- From: Daniel Dardailler <danield@w3.org>
- Date: Thu, 10 Oct 2013 08:52:14 +0200
- To: Jirka Kosek <jirka@kosek.cz>
- CC: liam@w3.org, public-xml-core-wg@w3.org
On 10/10/2013 00:51, Jirka Kosek wrote: > On 9.10.2013 20:41, Liam R E Quin wrote: >> There was some interest a while ago in submitting XML Schema 1.1 as an >> ISO standard, which may help adoption in some countries. >> >> It's not a large amount of work (obviously, it's editorial work). > I would be very careful with this, there are several gotchas: Hello all > > 1) The first submission to PAS can be published as is, but any > subsequent versions should adhere to ISO terminology (SHALL instead of > MUST and so on) -- which is a lot of work. Let me check on this one. Last I heard, it was mostly about a change of stylesheet, after 5 years. (it may be more that just styling, but I'd expect it to be an automatic process of some kind, and since we may submit new revision/version before 5 years, it may be a moot point anyway) > > 2) Who will be maintaining ISO version of XML schema? Usual target will > be ISO/IEC JTC1/SC34/WG1. Without close work between WG1 and XML Schema > WG (XML Core WG) there can be divergences introduced in a long term. > Ideally there should be some agreement between W3C and ISO stating that > all maintenance will be handled at W3C to avoid such possible > divergences. But then W3C has to guarantee timely processing of comments > channelled through ISO process. In our PAS submitter "contract", we retain full maintenance, and therefore, we do not submit our REC to a specific TC but to JTC 1 directly (it's a new rule in the PAS system) > > 3) XML Schema Recommendation is written in a unnecessary complex way. > There is a risk that some national bodies will raise comments against > this during PAS process. With an assumption that W3C doesn't want > retroactively change XML Schema 1.1 spec and doesn't want misaligned ISO > version this could lead to PAS voting failure. This probably will not be > a very good marketing for XML Schema. I don't think that this would be a matter of concern, it's more a political vote (see above). > > 4) Also with just two implementations, XML Schema 1.1 might be too > premature to be taken to ISO. Market adoption/acceptance is indeed an important criteria. Checking this is part of the PAS preparation report, and it may indeed be abandonned if there is not enough traction for it. > > In general it would be a lot of paperwork and boring editorial work. If > it will be successful, users will be able to buy useless PDF rendering > of XML Schema spec from ISO in two years. I don't think it is worth > effort. And in general there are many existing precedents where > publishing one specification by two standardization bodies introduced > maintenance issues. This section would seem to apply to all our PAS efforts, not just XML Schema, so this is not really the right forum to discuss it (for those of you who have member access, you can look at https://www.w3.org/2010/07/potential-pas.html). > > > Jirka >
Received on Thursday, 10 October 2013 06:52:43 UTC