[CMake] Documentation strategy

[Brandon Van Every]

On 6/20/07, Robert J. Hansen <rjh at sixdemonbag.org> wrote:
> Brandon Van Every wrote:
> > Counting on people to buy books to do evaluations is bad strategy.
>
> I'm not talking strategy.  I'm simply saying that your broad claim that

I am; note the subject line.  Strategy is about broad patterns of behavior.

> > will, when prodded by peers.  If you're the only guy with the CMake
> > book, and you're waiting for it, and it's about your schedule and your
> > ways of masking the shipping delay, and you being an "assigned" person
> > to deal with it in the 1st place, there are lotsa extra barriers.
>
> I can't parse this sentence.

It's long and run-on but properly constructed.  Is English not your
1st language?  Note the implicit "then," i.e. "[then] there are lotsa
extra barriers."

> >>> The projects that see CMake as a slam dunk, are the ones that did an
> >>> Autoconf build for the Unix stuff, and also had to maintain some
> >>> horrible hand rolled Visual Studio build, typically with .BAT files.
> >>
> >> At the time KDE converted to CMake it was a UNIX-only project, and they
> >> considered CMake to be a slam dunk.
> >
> > No, they were planning to port huge chunks of libraries to Windows.
>
> Yes, exactly.  Please note that "planning to port huge chunks of
> libraries to Windows" means they were not yet ported to Windows.  It was
> a UNIX-only project.

No, when you are definitely planning to port to Windows, and picking
tools on that basis, you are not doing a Unix-only project anymore.
You are at the planning stage of a cross-platform project.

>  They were not maintaining "horrible hand rolled
> Visual Studio builds".

Yes, they were smart enough to avoid that.  CMake is still a slam dunk
for people who weren't smart enough to avoid that.

> Again, I think you are arguing too broadly.

I'm not sure what point you're really trying to make here, other than
you don't like it when people make "sweeping generalizations."  I'll
try rephrasing the spirit of the argument:

- huge numbers of people don't and won't buy books to do evaluations
- there's a lot of competition out there for "modern" build systems
- without passable free docs, CMake will not become a de facto standard.
- intelligent early adopters don't create de facto standards.  The masses do.
- CMake does not have a good method for getting the community to write the docs.
- if CMake doesn't choose to solve this problem, other build tools will.


Cheers,
Brandon Van Every