend to fail, they should be avoided, or undertaken with the worst technology. Rather, he seemed to conclude (as do I), that the report shows only that neither technology gives you an edge, that is, significantly changes your chance of success.
I would however, argue that these technologies are, in some sense a success. Even if our chance of success hasn't changed, with newer technology, the scope of the projects have increased. That, to me, is an improvement.
--Mark

[Mark Fletcher]

Originally posted by Piyush Daiya:
HAHA Harpreet,
I would like to point out, few things:-
1) The article does appear in zdnetindia but it is not written by an Indian, if you would check it is written by Angus Kidman, ZDNet Australia.

Piyush,
Point taken, but remember these figures were created by a Gartner "analyst", and to be honest based on the track record of "analysts" particularly in the last few years Id be wary.
If these analysts predictions are so accurate, why didnt they fortell the current recession?

2) Everyone has their own way for looking at things.So, it wouldnt be right to jump to conclusions.Like, it is mentioned outsourcing the development, in the sense that move it to people who are experienced in that particular domain and field and not necessarily for cheap implementation.

I guess thats one way of looking at it. However, how can you be 100% sure of the experience of Developers at the company youre outsourcing to? Ok, sure you relatively safe with well known companies such as Wipro. But for other companies you cant be sure.
Furthermore, arent the native developers that are being replaced experienced enough?

3) As both technologies have their shortcomings, so none of the technologies will be leaders.Thats what the article says.--->Despite those problems for early adopters, Gartner is predicting that by 2005, the battle for enterprise development supremacy will be a neck-and-neck two horse race, with Java and .NET each commanding around 40 percent of the market.
Arent we going overboard with outsourcing fear.It is not just IT work that is getting outsourced, there are many other works too.What about medical transcription?It has been going on for years.WEll, i would like to point to another article and yes it is on zdnetindia and is written by an Indian.
Piyush

Piyush,
To be honest I dont really fear Outsourcing as a concept. I recognise that there are a lot of skilled developers in India who can do the same job for a lower price. There are cases where it makes sense to outsource. For example where you have a lot to cover and not enough manpower to cover it all internally.
What I do fear though is that Outsourcing could be viewed as a magic pill for solving all your IT problems. We need to step back from the whole situation and make a reality check. In the current climate, I fear that CTO's may be outsourcing *too* much in the desire to cut costs, and in the end this might create some serious repurcussions.
Ive only been in IT for about five years now, but in that time Ive come across many instances where development has been outsourced to companies either in the same country or overseas. There have been instances where this outsourcing has made sense, and there are others where it has brought more bad than good. Where you outsource to is irrelevant. Its what you choose to outsource that is of critical importance.
Best Regards,
Mark

[Fred Grott]

Originally posted by Piyush Daiya:
Hi,
I just got this article ( http://www.zdnetindia.com/techzone/trends/stories/70597.html ) thro some friend.
Interesting thoughts.
Wondering how many enterprise development architectures are based on Java currently and how many will be?
Piyush

ah hmm you need to read better articles with aactual facts..
Try the Computing Failure series by Robert L Glass..
You will find that most computer project failures have nothing do with programming! So the outsourcing is the only answer is not only smells bad ti does nto even attempt to state the true orginal problem in computer project failures..
In fact the same computer project failrues come up whether its java, MS .NET, python, Perl or etc..

[Harpreet Singh]

Mark Herschberg,
I think the article on zdnet was total BS! Just because a certain technology is vulnerable you should outsource the project?
All,
Outsourcing is an entirely different discussion. My personal opinion is that most corporations are just jumping on the bandwagon without considering the long term consequences. It may work well for a large majority of projects but in some projects it may come back to haunt you.
But I think we have all beaten the outsourcing topic to death.
All techies need to realize two things.
1. Technology career is a lifelong treadmill.
2. There is always going to be someone cheaper.
Trying to figure out niches in technology is not going to help. Sooner or later the niches will disappear. DBA and sysadmin jobs were considered safe haven but they are also disappearing. I know of a DBA in India who manages databases all around the globe. In fact I know Indian companies are considering outsourcing to Manila!
At the end of the day, the choice is yours, you can choose to run on the tech treadmill or you can choose to be on the business side(aka push paper)!
Just my $0.2

[Mark Herschberg]

Harpreet Singh,
I'm not sure if you misunderstood me and were trying to counter my argument, or just adding your own views, but we seem to be in violent agreement. :-)
--Mark

[Harpreet Singh]

Mark,
Its the latter. We are in violent agreement

[Jason Menard]

Originally posted by Piyush Daiya:
Arent we going overboard with outsourcing fear.It is not just IT work that is getting outsourced, there are many other works too.What about medical transcription?It has been going on for years.WEll, i would like to point to another article and yes it is on zdnetindia and is written by an Indian.
http://www.zdnetindia.com/news/national/stories/366,70849.html

I read that article. It's simply the projections of some analyst and you can't say for sure one way or the other what will happen between now and 2015. While this current trend is not good for the US worker, it is potentially disasterous for the economy of India should there be some reason Western companies decide to shift their outsourcing elsewhere. One reason that might happen would be if the region were to further destabilize politically. A war between India and Pakistan, for example, would very likely destroy India's tech industry as foreign customers get nervous and move their projects elsewhere. There was already some evidence of this earlier this year when tensions were very high over Kashmir.
It seems that some of these nations, India in particular, are willing to place most of their eggs in one basket (foreign contracts for IT support in this case). Unfortunately that basket is resting on a house of cards which could come tumbling down at any time.

[Pakka Desi]

Originally posted by Jason Menard:

While this current trend is not good for the US worker, it is potentially disasterous for the economy of India should there be some reason Western companies decide to shift their outsourcing elsewhere.

That is correct but there is no option. We've got to start somewhere. Local s/w market cannot provide enough incentive for the local industry to even think of learning and developing new things. But the s/w services exports is giving a very good head start. Once the ball start rolling, there are opportunities.

It seems that some of these nations, India in particular, are willing to place most of their eggs in one basket (foreign contracts for IT support in this case).

That is not correct. We are not willing to put *all* eggs in one basket. We have only one egg so it's got to be in one basket. But as I said, it's a very good start.
In any case, until US companies stop outsourcing completely (which is next to immpossible) or outsource from somewhere else (which is very much possible because of comptt. from China), it has provided a very good opportunity to earn money.

[Kyle Brown]

Originally posted by Pakka Desi:

That is not correct. We are not willing to put *all* eggs in one basket. We have only one egg so it's got to be in one basket. But as I said, it's a very good start.
In any case, until US companies stop outsourcing completely (which is next to immpossible) or outsource from somewhere else (which is very much possible because of comptt. from China), it has provided a very good opportunity to earn money.

This is just my own opinion, but I think that India has two things going for it that make China overtaking India as an outsourcing location unlikely:
(1) English is not only quite commonly spoken in India but very well spoken -- a legacy of the British empire -- the average CEO would probably find it easier to converse with an English-speaking Indian than an English-speaking Chinese person of roughly equal educational background.
(2) India has two other legacies of the British Empire -- British common law and parliamentary democracy.
These two together might tend to make companies who are doing business over the long term more comfortable in India than China.

[Randall Twede]

i wont even bother reading it. it is clearly propaganda.