Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA Path: utzoo!watmath!clyde!bonnie!akgua!sdcsvax!sdcrdcf!hplabs!hao!seismo!brl-tgr!tgr!GodFrom: God Newsgroups: net.unix-wizards Subject: Re: Hacking OS sources Message-ID: <6399@brl-tgr.ARPA> Date: Thu, 6-Dec-84 13:29:50 EST Article-I.D.: brl-tgr.6399 Posted: Thu Dec 6 13:29:50 1984 Date-Received: Sun, 9-Dec-84 03:23:45 EST Sender: news@brl-tgr.ARPA Organization: Ballistic Research Lab Lines: 35 RE: Buying and hacking VMS sources... I started to look into this to as we had some significant misfeature problems under VMS, like no CBREAK mode (ie. no XON/XOFF while passing chars reads, we want to use things like EMACS over our flow controlled tty net.) It seemed to me that: a) Yes, it is expensive but that's all relative. b) It's more than you think, as it is written in a few languages and you have to have and support all those languages just for this one purpose (eg. BLISS) c) I heard (tell me if I'm wrong) that a sysgen from source can take many many hours on a 780 and basically requires all those to be standalone. (I guess we could call buying a spare 780 just another expense.) In sum, the Turing Tarpit beckons, yes, you COULD do anything. The real point is that VMS was never MEANT or DESIGNED to be source supported by users as UNIX certainly is designed to be. Similarly, you COULD put the VMS manuals on-line but they were never designed to be on-line and would require something much more sophisticated than the UNIX 'man' command to make it useful. It's funny, TOPS-20 is widely source supported and the only thing that seemed to kill it was DEC, certainly not the users or the source sites. I am truly sorry VMS got off to such a bad start, but denying it's historical legacy doesn't fix it. -Barry Shein