Return-Path: Delivered-To: apmail-xml-general-archive@xml.apache.org Received: (qmail 31251 invoked by uid 500); 6 Feb 2002 10:32:46 -0000 Mailing-List: contact general-help@xml.apache.org; run by ezmlm Precedence: bulk list-help: list-unsubscribe: list-post: Reply-To: general@xml.apache.org Delivered-To: mailing list general@xml.apache.org Received: (qmail 31230 invoked from network); 6 Feb 2002 10:32:45 -0000 Content-Type: text/plain; charset="iso-8859-1" From: Guillaume Rousse Organization: =?iso8859-1?q?Universit=E9=20Pierre=20&=20Marie?= Curie To: general@xml.apache.org Subject: Re: [3rd party jar] w3c Date: Wed, 6 Feb 2002 11:30:25 +0100 X-Mailer: KMail [version 1.3.2] References: <20020205181745.30187.qmail@web12108.mail.yahoo.com> In-Reply-To: <20020205181745.30187.qmail@web12108.mail.yahoo.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: <20020206103025.7BEC1420@silbermann.snv.jussieu.fr> X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N Ainsi parlait Shane Curcuru : > ---- you Guillaume Rousse wrote ---- > > > Yes, it requires a net connection, but from > > where did you get your initial source tarball anyway ? > > Well, a couple of points: > -- The initial download requires a net connection somewhere, but many > folks only download once and then work on that copy for a while. I > don't mind that, but I would mind having to wait for the net connection > for every time I recompile something. (While having to do this only > the first time you build a new download is better, it's still quite > annoying and is difficult to document). > > -- What about the developer who works for a larger organization who got > the distro from some internal network at their site? That individual > may *never* actually have an external net connection - they may only > work on an intranet, where some other member on the team downloaded the > sources or distros for them to an internal server. Don't laugh, I have > dealt with exactly this kind of situation in the past. I know those kind of situation, i usually donwload many things from my desktop at university, and work offline on my laptop. And i've been badly suprised by some build using network connections, for additional sources or for javadoc linking :-) However, i was proposing to make it a distinct task, so checking dependencies could been made as a part of initial download. And a consequent README file would allow to check if you actually need to proceed this task. FYIThis is also standard src rpm behaviour: you download them, and only at build time it will bomb you out for missing build dependencies. However, you can also query the rpm for checking them. Anyway, having each used jar as an overridable property, whereas they are included or not, would be a big help for people wanting to build from external ones. Currently, we have to: - delete all provided jars, to make sure they not interfere - include installed one in classpath - deal with ant complaints about missing files. > I guess in the larger scale, this is part of a meta-issue of the > overall usability of Apache projects for any potential user out there - > including non-technical users. > My preference is to try to do the little extra bit of coordination and > work on our part to enable those folks who *don't* happen to already be > expert developers to be able to quickly and easily use our projects. > But that's just my perspective. 8-) My opinion on this subject is it overcomes a single (or even a group of single) software project capacities, and should be delegated (at least to some extent) to distribution project. Thus the need to cooperate. -- Guillaume Rousse GPG key http://lis.snv.jussieu.fr/~rousse/gpgkey.html --------------------------------------------------------------------- In case of troubles, e-mail: webmaster@xml.apache.org To unsubscribe, e-mail: general-unsubscribe@xml.apache.org For additional commands, e-mail: general-help@xml.apache.org