[dsdl-discuss] Fw: DSDL Part 5 (DTLL)

Ken

Seems I've been somewhat remiss. I was supposed to be passing this on to you
for publishing as a CD before I went on holiday earlier this month and
failed to do so. Can we get it out for a vote in time given that the fault
in late delivery has been a clerical error on my part?

    <doc n="(leave blank)" date="2006-02-27" project="is19757-5"
format="PDF">
    <title>Document Schema Definition Language (DSDL) Part 5 -
 Datatypes - Datatype Library Language (DTLL)</title>
    <short>CD - Document Schema Renaming Language</short>
    <requestor contact="brown"/>
    <status>Committee Draft (CD)</status>
    <action>For submission to ITTF for ballot</action>
    <distribution>SC34 and Liaisons</distribution>
   </doc>

Thanks

Martin

----- Original Message -----
From: "Alex Brown" <alexb@griffinbrown.co.uk>
To: "Martin Bryan" <martin@is-thought.co.uk>; "Jeni Tennison"
<jeni@jenitennison.com>
Cc: <francis@franciscave.com>
Sent: Saturday, February 11, 2006 4:21 PM
Subject: RE: DSDL Part 5 (DTLL)

Dear all,

> -----Original Message-----
> From: Martin Bryan [mailto:martin@is-thought.co.uk]
> Sent: 22 January 2006 13:35
> To: Alex Brown; Jeni Tennison
> Cc: francis@franciscave.com
> Subject: Re: DSDL Part 5 (DTLL)
>
> Alex
>
> Many thanks for this draft, but...
>
> >All being well, a version of this will become a draft international
> standard (DIS)
>
> Not so fast as you might like. What we have publish now is a
> Committee Draft, then a Final Committee Draft, and only then
> a DIS. So the first thing we need to do is to change the
> running head to replace "Draft International Standard" with
> "Committee Draft".

Argh, yes - I had elided my categories. I have made this change (though
the wording was hard-coded into James's stylesheets. BTW, any word on
WG2's work on these - the sourceforge project wasn't going last time I
looked)

> Then there are a number of changes that will need to be made,
> some of which should really be made before we put out the CD,
> and some which can be left until the FCD stage when we can
> updated them in response to UK comments on the CD. I've
> therefore split the following notes into two main groups:
> Changes needed now and Changes needed before the next edition:
>
> Changes Needed Now
> --------------------------------
> Need to add XPath 2.0 to the list of Normative References.
>
> Add statement explaining the convention used for
> documentation, referencing the Compact Syntax of Part 2 of
> this standard. In particular highlight that concepts such as
> ns are inherent in Part 2 and not specific to this part.
>
> Add some form of explanatory text to the start of Clause 4.

Okay, I have addressed these and attach a new draft.

I am working on a DTLL implementation that should aid my understanding
and benefit the editorial process too.

> Changes Needed Before Next Edition
> ------------------------------------------------------
> Wording of Foreword and Introduction needs extending to align
> with other parts and new ISO wording.
>
> Scope satement needs extending to provide a suitable abstract
> for the part.
>
> Assign unique identifiers to normative references so that
> references to them can be clearly identified (as in other parts)
>
> Add cross-reference to terms defined in Part 2 in Clause 3.
> Add DTLL to the list of terms defined.
>
> Clause 4: Replace namespace with one based on DSDL.ORG. Set
> Version Number to 1.0
>
> NB: I prefer the Clause 4 process of showing the productions
> then explaining them to the approach used elsewhere of
> explaining the productions and then defining them. We need to
> be consistent in what we do. I would suggest all elements and
> attributes should, whereveer possible, be formally defined
> before they are mentioned in the text.
>
> Clause 5: Change "elements includes" to "elements include".
> In last sentence of last paragraph change second occurrence
> of "can be used"
> to "can be referenced".
>
> Clause 6: In first para change "are" to "can be either". In
> definition of anonymous datatypes change "are" to "can be"
> and "cannot" to "does not need to". Could also add
> explanatory statement at end, such as "for example, because
> it is a subcomponent of a definition".
>
> Why is datatype-reference defined as xs:QName rather than
> ns:QName? This is the only case that xs is used. The
> difference between xs and ns is not explained. (Why does the
> preceding para have a cap N on Namespace? Is this somehow
> related to the use of xs rather than ns?)
>
> Clause 6.1: Change "a implementation" to "an implementation".
>
> Clause 7: Bracketed text at end of second paragraph should be a note.
>
> Clause 7.2.1: Bracketed text at end of first paragraph should
> be a note. In second paragraph both occurrences of
> "(document)" should be removed as they give the incorrect
> impression that each regex creates a new document whenever it
> identifies a match. The use of "element" for each named
> subexpression is also confusing. I would suggest "component"
> would be less ambiguous.
>
> Clause 7.2.1.2: Second sentence of first paragraph should be a note.
>
> Clause 10.2: The use of italicised name and regex in this
> clause is confusing. Elsewhere references to terms defined in
> the standard are not italicised. Why are they italicised here?
>
> Clause 10.3.3: Missing closing parentheses.

Thanks very much for these, I will queue them for action. But I guess
the priority now is to get the CD into the ISO machine.

- Alex.

------------------------------------
Alex Brown
Technical Director

Griffin Brown Digital Publishing Ltd
Orwell House
Cowley Road
Cambridge
CB4 0PP
United Kingdom

Tel +44(1223)425730
FAX +44(1223)425384
Web http://www.griffinbrown.co.uk/
------------------------------------

>>> How clean is your data? http://xmlprobe.com/ <<<

> ----- Original Message -----
> From: "Alex Brown" <alexb@griffinbrown.co.uk>
> To: "Jeni Tennison" <jeni@jenitennison.com>
> Cc: "Martin Bryan" <martin.bryan@csw.co.uk>; "Martin Bryan"
> <martin@is-thought.co.uk>; <francis@franciscave.com>
> Sent: Thursday, January 19, 2006 1:02 PM
> Subject: DSDL Part 5 (DTLL)
>
>
> Jeni hi
>
> We have been informed by ISO that there is an urgent need to
> get Part 5
> of DSDL into the standardisation process, and to this end I have been
> starting to turn DTLL 0.4 into a draft ISO standard as previously
> discussed.
>
> All being well, a version of this will become a draft international
> standard (DIS) which will then be subject to comment and one or more
> further rounds of drafting before eventual standardisation. I'm aware
> you have further modification in the pipeline but knowing that 'the
> excellent is the enemy of the good', and bearing ISO practicalities in
> mind, it is important now to get something on the table, even if we
> expect the final version to incorporate some substantive technical
> changes.
>
> I attach the draft as it is right now. This is based on your DTLL 0.4
> but with maps removed (these were felt by WG1 to be useful
> but likely to
> hamper initial uptake by making implementation too hard). I have also
> begun pulling the text about as required by ISO ... As well as bolting
> on ISO textual apparatus this also means trying to couch the spec in
> terms of terms which are all wholly defined somewhere. There are a few
> editorial comments, which I'd appreciate feedback on ...
>
> Francis, Martin - how far away is this from being a document
> the UK can
> submit to ISO?
>
> - Alex.
>
> ------------------------------------
> Alex Brown
> Technical Director
>
> Griffin Brown Digital Publishing Ltd
> Orwell House
> Cowley Road
> Cambridge
> CB4 0PP
> United Kingdom
>
> Tel +44(1223)425730
> FAX +44(1223)425384
> Web http://www.griffinbrown.co.uk/
> ------------------------------------
>
> >>> How clean is your data? http://xmlprobe.com/ <<<
>
>
>
>
>
>
>

--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)