Return-Path: X-Original-To: apmail-incubator-ooo-dev-archive@minotaur.apache.org Delivered-To: apmail-incubator-ooo-dev-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 36044928B for ; Mon, 23 Apr 2012 19:26:37 +0000 (UTC) Received: (qmail 91330 invoked by uid 500); 23 Apr 2012 19:26:36 -0000 Delivered-To: apmail-incubator-ooo-dev-archive@incubator.apache.org Received: (qmail 91254 invoked by uid 500); 23 Apr 2012 19:26:36 -0000 Mailing-List: contact ooo-dev-help@incubator.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: ooo-dev@incubator.apache.org Delivered-To: mailing list ooo-dev@incubator.apache.org Received: (qmail 91243 invoked by uid 99); 23 Apr 2012 19:26:36 -0000 Received: from minotaur.apache.org (HELO minotaur.apache.org) (140.211.11.9) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 23 Apr 2012 19:26:36 +0000 Received: from localhost (HELO mail-vx0-f175.google.com) (127.0.0.1) (smtp-auth username robweir, mechanism plain) by minotaur.apache.org (qpsmtpd/0.29) with ESMTP; Mon, 23 Apr 2012 19:26:36 +0000 Received: by vcbfl13 with SMTP id fl13so9333764vcb.6 for ; Mon, 23 Apr 2012 12:26:35 -0700 (PDT) MIME-Version: 1.0 Received: by 10.52.22.162 with SMTP id e2mr14218924vdf.76.1335209195535; Mon, 23 Apr 2012 12:26:35 -0700 (PDT) Received: by 10.220.118.147 with HTTP; Mon, 23 Apr 2012 12:26:35 -0700 (PDT) In-Reply-To: References: <4F95803C.4060304@gmail.com> <4F95A6DF.8050609@apache.org> Date: Mon, 23 Apr 2012 15:26:35 -0400 Message-ID: Subject: Re: Unifying download logic, especially from NL pages? From: Rob Weir To: ooo-dev@incubator.apache.org Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Mon, Apr 23, 2012 at 3:22 PM, Rob Weir wrote: > On Mon, Apr 23, 2012 at 3:00 PM, Andrea Pescetti wr= ote: >> Rob Weir wrote: >>> >>> It is the NL pages that I >>> am concerned about, since they have hard-coded logic that differs from >>> the above pages. >> >> >> The Italian one is probably the worst at the moment, since it has hard-c= oded >> links (to the Italian mirror; this is a relatively recent change to avoi= d >> depending on the changing mirror situation), hard-coded version numbers = in >> the download page and even a hard-coded directory for the download page. >> >> While I see it changing (at the very least, the download link will be >> pointing to the SF/MirroBrain URL and not to a specific mirror), I'm not >> sure that the convenience of an API to generate links will always outwei= gh >> the added complexity: a link to the 'latest' version would have to be >> changed, for example, if a version-specific security patch is released, = like >> it happened for 3.3.0. >> > > I think the convenience will always outweigh. =C2=A0Remember, we can neve= r > reduce the real world complexity (essential complexity) any more than > it is in the real world. > > All we can do are two things: > > 1) Put all of the real world complexity in one place so it can be > centrally managed, updated, debugged, etc. > > 2) Reduce the accidental complexity > > Since the essential complexity can never exceed the total complexity, > centralizing this logic will always be a net improvement. > (Sorry, unstated assumption I should make: It is not true that all users who want the Italian version will come through the Italian NL page. Therefore if there ever were some special rule that needed to be defined for the Italian download, it would need to be triggered from four different places: 1) Italian NL page 2) www.openoffice.org page 3) download.openoffice.org page 3) download.openoffice.org/other.html page This would be very ugly. So even in a world where you think there may be special rules that apply to only a single NL, it still makes sense to centralize that special rule into a centrally managed script. That reduces the net complexity of the system.) > -Rob > >> Regards, >> =C2=A0Andrea.