Mixing and matching a couple of posts.... > On Friday 23 February 2001 14:14, Stephen White wrote: > > > > Just a brief side-track... but I'm seeing "XML" everywhere and I > > just don't click onto why everyone's talking about it. > > > > It seems to me that XML is a really weak SQL, with the difference > > that anything is supposed to be able to parse it... but ends up > > with everyone having to create and re-create a database in their > > code instead of just using SQL or an external database. XML and SQL aren't commensurate. In fact, very broadly viewed, they're sort of at opposite ends of things. SQL is a mechanism for getting at data; XML is a data storage standard (or data-storage-standard standard). SQL requires that there be a database to query; XML documents require that some tool exist to query them and process what's in them. (That's painted with awfully broad strokes, but I think the gist is accurate.) XML is both a supposedly (but I'm not convinced really) "simplified" version of its sort-of-parent, sort-of-sibling technology, SGML, and an exponentially more powerful way to define and process documents and data than HTML. On Sat, 24 Feb 2001, W. Kent Starr wrote: > It can also be thought of a a meta-language, in that, with XML you > define customized tags and attributes for these. This 'defining' is > done via a 'document type definition' (DTD). As long as you and I > write to the same DTD (which can be located anywhere on the > Internet) then we can exchange even highly customized data unique to > our respective needs. One addendum: XML (unlike SGML) doesn't actually require a DTD. I really hope people do continue to understand their value, though, in spite of XML's wimp-out option :-) I never found DTD-writing particularly difficult; I'm not entirely sure how they earned their reputation as a scary part of SGML. ANYWAY... A good-faith attempt to reconnect this with Ruby will now follow. Earlier this week I was looking at the WWW Consortium site and noticed, among other things, a newly-released working draft of XQuery, an XML query language. Not the first such language, but supposedly a very comprehensive one that improves on old ones. It occurred to me to wonder whether there might be a worthwhile opening here for the implementation of this standard in Ruby tools and/or libraries. This is the vaguest possible idea -- I haven't even really looked over the standard in detail, or thought about what would be involved -- but it struck me as an area of potential rubification. David -- David Alan Black home: dblack / candle.superlink.net work: blackdav / shu.edu Web: http://pirate.shu.edu/~blackdav