Michael J. Lynch reportedly babbled: > Ok so several of the replies in this thread mention how hard it is for > *NIX to > become mainstream because of problems with packages. Furthermore, most > of these comments reference how when one gets a package for M$ or OSX it > just installs and works. > > Well...that's very easy for them to do, they only have to worry about > one processor > type and one machine architecture. Linux, in particular, has to be able > to deal > with just about every processor and machine architecture from embedded > systems > up to S/390 mainframes. This makes the task of QA testing packages for > linux > a formidable task. > > We here are the maintainers of LiS (Linux Streams) and everytime we make a > change we have to test it on at least 3 versions of at least 4 > distributions of Linux > for x86 architectures. While it is used on S/390's, Sun SPARC stations, > and other > machines running Linux, it would be completely impractical for us to > test the > package on any and all of the processors and/or architectures available. We > would NEVER be able to release any software. > > The developers of Myth have the same problem only worse. It would be > impractical and too time consuming for them to test the package on all > of the > different systems people are using. I'm also sure they don't have the > finacial resources > to go out a buy a pile of different machines with different video cards, > drivers, > etc. on which to test the package. > > -- > Michael J. Lynch
Hi, I fully agree with this.. It is a horrendous problem. I worked for Award Software for many years doing BIOS code and support, you don't want to know what that world is like... I later worked for BMG Software (formerly Boole & Babbage) and a major Internet backbone provider so I fully understand the testing and nightmares of doing so.. This IS why a formal release proceedure is so important. But with a formal procedure you can manage and mitigate these issues. Just doing a wildcard release because "it works on my machine and so what if it does not work on yours" just does not encourage others to help you out.. I am not new to any of the issues with *NIX in general and support VMS, MVS, *NIX, MS, and Novel systems, I have been hacking for 30+yrs. And yes I am new to THIS list.. Yes I may have missed something in the docs and some note about the audio sync problems, yes I am up at 7:30 in the morning and go to bed at 3:00 in the morning 7 days a week and sometimes I miss the fine points because there either is too/little much caffine and always too little sleep... I was not complaining about my issues with my install.. I will review what I have done and find the current problems, the first of which was doing both FC3 and Myth installs almost concurrently without figuring out the new issues with FC3 first.. My goal with my Myth install was to not break my transcode install and all the other software I have but hey, what the hell.. it gives me something to do.. right? :-) What I do believe in though is that to get the average guy who is interested in any package (whatever it is) to use it, it should be made available for the mainstream systems - BSD, OSX, and the most dominant Linux distro's (DEB/FC/MANDRK) - in a format that is "relatively" easy to install and manage (DEB's/RPM's). At MBG we provide system monitoring and management tools for any platform you can find and a few that most have certainly never heard of.. The support and packaging/testing issues are hugh.. I don't expect Issac or anyone else here to undertake that level at all.. this is after all a hobby for most of us. BUT a stable release process like most of the other Linux software packages would go really far for MythTV and a few pre-packaged bundles would go even further. I still think MythTV is awesome. Again.. JMHO -R -- ------------------------------------------------------------------------ Rich Hall [EMAIL PROTECTED] http://www.netlynx.us/rich/ ham radio: kf6arx ------------------------------------------------------------------------ No trees were destroyed in the sending of this message; however, a significant number of electrons were terribly inconvenienced. ------------------------------------------------------------------------ And remember - if it ain't broke, hit it again. _______________________________________________ mythtv-users mailing list mythtv-users@mythtv.org http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
