Date: Sun, 3 Aug 1997 19:47:00 -0700 (PDT) From: Tom <tom@uniserve.com> To: Michael Smith <msmith@atrad.adelaide.edu.au> Cc: current@FreeBSD.ORG Subject: Re: Make this a relese coordinator decision (was Re: ports-current/packages-current discontinued) Message-ID: <Pine.BSF.3.96.970803193617.5004C-100000@shell.uniserve.com> In-Reply-To: <199708040151.LAA16754@genesis.atrad.adelaide.edu.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 4 Aug 1997, Michael Smith wrote: > > > Whacko. While we're at it, let's just rename this list "msdos-current". > > > > What? What exactly are you trying to say here? > > I'd have thought that was obvious; let me try a longer version that may > have a better chance of getting through. Please don't pick holes in > my illustration; look at the _fundamental_ issue involved : > > If you accept the statement above, then you are accepting that there > is no future in developing FreeBSD. > > Suppose I propose some new, useful, possibly performance-enhancing or > otherwise beneficial change. This change will only be introduced in > -current, where it will receive a healthy shakedown before it becomes > part of a release. > > The proposed change alters a number of features of the system's behaviour, > in a fashion widely regarded as beneficial; everyone agrees that it would > be a Good Thing. > > But wait! It creates a compatability problem with the ports > collection. All of a sudden, I am faced with the demand that I must > independantly test over a thousand ports for compatability, and abandon > my change if any fail. > > Now let us suppose that the feature was incorporated, and somehow the > ports adapted to deal with it. Then it is discovered that the feature > was misdesigned, and in fact should be done differently; remember, > this is a basic fact of software development. > > And now I have to test all the ports over again. > > Can you see how stupid this is? No. We are talking about application level stuff here. Unless you are changing an API, how could you break applications? Also, your arguement has nothing to do with the ports-current vs. ports-stable debate. Either way something is going to be broken. Also, developers have never been required to test ports. It is silly to suggest that anyone is requiring you to test ports. Let's look at some real examples: most ports don't build under 2.1.7.1. Why? Mainly due to difference in bsd.port.mk, and "install" (grew some new flags in 2.2). Many of the ports will build on 2.1 with a new "install" and bsd.port.mk. It is easy to predict that changes to bsd.port.mk and "install" ports. > -- > ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ > ]] Genesis Software genesis@gsoft.com.au [[ > ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ > ]] realtime instrument control. (ph) +61-8-8267-3493 [[ > ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ > Tom
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.970803193617.5004C-100000>