Many people are much better software designers than they are implementers. Not everyone can design, or to say it differently, more people can implement software better than they can design software, than vice versa. Not every construction worker is an architect.
Most companies have roles for software architects whose primary job is to design software. But if you look at the world of open source, there really isn't any place for that role. Even the most revered of the open source "gurus" write and check in their own code.
In the corporate world, you have to prove yourself before you get to be an architect, but once you've earned that status, your company lets you design software and somehow gets those designs to teams of people to implement them (whatever method your company uses.. maybe the developers work for the architect, maybe not).
Implementation isn't as much fun as design. Developers who aren't getting paid don't want to work on stuff that isn't fun - I can't blame them. But I wonder what would happen if there was a system where software designs were submitted by, well, anyone I guess but preferably experienced designers (how do you filter for that), and then users would vote for which of those designs they'd like to see with their dollars - pledges, backed up by Visa cards or cash in escrow or some such system - and once it was looking like it would be profitable (or at least, not a loss), the actual implementation work is contracted out.
I can think of many flaws with this system - like how do you pick who gets paid to do the implementation - but then again maybe a team could sign up for the implementation, and not receive payment until they deliver. This would be a little encouragement on top of the usual motivation Eric Raymond suggested motivates developers - their ego.
Maybe this method isn't the best one but the goal is to let gifted software designers do what they do best, and let someone else do the implementation. Something to think about.
7:50:40 AM 
|
