My recent reply about the problem of Linux being so heterogenous set me thinking. There's not much point having standards, if they move as fast as Linux does! So, these standards would have to have quite solid roadmaps setting out the intended developments for the future, so that vendors could reliably target software. This in turn would have a strong impact on the development of Linux itself, as well as many of the supporting apps. If the GUI standard (hot topic) says that version 3.0 (target release date q3-1998) will use GTK not Motif, that will have a big impact on what RedHat labs does. And if the Linux Network Standard decides that AppleTalk should only be used as a kernel module (we assume for this example that there is a good reason for doing this...), that'll mean the Kernel guys (yes, _you_ Linus <g>) had better comply for kernel 2.2.x, ensuring 'make config' default correctly. Sooo... Such standards will work only if they come from the top, or from a very well respected body. As I've said, if these standards are to be useful to vendors and Linux distributions, they will, I think put something of a damper on the fits and starts developments we have seen so far. No more adding things cause they are cool - better check the standards first, and the standards people are cautious and slow moving. The benefit? A stronger, simpler, Linux for all of us? -------- Jon Peterson [EMAIL PROTECTED] +44 (0)171 613 5300 Internet Developer When she told me I was average, she was just being mean. -- PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES! http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe" as the Subject.
