Date: Fri, 02 Sep 2011 03:07:40 +0200 From: "Julian H. Stacey" <jhs@berklix.com> To: Michal Varga <varga.michal@gmail.com> Cc: freebsd-ports@freebsd.org Subject: Re: Ports system quality Message-ID: <201109020107.p8217eFJ089679@fire.js.berklix.net> In-Reply-To: Your message "Thu, 01 Sep 2011 23:47:40 %2B0200." <1314913660.1744.136.camel@xenon>
next in thread | previous in thread | raw e-mail | index | archive | help
Michal, Nice analogy ! > Nobody is really steering this ship anymore and it just happily > rams icebergs along the way, with volunteers occasionally throwing > buckets of water (and sometimes pieces of furniture) overboard to > somehow keep it afloat for a while longer. Furniture like sysutils/diskcheckd, mail/procmail etc was recently under threat of discard, from misguided preference to toss code alleged problematic, rather than investigate send-prs others said were problematic. Tossing to CVS Attic is no excuse, it's not something many release users have to hand, nor will they read ports@ etc between releases. Release users are what a company would call valuable customers. For FreeBSD they should be valuable too, & include firms that use releases to create jobs, sponsor developers, & promote BSD. But FreeBSD ports@ degrades releases with: Give no warning, dont mark deprecated on one release then wait till next release before removal, just toss out ports between releases with no warning. No principle of least suprise. Unprofessional. I've not known a clean build of ports/ I use in 10 years ( http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/ports/jhs/Makefile.local ) But as I dont build in chroots, probably I should be grateful the can of worms is not bigger, with `/bin/ls -1 /var/db/pkg | wc -l` > 1000 The mass waste of time with all major Unix projects maintaining their own ports/ packages shims/ skeletons to build generic software is scarey. Microsoft must grin at all us BSD, Linux, maybe Solaris & presumably even now http://www.minix3.org free source enthusiasts reinventing similar old ports shims for same old 3rd party wheels:. A ports manager made a good point: there's a big investment of people skills in doing things the FreeBSD way. But that could also translate as: Our Titanic is too heavy to steer. FreeBSD ports infrastructure/ future history could be any of these: - Struggle on as is. - Accept we're in trouble, (as some NetBSD too accept), co-operate, & get someone [ SoC students ? ] to design *BSD *Linux, a new cross/ OS shim set. (too scarey letting students do that > pay some one else. ). - Get FreeBSD & NetBSD foundations & maybe wider to sponsor a new build structure with a limited number of demo ports. - Wait & see if a firm does a Cygnus on us. ie what Cygnus did for FSF UX tools on MS, remove some pain & charge some money. - Wait & see if someone ports their Linux NetBSD or whatever ports shims to also work on FreeBSD. (just as Linux tools on a BSD kernel came as a suprise to us). - See if someone who produces RFCs takes a crack at defining standards for ports/ shims BTW I call 'em shims, cos the word ports causes problems in the NetBSD hardware ports camp, Cheers, Julian -- Julian Stacey, BSD Unix Linux C Sys Eng Consultants Munich http://berklix.com Reply below, not above; Indent with "> "; Cumulative like a play script. Format: Plain text. Not HTML, multipart/alternative, base64, quoted-printable. http://www.softwarefreedomday.org 17th Sept, http://berklix.org/sfd/ Oct.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201109020107.p8217eFJ089679>