Return-Path: pg_adm@postgres.berkeley.edu Received: by postgres.Berkeley.EDU (5.61/1.29) id AA05669; Fri, 22 Jan 93 11:39:33 -0800 Message-Id: <9301221939.AA05669@postgres.Berkeley.EDU> From: Marc Teitelbaum Subject: Re: Concern about new building environment To: postgres@postgres.berkeley.edu Sender: pg_adm@postgres.berkeley.edu In-Reply-To: Your message of Fri, 22 Jan 93 09:22:51. Date: Fri, 22 Jan 93 11:39:05 PST The new make is of course being shipped directly with the postgres source and will compile and run on the platforms that we support. When porting to a new platform you will have to make sure, obviousely, that the bmake compiles and runs there too (its directly in the postgres source tree with a V7 bootstraping makefile). The code is a whole lot more portable than the postgres code itself, so if you have any problems porting bmake, you can be sure you'll be spending 10 to 15 times as much time porting the postgres code - so i wouldn't worry :) But, like I said, it comes working with any supported platform. It's a much better solution, IMHO, than the imake solution that X uses (in retrospect that is, I'm not bashing X for trying the imake scheme). The make code does what we want and is freely redistributable and available off uunet. I haven't seen the GNU make, (and sorry john or rms) but the copyleft IS a restrictive license compared to the freely redistributable BSD license on bmake (actually, BSD just calls it make - *we* call it bmake). When offered with a choice between a copyleft program and a freely redistributable program of comparable quality, i will choose the latter as it causes less headaches for the end users who wish to reship the code. Putting a copyleft on a program is a personal (and political) statement that anyone can make, but no one should be forced to make. The university policy is to make code freely redistributable with no strings and I support that when there's no strong reason not to. Marc