Date: Thu, 3 Aug 2006 14:30:19 -0700 (PDT) From: <backyard1454-bsd@yahoo.com> To: Colin Percival <cperciva@freebsd.org>, John Rogers <ruralriver@gmail.com> Cc: freebsd-questions@freebsd.org Subject: Re: **SPAM** binary upgrade issues Message-ID: <20060803213019.90803.qmail@web81615.mail.mud.yahoo.com> In-Reply-To: <44D26614.7010704@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> > I also wonder why these binary update and upgrade > are not legitimized > > in the freebsd core distribution. An important > reason why linux is > > used by more is its easy update solution similar > to Microsoft's > > Windows Update. Sure "make world" is fun > especially to developers. > > But providing easy update and upgrade tools in > addition will attract a > > large user base who just need a stable and easy to > use operation > > system - and many of them can be companies who can > be potential donors > > to the freebsd project. So the effort to this > path will be well > > rewarded. > > We're moving in that direction. Everything starts > out by being experimental > before becoming officially supported and endorsed. > > Colin Percival I acutally find it better to do the "make world" then to deal with binary updates because if it builds on your system it will typically run on your system, as well as there not being silly little incompatabilities with the system libraries and binaries. I find updating linux to be the most god awful prospect on earth which is why I switched to FreeBSD for the most part. It's probably gotten a lot better since Redhat 7.x which is what I was using. Gentoo is a lot better but I haven't had a working system since they updated the kernel to xx.xx.15 and put gcc 4.x into the base system... but to each their own, I know a binary update would be nice when I start deploying things like desktopbsd on my friends PCs who don't get formatting a harddrive let alone building software. However this would mean the builds would have to be generic i586, i686 and on an old p3 500mhz machine building for a specific processor with specific optimizations can make a huge difference in performance. Even more so on p2 166 machines. Again to each their own. But I wouldn't tout Microsoft update as a good thing becuase there are known bugs where updates can erase previously updated code with old buggy code... sorry been a long day had to end it with blasting microsloth... -brian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060803213019.90803.qmail>