From maine@cello.qnet.com Fri Jun 9 03:14:19 2000 Received: from cello.qnet.com (56k-palm-01-09.dial.qnet.com [209.221.196.72]) by dkuug.dk (8.9.2/8.9.2) with ESMTP id DAA23049 for ; Fri, 9 Jun 2000 03:14:17 +0200 (CEST) (envelope-from maine@cello.qnet.com) Received: (from maine@localhost) by cello.qnet.com (8.9.3/8.9.3) id SAA00902; Thu, 8 Jun 2000 18:15:29 -0700 From: Richard Maine MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14656.17713.847820.623862@vega.qnet.com> Date: Thu, 8 Jun 2000 18:15:29 -0700 (PDT) To: sc22wg5@dkuug.dk Subject: (SC22WG5.1848) SC22WG5.1842) N1385 AFNOR proposal for August WG5 meeting In-Reply-To: <200006080207.EAA19090@dkuug.dk> References: <200006080207.EAA19090@dkuug.dk> X-Mailer: VM 6.72 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid Robert Corbett writes: [re: XDR files] > It is currently RFC 1832. Thanks. I took a quick look at the RFC. I don't see anything at all defining a data file format there. There are parts that could be incorporated into a data file, but as best as I can tell from looking at the RFC, just defining a file to be an XDR_UNFORMATTED file would be a completely inadequate definition and would give no guarantee of any kind of portability. Its not even clear to me what approach one would best take to such a project. The RFC describes a data definition language. Would one actually do something with that language (perhaps something analogous to a format specification), or would one presume that the XDR specification was implicit in the types of the data being read/written? Yes, there is material there that could be used. But there would be a lot of work yet needed to do something with this. After looking at this, I'd say that Tony's one meeting estimate is low. I'd have perhaps believed that if XDR actually defined a data file format, and all that was necessary was to bind that to Fortran. -- Richard Maine maine@qnet.com