[dsdl-comment] Re: [davidc@nag.co.uk: Re: Re: Entities for XML]

Without having seen the draft, I have a few comments. I made one of the
original ISO Entity to Unicode mappings, which was used by several
vendors such as Arbortext and still occassionally crops up. I stopped
maintaining it early, because there were better experts doing this and I
am largely out of typesetting. A few years ago, I made an effort to make
a draft for standard set, but I found that the handful of differences
between the various mappings in adoption had quite a lot of traction by
their users and installed base. (The lion's share of entities have
obvious and agreed mappings.)

I still think it would be a good idea to have a standard mapping. I
don't see that W3C needs to do anything other than quote the ISO
copyright. However, I would suggest:

1) Major variant mappings should be fully noted, and where there are
 significant variants, the use of that entity should be deprecated and
unambiguous alternatives provided.

2) Look at using ISO DSRL mechanism to handle the mappings and variants?

3) It certainly would be friendly for W3C to acknowledge Anders etc. and
we need to value friendliness and collegiality.

4) We should actually look at making W3C the maintenance body for
character entity sets.

Cheers
Rick Jelliffe

On Thu, 2007-11-15 at 19:21 -0500, John Cowan wrote:
> [I have posted these at the request of David Carlisle, who for
> whatever reason cannot post to this list.]
>
> ----- Forwarded message from David Carlisle <davidc@nag.co.uk> -----
>
> Date: Thu, 15 Nov 2007 23:39:50 GMT
> From: David Carlisle <davidc@nag.co.uk>
> To: cowan@ccil.org
> Subject: Re: [dsdl-comment] Re: Entities for XML
>
>
> Thanks, I'd appreciate it if you could send the following as apparently
> I can't post to the list. I've spotted more than a few typos and
> probably would reword a few things said in haste, but I think the
> replies should be archived as they were so I haven't edited them.
> I'd appreciate it if you could post the following (and this covering
> paragraph). For what it's worth, I've updated the reference section in
> the draft under discussion....
>
> David
>
> Date: 15 Nov 2007 20:15:36 +0000
> From: David Carlisle <davidc@nag.co.uk>
> To: martin@is-thought.co.uk
> CC: dsdl-comment@dsdl.org, member-math@w3.org
> In-reply-to: <39A27A6184714BA5834DB38884B41746@martinsPC>
> Subject: Re: Entities for XML
>
>
> > I note with dismay, but not with surprise, that once again a W3C
> > recommendation based on earlier and extensive work by ISO carries not only
> > no reference to the source of the material but no acknowledgement of where
> > copyrighted material has been stolen from. I certainly do not want my name
> > associated with such illegal activities.
>
> Hey, I think that's rather unfair, firstly it;s not a w3c recommendation it's a
> private first draft of an update almost all of the work of which has
> gone into the technical details of the lists and mappings to unicode.
> The text of the covering document is a minimal skeleton at
> present. Clearly it will need to reference the iso specs as well. and
> __already__ the actual entity files all say:
>
>
> (C) International Organization for Standardization 2003
> Permission to copy in any form is granted for use with
> conforming SGML systems and applications as defined in
> ISO 8879, provided this notice is included in all copies.
>
>
> which is hardly a lack of reference.
>
> I have no idea what you think has been "stolen" certainly not the
> mappings to unicode (which is the sole point of this proposed spec)
> as that was never part of any ISO spec.
>
> I think it is vitally important that if the ISO does have offical
> mappings to unicode that they match the ones mathml and other w3c specs
> use as having two incompatible specs would be worse than having no spec
> at all. However if the ISO decides to kill off the entity spec drafts on
> the ISO side so be it, I'll need to change the copyright notice from the
> ones we have in the (2003 and current) draft which proposes ISO
> copyright and ISO FPI to what we previously used in MathML which is a
> non ISI FPIf starting with "-" and a copyright notice of the form
>
>
>
>
> ...
> Entity names in this file are derived from files carrying the
> following notice:
>
> (C) International Organization for Standardization 1991
> Permission to copy in any form is granted for use with
> conforming SGML systems and applications as defined in
> ISO 8879, provided this notice is included in all copies.
>
> -->
>
>
>
> I must say, I'm rather shocked and disheartened by your respnse.
>
> David
>
>
>
> Date: 15 Nov 2007 20:50:06 +0000
> From: David Carlisle <davidc@nag.co.uk>
> To: cowan@ccil.org
> CC: dsdl-comment@dsdl.org, member-math@w3.org
> In-reply-to: <20071115202506.GF12525@mercury.ccil.org> (message from John
> Cowan on Thu, 15 Nov 2007 15:25:06 -0500)
> Subject: Re: [dsdl-comment] Re: Entities for XML
>
>
>
> > Let's take this as strongly worded advice to add such a reference
> > and an explanation that the entity sets with names beginning "iso-"
> > do in fact come from ISO.
>
> I'm not sure of the status of dsdl-comments as to whether it accepts
> posts from non subscribers. If not hopefully you'll forward.
> I'm a little perplexed to be discussing this pre-alpha draft on a new
> list when I just gave Martin a private look before we even asked the w3c
> if we could publish such a draft on rec track (the draft is hosted on a
> w3c site but as it says it currently has no standing at all).
>
> As for the iso provenance yes that has to be explict and is already very
> explicit in the entity files, just currently less so in the references
> section (which is almost all wrong anyway, at present) and one reason for
> re-contacting the ISO (in the form of Martin) _before_ making this draft
> public is to see if there is a chance of gettingthe iso entity activity
> restarted. If the entity sets can have ISO FPI and ISO copyright, if not
> they presumably have to have non-ISO "-" FPI and a copyright notice that
> (following the licence on the existing ISO 9573 files) says that teh
> names are copied from there,and contains a copy of the ISO notice.
>
> personally I'd much rather make the entity name to unicode mappings a
> joint w3c/iso publication as then it's just clearer for everybody, but
> if that can't work then (as we need these now for mathml3) we'll just
> have to press on.
>
> David
>
>
>
> Date: 15 Nov 2007 22:10:32 +0000
> From: David Carlisle <davidc@nag.co.uk>
> To: martin@is-thought.co.uk
> CC: dsdl-comment@dsdl.org,member-math@w3.org
> In-reply-to: <C93924917E794C949B54DF88F0A1EDDC@martinsPC>
> Subject: Re: [dsdl-comment] Re: Entities for XML
>
>
> Reply-To: "Martin Bryan" <martin@is-thought.co.uk>
> From: "Martin Bryan" <martin@is-thought.co.uk>
> Cc: "David Carlisle" <davidc@nag.co.uk>
> Date: Thu, 15 Nov 2007 21:30:35 -0000
> Organization: IS-Thought
>
> Exactly John. Simply acknowledging source as a copyright statement in a
> referenced file that is linked to the printable document is not sufficient
> to absolve someone from blame for copyright infringement. David must
> acknowledge the earlier work over many years by ISO members such as Anders
> Berglund if he is to publish this as a W3C Recommendation
>
> Martin
>
> Martin, I think your reaction is frankly bizarre, the references
> section is at best a skeleton and actually I specifically dropped
> references from this draft, which I repeat is only being seen by a wider
> audience because _you_ posted its URL to an apparemtly public list,
> because the form of the reference takes on a very different natuure
> depending on whether this document is intended to be an update to ISO
> 9573-13. If it's an ISO document with an ISO copyright and all the
> entity files have ISO copyriles on them, that is one thing, if on the
> other hand it's a purely W4C document then the form of refernce to the
> previous version of iso 9573-13 needs to take on a rather different
> form. I mailed you asking if you were still interested in co-editing and
> if you thought that the ISO activity on entities could be restarted.
> The draft hasn't been publically announced or lined to from any public
> page because it is not ready for public yet. to talk of copyright
> infringement is nonsense. It's on a public rather than W3C server
> but primarily that is so you (personally) could see it while in draft
> form. To talk of copyright infringement and theft is just nonsense.
>
> As far as work goes, these mappings have as a matter of necessity been
> maintained by the Math WG for almost 10 years, mapping the existing
> names (and they are just names in the existig ISO spec) through multiple
> incarnations of stix unicode submission then changes in unicode 3.1 3.2
> 4 5 and the current pipeline, _all_ of which involved code points that
> affect these character mappings. Clearly the origin of the names needs
> to be cited, but it is also clear where and by whom most of the work has
> been done. To try to portray this as the W3C jumping up and stealing
> some existing body of work is just nonsense.
>
>
>
> I see from
> http://lists.dsdl.org/dsdl-comment/2007-11/0005.html
> that this list is being archived publicly, but my replies are not
> appearing, so I'd appreciate it if you'd post all the messages that I
> have replied to you to the list, so people see both sides of the
> conversation. Given that you've made thie existence of this document
> public I will probably have to announce it sooner rather than later.
> which is a bit of a pain, as it isn't really ready to go out yet.
> But so be it.
>
>
> David
>
>
>
>
> ________________________________________________________________________
> The Numerical Algorithms Group Ltd is a company registered in England
> and Wales with company number 1249803. The registered office is:
> Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.
>
> This e-mail has been scanned for all viruses by Star. The service is
> powered by MessageLabs.
> ________________________________________________________________________
>
> ----- End forwarded message -----
>

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