Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!sdd.hp.com!elroy.jpl.nasa.gov!ncar!csn!stan!kucharsk From: kucharsk@solbourne.com (William Kucharski) Newsgroups: news.software.b Subject: Re: CNEWS MUST DIE! Message-ID: <1991May20.215138.12821@solbourne.com> Date: 20 May 91 21:51:38 GMT References: <1991May18.041319.29490@zoo.toronto.edu> Organization: Solbourne Computer, Inc., Longmont, CO Lines: 30 In article mathew@mantis.co.uk (CNEWS MUST DIE!) writes: >henry@zoo.toronto.edu (Henry Spencer) writes: >If the input to relaynews is *really* unparseable (lacking *essential* >information and containing *uncorrectable* errors) then it is reasonable to >drop the input. The sorts of errors C News drops articles for are neither >uncorrectable, nor are they errors which cause the article to lack essential >information. Well, hell then, let's just throw out all protocol specs and allow things which are "close enough." This is a rather rash statement, but when there is a published document stating how things _should_ be, why should that software then have to correct information not in that format to conform to the published one? Yes, it's great for compatibility purposes, but there's a certain point when you have to stop being backwards compatible and take the leap forwards. It kind of strikes me as the whole war that some Mac developers have waged in the past. Yes, they'll do really nifty things on the Mac but they'll do things that _Inside Macintosh_ told them not to do, and when a new System comes out and their app breaks, they're surprised. The RFC's been around for a few years; software should have been changed to follow it by now. As far as mailing the originator in case of an article with a bad header, try this experiment: Send out a sendsys message. Note the number of replies you get. Now multiply this by the number of news articles which may get sent in a given day before you fix your software. Do you have that kind of mail spool space? I know my system doesn't... -- | William Kucharski, Solbourne Computer, Inc. | Opinions expressed above | Internet: kucharsk@Solbourne.COM | are MINE alone, not those | uucp: ...!{boulder,sun,uunet}!stan!kucharsk | of Solbourne... | Snail Mail: 1900 Pike Road, Longmont, CO 80501 | "It's Night 9 With D2 Dave!"