Date: Tue, 10 Jun 2008 15:09:53 -0400 From: Miles Nordin <carton@Ivy.NET> To: freebsd-sparc64@freebsd.org Subject: Re: status of freebsd on ultrasparc? Message-ID: <oqy75dxg1q.fsf@castrovalva.Ivy.NET> In-Reply-To: <20080609234022.GA18959@soaustin.net> (Mark Linimon's message of "Mon, 9 Jun 2008 18:40:22 -0500") References: <6e5cf6a70806051341x95eb814j6222c04dbfd7fb2d@mail.gmail.com> <20080609211208.GA66541@alchemy.franken.de> <20080609231419.F13655@ury.york.ac.uk> <484DBB64.5050607@FreeBSD.org> <20080609234022.GA18959@soaustin.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--pgp-sign-Multipart_Tue_Jun_10_15:09:41_2008-1
Content-Type: text/plain; charset=US-ASCII
>>>>> "ml" == Mark Linimon <linimon@lonesome.com> writes:
ml> We need more people using and testing on sparc64 and
ml> contributing fixes back. It's as simple as that.
IMHO if we deliver a platform with a working web browser for two minor
stable releases in a row, we're likely to have more testers.
but the key phrase is ``and contributing fixes back.'' More testing
by users who don't have the time, interest, or ability to create their
own fixes doesn't seem to me useful to sparc64 at this stage.
Another thing that might help is to pick some favoured pieces of
hardware and clearly specify what works and what doesn't. like ``the
IDE controller works but can't burn CD's'' or ``the network card works
but doesn't autodetect duplex reliably.'' The current plan with most
less-equal BSD architectures seems to be: advertise as ``working''
whatever you'd like people to test, and wait for them to report
otherwise. When they do, ignore the report and ask if it's still
broken in the latest version which of course they haven't tested yet.
``buy a 200lb 1000watt machine on eBay, and test it for us and let us
know what you find,'' just doesn't work for me at all. I think only
someone who wants to be a FreeBSD/sparc64 developer would (or should)
do that. I guess it would be nice to run on more SPARC hardware but
I'm not sure how sustainable it is to get and keep FreeBSD running on
hardware that developers don't posess themselves, or can't afford to
leave plugged in.
Ultimately it looks more and more to me like, whether the software is
free or not, achieving high enough quality to get work done requires a
paid support contract where attention gets focused onto the bugs
reported by the people who are paying. With Solaris or Linux you can
sometimes get high quality without paying by looking over the
shoulders of the people with support contracts and trying to get as
close as convenient to what they're doing so that the two of you will
run into the same bugs.
I don't know how to pull this trick with FreeBSD. Are there specific
bits of sparc64 hardware FreeBSD uses for build labs, that non-fixing
users can buy the exact same thing and benefit from fixes developers
do to keep the build lab running? The problem is, even if you do
this, you gain the ability to: build FreeBSD. not to run a web
browser, because all the developers just use some proprietary browser
under Mac OS X that supports FLASH. so it's a useless strategy.
--pgp-sign-Multipart_Tue_Jun_10_15:09:41_2008-1
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (NetBSD)
iQCVAwUASE7RgYnCBbTaW/4dAQKzAAQAoYc7/1dBpvMBfdCrFoWXmLXS4GUcTeTB
fMxWHoyn+VOuMMKpFxa9zgnbtpFZPBRDRiafHJRQSCcZuiyAZ9bwQGETP2JLcFTZ
ruQaBKy6M3HSVAEpxfJIIJZEKjfF0+wu6R84ED2EzCLfBcOvcWtlvS+GHrUgpqXZ
0FYbJK8n8ok=
=lzF6
-----END PGP SIGNATURE-----
--pgp-sign-Multipart_Tue_Jun_10_15:09:41_2008-1--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?oqy75dxg1q.fsf>