Ugh... Outsourcing is only a good alternative when you have a reputable supplier of resources overseas AND you have enough staff to deal with the overhead of:
- HEAVY hand-holding type Project Management
- Code check-in
- Code rewrites
- QA of outsourced code
- Gaps of internal knowledge of code
- Skype phone calls for meetings
- Time differences
- Spotty internet coverage in many of these countries (google recent articles about undersea cables and internet outages)
.... just to name a few issues.
If you had one technical and QA resource from the outsources team housed (physically) at your shop who could then interface with the remainder of their team, you might be able to mitigate these issues.
Most who look to outsourcing as an answer do not calculate the true cost to the business. Also, it is EXTREMELY hard to find a good, solid, competent team overseas that is also reliable.
The good resources -- just like in our country -- job hop as soon as they get a better offer.
Submitted by softwarejanitor on Mon, 02/18/2008 - 11:59am.
I think you are right on Audrey... one problem is that if an offshore team is really good they will either get oversold or they will get broken up as key team members will get hired away. So even if an offshore team has been reliable in the past there is no guarantee they will stay that way. When I was dealing with offshore developers the turnover problem was major. India in particular is going through a job market like the US during the .com boom where a guy can go across the street and get a big raise.
Few organizations I've ever been a part of could ever specify what they wanted well enough to make offshore development work. You are more or less forced into mini-waterfall type project management when you have people 1/2 way around the world and that has big problems.
Absolutely agree here .. I can't believe I forgot to mention that in my first list. Documentation at this level becomes so absolutely ridiculous -- even for the simplest of projects. Huge and burdensome waste of time.
...Few organizations I've ever been a part of could ever specify what they wanted well enough to make offshore development work...
You say it is extremely hard to find a team and it sounds like you have some experience with this, Audrey? How do you find a team? And how do you know if they're any good?
Ha! Well trial and error really. We had a project that was bid out by one firm -- who failed to do just about anything they promised. We moved the 'code' they had done so far over to another outsourcing company, who promised to take it over and finish the project by the originally projected time-frame.
They 'finished' by the time promised, but with code that our in-house team says is unmaintainable and needs to be rewritten. It was also missing several pieces of core functionality that was in the design doc.
Oh another HUGE piece of overhead I forgot - documentation - detailed and explained in the most basic terms possible to reduce opportunities for misunderstandings across languages. Spreadsheets to manage any issues or changes to documentation, along with fully documented decision history, is also critical to project success.
Finally (and thankfully) the realization that the cost 'savings' by outsourcing was more than wiped out by our internal cost to manage those resources helped squash any idea that we should continue to invest in this direction.
If your company is setup today so that any 'code monkey' could write code from your specs -- and you have the internal code architecture in place that requires no critical thinking -- then it might work.
Personally, I'd trade one 1/2 decent programmer in my office with whom I can interact in person and has the ability to solve business problems with creative technical solutions - for 10 overseas people in a heartbeat.
I've worked with overseas teams and made it work to some level of success... but I would never, ever choose that over in an in house team if I had the option.
Submitted by softwarejanitor on Thu, 02/21/2008 - 12:02pm.
I really wish that more people had your opinion about using offshore development teams. I'd be a lot more optimistic about about the future of IT employment in the US. Unfortunately too many PHB types only seem to see the lure of low hourly rates and are willing to endure the pain to get them especially if they can make others deal with it.
Submitted by softwarejanitor on Thu, 02/21/2008 - 1:16pm.
O.K., then Subversion is the place to start. CVS is also still in common use although many CVS users are switching and most people would go with Subversion if they are picking something new.
Ugh... Outsourcing is only a good alternative when you have a reputable supplier of resources overseas AND you have enough staff to deal with the overhead of:
- HEAVY hand-holding type Project Management
- Code check-in
- Code rewrites
- QA of outsourced code
- Gaps of internal knowledge of code
- Skype phone calls for meetings
- Time differences
- Spotty internet coverage in many of these countries (google recent articles about undersea cables and internet outages)
.... just to name a few issues.
If you had one technical and QA resource from the outsources team housed (physically) at your shop who could then interface with the remainder of their team, you might be able to mitigate these issues.
Most who look to outsourcing as an answer do not calculate the true cost to the business. Also, it is EXTREMELY hard to find a good, solid, competent team overseas that is also reliable.
The good resources -- just like in our country -- job hop as soon as they get a better offer.
I think you are right on Audrey... one problem is that if an offshore team is really good they will either get oversold or they will get broken up as key team members will get hired away. So even if an offshore team has been reliable in the past there is no guarantee they will stay that way. When I was dealing with offshore developers the turnover problem was major. India in particular is going through a job market like the US during the .com boom where a guy can go across the street and get a big raise.
Few organizations I've ever been a part of could ever specify what they wanted well enough to make offshore development work. You are more or less forced into mini-waterfall type project management when you have people 1/2 way around the world and that has big problems.
Absolutely agree here .. I can't believe I forgot to mention that in my first list. Documentation at this level becomes so absolutely ridiculous -- even for the simplest of projects. Huge and burdensome waste of time.
...Few organizations I've ever been a part of could ever specify what they wanted well enough to make offshore development work...
You say it is extremely hard to find a team and it sounds like you have some experience with this, Audrey? How do you find a team? And how do you know if they're any good?
Ha! Well trial and error really. We had a project that was bid out by one firm -- who failed to do just about anything they promised. We moved the 'code' they had done so far over to another outsourcing company, who promised to take it over and finish the project by the originally projected time-frame.
They 'finished' by the time promised, but with code that our in-house team says is unmaintainable and needs to be rewritten. It was also missing several pieces of core functionality that was in the design doc.
Oh another HUGE piece of overhead I forgot - documentation - detailed and explained in the most basic terms possible to reduce opportunities for misunderstandings across languages. Spreadsheets to manage any issues or changes to documentation, along with fully documented decision history, is also critical to project success.
Finally (and thankfully) the realization that the cost 'savings' by outsourcing was more than wiped out by our internal cost to manage those resources helped squash any idea that we should continue to invest in this direction.
If your company is setup today so that any 'code monkey' could write code from your specs -- and you have the internal code architecture in place that requires no critical thinking -- then it might work.
Personally, I'd trade one 1/2 decent programmer in my office with whom I can interact in person and has the ability to solve business problems with creative technical solutions - for 10 overseas people in a heartbeat.
I've worked with overseas teams and made it work to some level of success... but I would never, ever choose that over in an in house team if I had the option.
I really wish that more people had your opinion about using offshore development teams. I'd be a lot more optimistic about about the future of IT employment in the US. Unfortunately too many PHB types only seem to see the lure of low hourly rates and are willing to endure the pain to get them especially if they can make others deal with it.
Great advice! I was on the fence on this issue but you've confirmed what I suspected that outsourcing isn't a bed of roses.
depends on your goal. i know of a great group if you're looking for people.
also, try using http://smartbear.com (austin co) for code review -
How is smartbear set up for code review? Do they use subversion? How does their code review process work?
If you look at their web site they have proprietary tools and a methodology for doing code reviews which they sell.
I'm trying to figure out subversion.
What are you familiar with? If you know CVS it shouldn't be too hard to make the switch.
Our team is switching from CVS to Subversion right now...
Are you doing any hands on with this, audrey?
No -- the development team is..
I don't... starting from scratch.
O.K., then Subversion is the place to start. CVS is also still in common use although many CVS users are switching and most people would go with Subversion if they are picking something new.
I need on of those "perfect subversion install guides" like they have for ubuntu. Those guides are awesome.
You might try here:
http://svnbook.red-bean.com/en/1.4/svn-book.html
OK I'll check it out, thanks.