Re: Custom undocumented(7)s are just as bad.
On Sat, Jan 29, 2000 at 09:33:54AM -0500, Michael Stone wrote:
> On Sat, Jan 29, 2000 at 03:23:27PM +1100, Brian May wrote:
> > >>>>> "Manoj" == Manoj Srivastava <srivasta@debian.org> writes:
> >
> > Manoj> a) If the maintainer does not know what every binary in the common
> > Manoj> PATH is doing, he should give the package to someone who is
> > Manoj> willing to spend the time to learn the package.
> > Manoj> b) If the binary is an internal binary, it should not be in PATH,
> > Manoj> but in /usr/lib/<pkg>
> >
> > I disagree. Take Heimdal for instance. It is a very complicated
> > package (at least by my standards), has numerous libraries and
> > clients. Most of the binaries are obvious, eg telnet, ftp, and even
> > come with man pages. However, some don't, eg, des, verify_krb5_conf,
> > string2key.
>
> I disagree with your disagreement. If you don't know what the binaries
> do, how is the user supposed to know? It's your job to do the research
> so every user doesn't have to do it on his own. I agre with manoj that
> maintainers should not be packaging things they don't understand--that
> just has all sorts of bad implications.
Manoj, Michael, I can see where you are coming from, and it makes a good
deal of sense -- but part of me thinks it is folly to require *that* level
of understanding before packaging a program, and expect it after being
packaged.
Some programs are just complex -- not many people can claim to know what
every file in the linux kernel can do. Not many people can claim to know
sendmail/qmail/exim/smail/postfix inside and out. X? Ouch. (If you, the
gentle reader, do know s/q/e/m/p or X inside and out, perhaps then a nice
feeling of smugness is called for. :)
While I agree that maintainers should know more about their packages than
the average end user, to expect them to know it as well as the upstream
authors just doesn't seem realistic. In the end, the maintainers of packages
are end users, people that used a package often enough that they desire to
maintain it, for the good of the project. Maintainers aren't employed by
anyone to know something inside and out, so they must spend their time
earning a living. I think this must be understood.
Perhaps what we should be after, is having maintainers that know their
packages well enough to track down user's questions in a reasonable amount
of time. Isn't that what we are really after? Who cares if the maintainer
knew the answer before or after the question was asked, as long as the
answer comes back in a reasonable amount of time?
Our large community model just doesn't allow for maintainers to spend all
day working on their programs -- to expect otherwise would alienate many.
ObDisclaimer: Of course, all this is IMHO, IANADD(Y), YMMV.
:)
--
Seth Arnold | http://www.willamette.edu/~sarnold/
Hate spam? See http://maps.vix.com/rbl/ for help
Hi! I'm a .signature virus! Copy me into
your ~/.signature to help me spread!
Reply to: