Re: [Groff] Status of the portability work, and plans for the future

[Eric S. Raymond]

Gunnar Ritter <address@hidden>:
> > > Maybe. But still the tools should not complain if a troff
> > > expert has decided that something is safe.
> >
> > And how are they going to do that?  Mental telepathy? :-)
> 
> As I wrote in the text following, by separating a "lint"
> (or "-Wall") mode and silent regular use.

That's not a bad idea, but it seems to me to solve a slightly different 
problem than the one I thought we were discussing.  Anyway, the
intention would be better expressed by having an "expert" flag that
switches *off* lint-like warnings -- that way, authors (knowing that
users would not in general use that flag) wouldn't be tempted
to simply skip the validation step entirely.
 
> > And that is what, <1% of our potential userbase?
> 
> >50% of those users that actually develop software, write
> useful bug reports, maintain the documentation, . . . ?

And you think these wise and savvy hackers are going to be terminally
offended by an often-unnoticeable drop in the quality of nroff output to
a terminal emulator, especially when they will know that in return
they can go to a browser and get (a) prettier fonts with proportional
spacing, (b) images, (c) color, and (d) *hyperlinks*?

Sorry, but that's just *dumb*.  It's as if you were balking at a
proposal to optimize a writing system for paper because you think some
of the (other) scribes might be stuck on cuneiform tablets.  You are
defending reflexive conservatism here, not thinking.  What's worse is
that you're not even being honest enough to defend your own
conservatism (if that's what the problem is). Instead, you're
attributing an absurd degree of touchiness and outright stupidity to a
class of people you implicitly claim to respect.

One of those people is me.  Because I am an ancient crusty Unix
grognard who *should* by all rights be as retrograde about how I read
my documentation as you are describing (if anyone is going to be), I
get to call bullshit on this kind of error.  1976 was thirty years
ago.  man(1) is not sacred, and I actually find attitudes like yours
kind of insulting, as though you think long-time man users like me are
so utterly lacking in mental flexibility as to be unable to cope with
even a tiny speedbump on the road to better things.

> I would rather risk to lose 100 non-contributors than one
> contributor. But since we are just developing an approach
> that satisfies both camps, I do not really understand what
> point you are trying to make here?

It's a point for the future, really, and goes back to the
philosophical question I opened up at the beginning of this
discussion: is the groff community ready to accept that the future of
on-line documentation belongs to hypertext and that man is a legacy
format that must adapt itself to the new reality, rather than holding
it back?

(Again, this is not a brief for the obsolescence of *roff; typesetting
is still typesetting and involves a rather different kind of
constraints than formatting for on-line viewing.)

This is what the discussion of viewer portability comes down to,
because the new viewers all render to HTML amnd it is dead obvious
that the trend in documentation browsers like KDE's is towards
integration with the Web, not away from it.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>