[EMAIL PROTECTED] (Michael Tulloch) writes: > > On Wed, 19 Jun 2002, John Keiser wrote: > > > Let's solve the problem we have now, and solve the imaginary ones later. > > Docs and code are extremely different than code in that *no one is > > doing them*. Thus the situation requires *less* process, a smaller > > That's not the point. The point is that they are both deliverables and > thus are naturally handled in similar ways. > > > review structure. There are other qualitative differences too, such as > > the fact that many more people read and understand documentation than > > code (and thus you have a natural reviewer base who will catch errors > > and tell you about them). > > That's not a sufficient reason to ditch planning. By the point the docs > have reached the natural user base, they are deployed. Typos, > incomprehensible sections, grammar problems all lead to major egg upon our > faces. People are going to be harsh upon help anyways. Why give them > further ammo to shoot us?
Crawl, walk, run. I think you're missing John's key point, which I strongly agree with: anything that makes the documentation publishing process more difficult is almost certain to guarantee that less documentation is going to be written. Given that the current documentation output is so low, this is something that we just can't afford right now. Dan
