M. Warner Losh wrote: > This thread should end here. Enough people want to support floppies > that they aren't going to go away before 6.0 at the earliest. It > doesn't matter what arguments are brough up for the removal, it simply > won't happen anytime soon. This is bikeshed. Concentrate these > energies on testing and fixing problems. > Okay, floppy support must stay, and by the sound of it we will have to support them for a long time to come. I myself have typically taken advantage of the floppies for installing FreeBSD. That is not the end of our problem, however. I am not new to FreeBSD at all. I've tested the hell out of 5.0-CURRENT, and at times have even helped fix problems. I have done this by supplying bits of information to FreeBSD commiters so that they can use their vast knowlege and understanding of FreeBSD to progress it to a state of stability and reliability. Since we need to keep the floppies, why don't we create a number of custom kern.flp images? This is a current issue of vital importance and must not be ingnored any longer. Why, a simple rackmount system compatible kernel.gz could easily be less than 1MB in size. Why insist kern.flp to be all things to all people? That is what seems to be the root of the problem here. Please forgive me if there are further binding issues holding the floppies in the position that they're currently in, but I don't see such a big problem in having some custom kernels defined for creating a set of kern.flps that people can choose from in order to match their set of requirements. FreeBSD is already quite complex and to add a few more options in this area is not too much to ask of the installer's brain capacity. Pete...Received on Sun May 04 2003 - 04:58:31 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:06 UTC