Newsgroups: news.software.b Path: utzoo!henry From: henry@utzoo.uucp (Henry Spencer) Subject: Re: Where does cnews live? Message-ID: <1989Aug9.164003.20669@utzoo.uucp> Organization: U of Toronto Zoology References: <28272@teknowledge-vaxc.ARPA> <1989Aug7.195904.13429@utzoo.uucp> <1989Aug8.170802.20975@vicom.com> Date: Wed, 9 Aug 89 16:40:03 GMT In article <1989Aug8.170802.20975@vicom.com> lmb@vicom.COM (Larry Blair) writes: >Which brings us back to the issue of patch numbering. If the patches for >C News were numbered like every other piece of net.software, we would >know that there were only 3 and we would know if we had missed any. Oh, nonsense. If they were numbered, you'd be asking what the latest number was. This way, if the patch is dated yesterday, you have pretty considerable confidence that you are current. And to find out whether you've missed any, you simply look at the list of prerequisite patches that is in *every* patch. (I agree this would get unwieldy if there were 57 of them, but there won't be.) >I still don't understand what the problem with the conventional numbering >is. It is not to late to change. Is this just a case of NIH? No, it's a case of NLH (Not Liked Here), plus an interest in finding out whether an alternative would work. I must say, if we'd anticipated that there would be this much mindless whining about it, we might have thought twice about the experiment... I would welcome *carefully thought out* comments on it, but please stop nattering about how it's not what you're used to and therefore it must be bad. I thought about this scheme at some length, and so far *nobody* has come up with an argument that I didn't think of in advance. I still feel that (a) the problems are nowhere near as serious as claimed, and (b) it works about as well when you think about it. Patch numbers make it easier to list all the patches that preceded a given one; that is their sole advantage. Patch dates tell you how current the latest patch is; that is their major advantage. In either case, the missing information can be had by referring to the text of the patch itself, if it's built properly (and ours are). We may change if it becomes clear that the date-based scheme is causing more problems than it solves. So far I see little evidence for this. Convincing us is going to require good arguments, not endless repetition of poor ones. -- 1961-1969: 8 years of Apollo. | Henry Spencer at U of Toronto Zoology 1969-1989: 20 years of nothing.| uunet!attcan!utzoo!henry henry@zoo.toronto.edu