Date: Wed, 30 Jun 2010 13:01:41 -0700 From: Tim Kientzle <tim@kientzle.com> To: Erik Cederstrand <erik@cederstrand.dk> Cc: Julian Elischer <julian@elischer.org>, Steve Kargl <sgk@troutmask.apl.washington.edu>, freebsd-arch@freebsd.org Subject: Re: dwb : groff replacement proposal Message-ID: <D1649004-5B2C-42C7-8483-DC878AA69AC0@kientzle.com> In-Reply-To: <877EF62B-D2BE-422F-8545-DBC63E5AA682@cederstrand.dk> References: <4C2AB793.4040601@laposte.net> <20100630034622.GA48794@troutmask.apl.washington.edu> <4C2B11A9.4020206@laposte.net> <4C2B6922.6090300@elischer.org> <20100630170757.GA52509@troutmask.apl.washington.edu> <877EF62B-D2BE-422F-8545-DBC63E5AA682@cederstrand.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Jun 30, 2010, at 11:48 AM, Erik Cederstrand wrote: > Den 30/06/2010 kl. 19.07 skrev Steve Kargl: >>=20 >> The fact remains that there is no available alternative that >> contains the functionality of groff. >=20 > I still can't read from this discussion if FreeBSD base actually needs = all the functionality that groff provides, and if the proposed = alternatives are lacking needed functionality which cannot be worked = around by simple changes to the distributed man-pages like the ones = committed in the last weeks. >=20 > I may be horribly misinformed, but man-page rendering does seem like a = fairly simple task. As many have pointed out, "replacing groff" is certainly not a priority for the project. Our current groff works, works reasonably well, and is likely to meet our needs for at least another decade (unlike C or C++, nroff functionality is not a moving target). But more importantly "correct man-page rendering" is actually pretty hard. The issue is not the manpages in the FreeBSD base---we can and should clean those up and experimentally rendering them with other tools is a good way to verify them. The problem comes with third-party manpages, including those installed by ports. Either the in-base replacement is pretty much bug-for-bug compatible with groff or else everyone will have to install groff anyway, which defeats most of the point of replacing groff in base. That said, if someone has tested an alternative to ensure that it provides the same quality output as groff across a wide swathe of base and ports-installed manpages, and there are other real advantages (license, size, features, complexity), then I think it's = worth considering. Tim
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?D1649004-5B2C-42C7-8483-DC878AA69AC0>