Re: XP and Pair Programming
- From: "Phlip" <phlipcpp@xxxxxxxxx>
- Date: Mon, 22 Aug 2005 17:22:11 GMT
Christopher Barber wrote:
> I have no doubt that you can produce better software w/ two heads instead
> of one. However, it is not at all clear that two people working a pair
> are twice as productive as one person, which is what it would take to
> break even. I doubt it. I think you are better off doing it only rarely
> and relying on reviews and frequent programmer interaction the rest of the
> time.
I use a recurring anecdote when someone asks that; I have probably posted it
here more than once. Here goes...
A former colleague of mine announced he intended to fly with our sprint's
profits to Las Vegas and play poker with it. I scoffed "Vegas??", and he
said proudly, "Don't knock it - I'm ahead on my lifetime total."
I asked, "Did you factor in the cost of airplanes and hotels?"
"No!"
Pairing may or may not make implementing one feature in one module faster.
The overall cumulative effects on an entire project are profound. Regardless
of me pushing TDD to lower bug rates, teams can easily test-last, pair
program, and achieve similar benefits.
--
Phlip
http://www.greencheese.org/ZeekLand <-- NOT a blog!!!
.
- Follow-Ups:
- Re: XP and Pair Programming
- From: Christopher Barber
- Re: XP and Pair Programming
- References:
- XP and Pair Programming
- From: Ashima
- Re: XP and Pair Programming
- From: christian9997
- Re: XP and Pair Programming
- From: Nick Malik [Microsoft]
- Re: XP and Pair Programming
- From: Christopher Barber
- XP and Pair Programming
- Prev by Date: precondition and postcondition
- Next by Date: Re: chooses not to generate code at all
- Previous by thread: Re: XP and Pair Programming
- Next by thread: Re: XP and Pair Programming
- Index(es):