- From: Jonathan Robie <jonathan.robie@redhat.com>
- Date: Thu, 19 Jul 2007 13:17:01 -0400
- To: Harry Halpin <hhalpin@ibiblio.org>
- CC: Dan Connolly <connolly@w3.org>, "C. M. Sperberg-McQueen" <cmsmcq@acm.org>, Jim Melton <jim.melton@acm.org>, Andrew Eisenberg <andrew.eisenberg@us.ibm.com>, public-grddl-comments@w3.org, w3c-xsl-query@w3.org
Harry Halpin wrote: > XQuery is a wonderful language that has my full support, as I have > mentioned to you before. However, current GRDDL implementations all > currently support only XSLT 1 [1], and so we do not mention XQuery per > se in the specification. Is the goal of the document to specify GRDDL or to describe current implementations? > However, the specification is written and the > Working Group decided (issue #which-langs [2]) to allow different > transformation languages, such as XQuery, to be used.. Yet as current > GRDDL transformations "in the wild" (as in, before GRDDL entered W3C > process) have been written in XSLT, the specificiation uses XSLT in its > examples. As language use changes over time, it is a possibility that > XQuery will be used for more and more GRDDL transformations, and I for > one would be happy to see this, but at this point it may not necessitate > a change in the specification. > For one thing, I think your specification is currently vague. Does GRDDL require that an implementation support XSLT 1.0? XSLT 2.0? Does it require that an implementation support any other language? Is there a way to specify the name of the language being used? Since GRDDL is intended to be a standard, and the only thing that it does is transformations, what interoperability do you expect among implementations that specify their transformations in particular languages? If you intend to allow only one language, say so. Currently, you do not give enough information to be able to do this with If you intend to support more than one language, say how this is done - for instance, you seem to have the assumption that a transformation is identified by a URL, and I am guessing you expect the text of the transformation, e.g. the XSLT stylesheet, to be physically located there. True? You say this can also be done with C - how? Should I download the program text, compile it, and use it, or are compiled languages really out of scope for this mechanism? Also, you didn't really respond to my original statement. Here's the text that I was responding to: > While technically Javascript, C, or virtually any other programming > language may be used to express transformations for GRDDL, XSLT is > specifically designed to express XML to XML transformations and has > some good safety characteristics. This is not really normative text - it gives no clues about how to use Javascript or C for such transformations. It argues that XSLT is the right language because it can do two things that XQuery was also designed to do, but does not mention XQuery, which seems to indicate that the WG hasn't thought about XQuery, or thinks it unimportant, or whatever. I think it would be simple to change this to say "XSLT and XQuery are specifically designed to express XML to XML transformations and has some good safety characteristics" or add a sentence like: "XQuery, another language designed specifically for such tasks, is not yet used in existing GRDDL implementations." Beyond that, though, I think you really do need to either say clearly that transformations are expressed in XSLT (in which case, just delete the above paragraph) or flesh out how they could be specified in other languages. Jonathan P.S., as you probably guess, I'm not satisfied with your answer.
Received on Thursday, 19 July 2007 17:18:12 UTC