Why are XQueries relegated to 2nd class status?

Let me start by saying that I am very grateful to see the progress being made
now on XML queries, and appreciate the great deal of effort that has gone into
developing the functional specification for the query language.  I want to make
it clear that I have no concerns with the functional specification for XQuery as
described at "http://www.w3.org/TR/xquery/", but I am disappointed that the
syntax that has been published is SQL-based rather than XML-based.

I am aware that there is a stated intention to create an equivalent query syntax
in XML.  I am also aware that at least one talk on this work given at XML 2000
in Dec 2000 made no mention of the possibility of an alternative XML syntax, and
this apparent bias towards SQL syntax (whether intended or not) makes me uneasy.
A definitive XML syntax that arrives within a week of the definitive SQL-based
syntax would be OK, as use of the XML-syntax only could be mandated.  An XML
syntax that arrives 1 or 2 years after the SQL-based syntax would simply make
the situation unmanageable.

I would be far happier if the XML syntax for the queries were created first, and
an SQL-ish version created afterwards for those who really insist on doing
everything in "vi".  Let's face it, like many other programming languages, SQL
has the kind of syntax you get when the only decent editor you have around is
"vi", "Emacs" or "notepad" - i.e. a text editor.  In a world where there now are
really good XML editors available, non-XML-based syntaxes do little except
relegate files to second-class status, and this is what worries me most about
FLWR.  OK, I am sure all of the Query WG are hard core experts in SQL, and so it
must seem appealing to continue to have an SQL-like syntax for queries, but I
sit in a position where I have to worry about the bigger picture of how to deal
with XML and all file types relating to it, and the problem with FLWR is that it
requires a separate, non-XML parser.  The fact that someone might write such a
parser would hardly be any comfort either, because if we had to install new
parsers for every different XML-related file format we deal with, it would be a
nightmare.  The XSL-T people made the right decision in using an XML-based
syntax.  Sure, "vi" users may not like it so much, but that just means there
should be better editors/GUIs.  The great thing about XSL-T is that it can be
automatically processed itself as XML, which is a big plus in a large production
environment.  The difficulty in doing the same thing with FLWR will relegate
XQuery to lesser status in the XML world, and that would be a great pity indeed,
because good and standard query tools something we need and want.

I would strongly urge you to consider delaying further work on the FLWR syntax,
and instead consider whether the XML-based syntax might not be the better one to
focus on, looking ahead 2-5 years.

     Cheers,
          Tony.
========
Anthony B. Coates
Leader of XML Architecture & Design
Chief Technology Office
Reuters Plc, London.
tony.coates@reuters.com
========


-----------------------------------------------------------------
        Visit our Internet site at http://www.reuters.com

Any views expressed in this message are those of  the  individual
sender,  except  where  the sender specifically states them to be
the views of Reuters Ltd.

Received on Monday, 19 February 2001 07:40:27 UTC