From owner-freebsd-current Fri Apr 12 09:10:53 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA15019 for current-outgoing; Fri, 12 Apr 1996 09:10:53 -0700 (PDT) Received: from mongoose.bostic.com (bostic@mongoose.BSDI.COM [205.230.230.129]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id JAA15014 for ; Fri, 12 Apr 1996 09:10:46 -0700 (PDT) Received: (from bostic@localhost) by mongoose.bostic.com (8.7.4/8.7.3) id MAA13184; Fri, 12 Apr 1996 12:03:26 -0400 (EDT) Date: Fri, 12 Apr 1996 12:03:26 -0400 (EDT) From: Keith Bostic Message-Id: <199604121603.MAA13184@mongoose.bostic.com> To: mike.long@analog.com Subject: Re: terminfo-less ncurses Cc: current@FreeBSD.org, ncurses-list@netcom.com Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > All that aside, I don't think it should matter which of the two > description formats you use. Zeyd and Eric think termcap sucks; > that's their opinion. Many BSD users disagree, and *WILL* *NOT* *USE* > ncurses unless it can use termcap instead of terminfo, no matter what > Bostic has said. Ummmm, Bostic has been trying to remain resolutely silent on the issue, and he's certainly NOT made any statements about termcap. My only goal was get rid of the 4BSD version of curses -- its internal structure and some of its internal algorithms are poor. Whether that was done with termcap or terminfo makes no difference to me, although having terminfo support is good because it makes System V applications more portable to the free systems. Regardless, I was one of the people working to convince the ncurses maintainers to support both terminfo and termcap equally. And, as far as I know, the ncurses distribution now supports them both, and since Eric used the Berkeley code, it's going to be as fast as the historic implementations. I think the arguments are somewhat pointless -- since ncurses supports both, we can all use what we like. My guess is that BSDI will convert to terminfo in our base release because it's more powerful, but you can bet anything you want that we'll not break support for the user's local termcap files. Finally, I don't think it's worth extending termcap to support the rest of the terminfo capabilities. It seems as if that would make make termcap files less portable to other systems, and I don't see any strong reason to do so. --keith bostic@bsdi.com