From owner-freebsd-stable Mon Oct 21 17:31:47 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id B35AF37B401 for ; Mon, 21 Oct 2002 17:31:45 -0700 (PDT) Received: from mrburns.nildram.co.uk (mrburns.nildram.co.uk [195.112.4.54]) by mx1.FreeBSD.org (Postfix) with ESMTP id EE37843E6E for ; Mon, 21 Oct 2002 17:31:44 -0700 (PDT) (envelope-from muttley@gotadsl.co.uk) Received: from [192.168.0.4] (muttley.gotadsl.co.uk [213.208.123.26]) by mrburns.nildram.co.uk (Postfix) with ESMTP id 7B4321E1A92; Tue, 22 Oct 2002 01:31:37 +0100 (BST) Date: Tue, 22 Oct 2002 01:31:39 +0100 From: Matthew Whelan To: "R. David Murray" , FreeBSD-Stable Subject: Re: freebsd test matrix In-Reply-To: <20021021191920.U21141-100000@twirl.bitdance.com> References: <20021021235512.C6F1.MUTTLEY@gotadsl.co.uk> <20021021191920.U21141-100000@twirl.bitdance.com> Message-Id: <20021022012432.3866.MUTTLEY@gotadsl.co.uk> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: Becky! ver. 2.05.04 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Mon, 21 Oct 2002 19:22:23 -0400 (EDT) "R. David Murray" wrote: > On Tue, 22 Oct 2002, Matthew Whelan wrote: > > I couldn't agree more. Testing discovers bugs at a faster rate than > > debugging, at least until the test becomes 'too big'. As a result, it > > actually *SAVES* developer time. If this weren't the case, noone would > > do it outside of safety-critical systems. The clever bit is knowing when > > to stop (ie. how big is too big). Starting should be a no-brainer. > > The solution to the too big problem is to make many smaller tests, > and to have a test harness that allows you to run them selectively. Not quite what I meant... there always comes a point when testing where the rate at which you find bugs drops below the cost/benefit threshold of removing them. It's no good delivering a bug-free product 3 years after its usefulness has expired. It's also no good delivering that piece of perfection for triple the price anyone's willing to pay. > In XP, you write the tests *first*, and then write the code to make > the tests pass. This also saves developer time, in my experience (not > that I always do it that way, mind ). This is a good way of disguising the fact that most people don't do enough detailed design - you effectively encode the design in your test suite instead. -- Matthew Whelan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message