The Joel on Software Discussion Group (CLOSED)

A place to discuss Joel on Software. Now closed.

This community works best when people use their real names. Please register for a free account.

Other Groups:
Joel on Software
Business of Software
Design of Software (CLOSED)
.NET Questions (CLOSED)
TechInterview.org
CityDesk
FogBugz
Fog Creek Copilot


The Old Forum


Your hosts:
Albert D. Kallal
Li-Fan Chen
Stephen Jones

Offshoring benefits?

Why would a company decide to off-shore their software development?  I understand the hourly rate is cheaper but is the cost for a specific project really that much more inexperience?  How much are you really saving once you take into account the time it takes to get something back that works, the overhead of managing the project and the rewriting of any pieces that weren't developed properly?  I just don't see a benefit with off-shoring.

Thoughts?

One a side note; If companies could find a way to off-shore the CEO function reducing that cost, would they?

Thanks
Dan Murphy Send private email
Monday, August 24, 2009
 
 
1.  Managers like LOTS of things that seem to have a cost-savings, seem to give them an illusion of something-for-nothing, seem to be "more reliable" than what they've been doing.  Off-shoring is simply one of the latest approaches that seem to work.

2.  To my knowledge, the seeming advantages of off-shoring have not occurred.  You don't get order-of-magnitude cost savings, order of magnitude quality or productivity improvements.

3.  And no, if the "off-shoring" process resulted in off-shoring their OWN jobs, they would probably not pursue it.

4.  Sadly, off-shoring the jobs the managers used to manage develops a cadre of off-shored managers to manage them.  Once they have one or two successful projects under their belt, why do they need the on-shore managers any more?  Answer, they don't, which is yet another unintended side-effect of off-shoring.
AllanL5 Send private email
Monday, August 24, 2009
 
 
It always struck me as slightly odd. You have two groups of employees:
One set are required to be completely focused on every tiny detail of your product, it's interaction with your business and your customers. They are relatively cheap and work long hours. They have specialist skills that are hard to find or measure.
Then you have another set that are board certified to be exactly as well qualified and interchangeable with any other board certified accountant or lawyer - they have spent 500 years developing systems that are completely decoupled from your business and will work equally well for any company. They are expensive to hire and cost you a fortune in golf club membership.

And which set do you offshore?
Martin Send private email
Monday, August 24, 2009
 
 
there will always be a maintenance cost Dan. even if the project was developed onsite.

if the total cost of offshoring + maintenance cost <= onsite development + maintenance cost, then cheap managers would be willing to go offshoring.

it might seem a little profit, but a profit nevertheless. as a manager you are judged by the additional beef you bring home.
victor a.k.a. python for ever Send private email
Monday, August 24, 2009
 
 
Good one, Martin! The trouble is that during the 500 years of process development, they were also figuring out other stuff. Among other things, they've figured out that controlled information is power and how to manage power structures for their benefit.
Ron Porter Send private email
Monday, August 24, 2009
 
 
No one reads history.

Remember the American revolution? England owned the land, directed the production, owned the ships, sold everything all over the world, but didn't actually *do* anything except own the whole setup and take the profits. Until the colonists said, "Fuck this, we'll keep it for ourselves."

England tried to say, "They hell you will!" but they were a little busy fighting a war with the French at the time. And the U.S. became a huge world power.

This is about what England did to the Dutch, and what the Dutch did the the Spanish before them. (http://www.thirdworldtraveler.com/Kevin_Phillips/Transformation_WAD.html The excerpt starting from p. 175)

Every great power eventually outsources the "dirty jobs", keeping management and ownership for itself. Until the people doing the work decide they don't need the foreign ownership any more. This isn't a new phenomenon.
Drew Kime Send private email
Monday, August 24, 2009
 
 
good points Drew.
victor a.k.a. python for ever Send private email
Monday, August 24, 2009
 
 
Except you guys are all focusing on the *bad* aspects of outsourcing - in reality true outsourcing, offshore or onshore, is not about doing the *same* thing cheaper; it's based on the efficient allocation of labour: you stop doing relatively lower value tasks and focus on those in which you generate higher marginal utility. This is truly a good thing.

The examples here are about outsourcing essentially everything but some modicum of control - which I agree inevitably leads to the "do-ers" taking over the process. BUT if I outsource something at which I'm competent, say accounting, but not as productive as say AI programming then (a) I do better, (b) someone else presumably good at accounting does better and (c) the entire system is more productive. Same goes for help desk, maintenance programming, etc. There are lots of problems and challenges with this process but not in its fundamental nature. What we need to work on is the "friction" caused by changes in the system; i.e. what happens to the people who's rolls get migrated? No one benefits if they go from *some* utilization to none (ex: unemployment). In my mind this is what separates good companies form bad ones.

This is essentially the foundation of capitalism and the efficient allocation of labour. You can debate if this is the *best* system (which many are following recent world turmoil) but if you believe in capitalism as (to paraphrase) a system that's the worse than everything but all the alternatives, this is what we've got to work with.
Mark Dochstader Send private email
Monday, August 24, 2009
 
 
"And which set do you offshore?"

Hmmm, which group is the one making the decision who to fire?
Scott Send private email
Monday, August 24, 2009
 
 
Mark, that should be true of *outsourcing*. But the question was about *offshoring*. And most companies talking about it start the conversation with cost savings.
Drew Kime Send private email
Monday, August 24, 2009
 
 
>>but if you believe in capitalism<<

There was a time when the US got practically all of it's income from tariffs, and the US was much more capitalistic then.

Capitalism within our own boarders is one thing, capitalism on a global scale is something else - other countries don't play by the same rules. For example, if a certain country has very protectionist economic policies, would it not be fair for the US to protection right back?

Furthermore, the present situation may be more like "dumping" than honest capitalism. Japan used dumping very effectively against the US in the 1980s (then Ronald Regan went to Japan and picked up a two million dollar "speaker's fee").

Think about it: other nations heavily subsidize the cost of education for their citizens. The US then hire those workers "below cost."
walter byrd Send private email
Monday, August 24, 2009
 
 
Good points Drew, Martin.

I would add that you can successfully offshore certain kinds of work - brain-dead simple reports (crystal, oracle reports, etc) and database monitoring come to mind.

Of course, when the work becomes so simple it's easily offshored, it's also a candidate for automating.

Just sayin'.

Another approach is to offshore or outsource spill-over work.  In other words, in the heat of the big ERP conversion, you don't want to hire employees because you won't have a job for them 12 months from now.  In that case, finding an offshore or at least outsourced partner may make sense.

Again, just me talkin'.
Matthew Heusser Send private email
Tuesday, August 25, 2009
 
 
"but not as productive as say AI programming..."

The reason we focus on the negative aspects of outsourcing is because of statements like the above.

If YOU are not "productive in AI programming", what makes you conclude that you can HIRE somebody off-shore who's "productive in AI programming"?  What about the entire chaotic state of Software Engineering makes you think that off-shoring or even out-sourcing is an effective strategy in the end?

Is there an evaluative criteria for company's effectiveness at software development?  Well yes, many, but are any really predictive of future success?

Is there a track record of successful projects completed?  Not to my knowledge.

If you focus on the positive aspects, and pooh-pooh those who are trying to raise issues because they're "being negative", then you'll duplicate the failures of those who've gone before you.

Just saying "it's better to pay others to do better what I can't do myself" is both a true-ism, and besides the point if you can't actually find somebody who CAN do it better.  Sure, you can find people who'll SAY they can do it better, but so far the RESULTS have not been promising.

That's why.
AllanL5 Send private email
Tuesday, August 25, 2009
 
 
@AllanL5:

Looks like you misread Mark's post.  What he said was:

"BUT if I outsource something at which I'm competent, say accounting, but not as productive as say AI programming then (a) I do better, (b) someone else presumably good at accounting does better and (c) the entire system is more productive."

So in other words, he's perfectly able to do his own accounting, but he gains more (is more productive) doing AI programming.  So, he hires someone else to do the accounting work, and because he's competent in it he can tell whether he's hired a halfway decent accountant.  If his additional productivity offsets the cost of hiring an accountant, then he should outsource.

The problems arise when companies outsource/offshore the very things that are tied to their competitive advantage.  That's when it comes back to bite you in the butt.
Justice Walker Send private email
Tuesday, August 25, 2009
 
 
Ah.  Good point.

So, if you outsource an identifiable skillset like "Accounting", then you should be alright.

Unless "Accounting" IS your core skillset, in which case you'd be shooting yourself in the foot.
AllanL5 Send private email
Tuesday, August 25, 2009
 
 
Evil Overlord List
http://tvtropes.org/pmwiki/pmwiki.php/Main/EvilOverlordListCellblockA

#179: I will not outsource core functions.
Drew Kime Send private email
Tuesday, August 25, 2009
 
 
Good point, Allan5 -

I did my MS Capstone on outsourcing, and the conclusion was basically "if you outsource your core competance or strategic differentiator, you're stupid."

It took more words than that:

http://www.xndev.com/CS/CS692/TheOutsourcingEquation_ABIT.doc
Matthew Heusser Send private email
Tuesday, August 25, 2009
 
 
"Three years ago, during my visit to India, the country was emerging as an IT superpower. Today, the country is handling the most sophisticated projects in the world. I am impressed with the talent we have in our India Development Centre and the quality of software being developed." - Bill Gates, Chairman and Chief Software Architect, Microsoft Corp.

I think this will give you clarification with respect to all of your queries regarding why to do Offshore Outsourcing.
Mark Spenser Send private email
Wednesday, August 26, 2009
 
 
Mark, I may be cynical, but what did you expect him to say? "I know we've just moved several thousand jobs out of the U.S., and the quality hasn't been as good, but you have to understand: We really saved a *lot* of money."
Drew Kime Send private email
Wednesday, August 26, 2009
 
 
Some managers are obsessed with processes: do you arive on time, do you wear a tie, did you go to university, did you staple the correct cover sheet onto the TPS report.

Other managers are obsessed with results: are you smart and do you get things done.

Process-oriented managers believe that every thing is controllable via the correct process, and to them, all developers and engineers are interchangable. These are the managers that are in favor of offshoring because they're convinced that with equal processes, the only variable then becomes one of cost.

While process control is extremely valuable in manufacturing, when it comes to development (which is more of an art or craft), then process-oriented management becomes the problem, not the solution. However, as it took decades for process to become entrenched in b-school, it will most likely take decades for results-oriented managment to become dominent. But I'd rather have things balanced out. During WW2, we in the US pretty much made statistical quality control (the epitome of process orientation), and then blew it off after the war, while the Japanese took statistical quality control to heart (thanks to Gen McArthur ordering Deming to Japan after WW2). It took time, but within a few decades, Japanese manufacturing went from being really bad to being the best in the world, while US manufacturers became obsessed with financials and US quality went from number 1 to really sucky. The reputation that the Japanese car manufacturers earned by the 1980s is still a deciding factor for many car buyers today, and it would take a lot of effort for them to squander the lead in quality they've got now.
Peter Send private email
Thursday, August 27, 2009
 
 
Ive managed Offshore labor as Team and Project lead, and it is great for ramping up body counts and getting a lot of work done cheaply and quickly. Its is HORRIBLE for quality and innovation. O worked with an engineer who was fully certified with 5 certs from Microsoft, yet the guy was not able to problem solve new ideas and problems. That has since convinced me that Americans excel at risk taking and are constantly pressured to go against the grains and compete and thrive on creativity. We dont like following the sheep....which has been the rest of the work. That directly applies to IT and why we gave birth to the Internet, Intel, Microsoft, Google, Apple, Twitter, etc etc. Ask yourself why all that was built and innovated here, and not India, Costa Rica or China???

Certainly the US doesnt have an edge in terms of smarts, brains, skills, or talent. Whats the difference....answer that and you answer the question posed above. 

We will continue down the cost cutting offshore track but there will always be huge demand for IT engineers and professionals here simply because the majority of new businesses, IT innovation and quality software comes from the West. As long as thats the case a large volume of IT innovation will continue to come from here.
Stormy Send private email
Friday, August 28, 2009
 
 
My company does business with a development group in India.  They do a very good job; after a couple of years of working with us they're starting to understand some of the gnarly details of our application space. It would take new hires that long anywhere.  Mr. Spolsky maintains that the job of a software company is to convert capital into excellent software with good efficiency.  Our partners in Bhubaneswar are a big part of doing that.

It's easier to understand this issue when we get past the ideologies of capitalism and nationalism.  The bottom line is this:  in the USA the smartest and most ambitious college kids have been going to Wall Street and law school for a generation now.  In India they've been going to engineering schools with rigorous standards.

Which bunch do you want for co-workers?
Oliver Jones Send private email
Saturday, August 29, 2009
 
 
Our parent corporation has dictated that when staff quit, they cannot be replaced by anything other than Indian L1Bs. The nationality *is* specified in the edict from above. They want to go public in the next 2 years, so they are squeezing every bit of profit out to appease potential investors.

http://en.wikipedia.org/wiki/L-1_visa
Peter Send private email
Wednesday, September 02, 2009
 
 

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
 
Powered by FogBugz