Problems of Symbol Congestion in Computer Languages

[Tim Wintle]

On Fri, 2011-02-18 at 12:27 +0000, Ian wrote:
> 2) Culture. In the West, a designer will decide the architecture of a 
> major system, and it is a basis
> for debate and progress. If he gets it wrong, it is not a personal 
> disgrace or career limiting.  If it is
> nearly right, then that is a major success. In Japan, the architecture 
> has to be a debated and agreed.
> This takes ages, costs lots, and ultimately fails.  The failure is 
> because architecture is always a trade off -
> there is no perfect answer.

I find this really interesting - we spend quite a lot of time studying
the Toyota production system and seeing how we can do programming work
in a similar way, and it's worked fairly well for us (Kanban, Genchi
Genbutsu, eliminating Muda & Mura, etc).

I would have expected Japanese software to have worked quite smoothly,
with continuous improvement taking in everybody's opinions etc -
although I suppose that if production never starts because the
improvements are done to a spec, rather than the product, it would be a
massive hindrance.

Tim Wintle