- From: Geoffrey Sneddon <geoffers@gmail.com>
- Date: Sat, 15 Nov 2008 18:39:27 +0000
- To: Matt Mullenweg <m@mullenweg.com>
- Cc: www-archive <www-archive@w3.org>
On 24 Sep 2008, at 21:01, Matt Mullenweg wrote: > [Quote removed as this is publicly archived; I would've archived my > original email had it been possible.] I can see no way to improve the XML export without breaking backwards compatibility as WXR is so far from XML. It is also impossible to change the version in the URI you use as a namespace, as currently even unknown major versions are attempted to be parsed. Also, just because it works and you say it is XML doesn't make it XML: I could say HTML is SGML, just all browsers are broken in some edge- cases. The fact that WXR would break if it were treated as XML, similar to HTML breaking if treated to SGML, rather disproves it is XML. If I were a cynic, I would suggest that WXR deliberately didn't use a feed parser (you have one in every copy of WP, it's called MagpieRSS and works fine for RSS such as WXR) or a XML parser (there's one of those in PHP too, <http://php.net/xml>) so that you could claim there was no vendor lock-in, while coming up with a format nothing can parse (and especially for anything not licensed under the GPL to remain fully compatible, being unable to copy any code). In fact, seeming you had a perfectly fine feed parser for what you needed, I would say how WXR is implemented is more complex than it had to be, I am almost such a cynic. Lastly, unlike your claims of it just being edge cases you fail to support, there are plenty very large omissions: XML requires support for UTF-8 and UTF-16, of which you only attempt to support UTF-8; fatal errors on non well-formed XML; to name but two. -- Geoffrey Sneddon <http://gsnedders.com/>
Received on Saturday, 15 November 2008 18:40:06 UTC