Re: [xml-dev] Transformation of RelaxNG syntax into presentationof requi

[Chin Chee-Kai <cheekai@softml.net>]

Rick Jelliffe wrote:
> Actually, the current implementation works by renaming the instance, 
> not the schema. The same DSRL script could be used to rename schemas, 
> but I don't know that anyone has done this.
You did this @:    
http://www.oreillynet.com/xml/blog/2008/06/a_dsrl_script_for_mapping_from.html

and didn't do any illustration about instance.  A blob in your blog?


> There is an idea floating from certain ODF-ers around that UOF exists 
> because of what are called "political reasons". I think it is more 
> than that: UOF meets certain objective Chinese technical requirements. 
> I hope these can be added to ODF (and to OOXML too). But getting an 
> awareness of these issues is the first step.  Sun's Scott McNealy 
> called for a merger of ODF and UOF, so there is certainly good talk; 
> Novel (IIRC) have added grid layout support to OpenOffice, so there is 
> some action there, but not much.
I think you brought up a mix of interesting and very related issues 
whose answers, if any, are never going to be quite as straight forward 
as a one-liner.  Ultimately, it would be the users who decide whether a 
standard is useful.

The fact that major supporters and head-honchos of  existing ISO 
standards are paying attention and trying to influence its growth path 
reflects the potential effect UOF has on the community of developers, 
users and market.  Those who are already "there" in ISO's camp would 
point to the potentially vast overlap of specifications UOF has and hint 
at its redundancy.  But as you pointed out, specific technical needs by 
the Chinese people which do not so much as to affect other users would 
be less likely accepted by those who are "there", thus frustrating  the 
Chinese and pleasing no one.

On the other hand, with UOF, the Chinese has the ability to decide the 
necessary technicalities applicable to their usage.  It may be just word 
editing/presentation for now as a start, but could potentially be 
expanded in future to useful things that electronically reflect Chinese 
culture.  That's just my guess, but the point is that Chinese could then 
decide on the things relevant to them, instead of having to convince the 
other 80% of world population about why Chinese needs certain technical 
features and why all users must accept those same features even though 
it implies more cost (of development, training, delay, etc) with no 
visible gain to them.


> So I would reverse the question: rather than saying "UOF exists 
> because of politicial reasons" (or unco-operativeness) perhaps we 
> should ask "Why hasn't there been effective Chinese involvement in the 
> ODF process in its core formative days?"  Part of the reason must be 
> language. (And there are many others, opportunities, contacts, 
> scheduling, perceived benefit).
One could hurl many negatives based on subjective inclinations, but the 
arguments become very weak and do not stand up well because there is no 
proof and no consensus.  What does it really mean if a standard proposal 
is developed "without political reasons"?  ODF?  If ODF is so useful, it 
doesn't have to seek ISO's recognition but through sheer openness, 
free-of-charge and word-of-mouth, it would become the de facto.  OOXML?  
It's a fact that it is already a dominant standard whether or not it has 
ISO status.  So these are non-politically motivated?

You may be right in asking the reverse question.  However, I think it's 
good for historical study on standards development rather than forging 
ahead. Now that there is a well-primed candidate UOF, we should be 
looking at how to help make UOF accepted as an ISO standard, such as 
fast-tracking and receiving ISO status in record time.  (deja vu?)  ISO 
has certainly had a practice on that already, so it should be do-able.


> But who would use a format in simplified Chinese?  The PRC. Singapore 
> also uses simplified characters, but it is multilingual, and you would 
> expect English-language names to be more practical.  Taiwan treasures 
> its use of traditional characters somewhat, so I expect there would be 
> mixed feelings about UOF. HK: who knows?   So a standard that is only 
> usable by PRC mainland, HK and Taiwan will have problems getting 
> positive votes from other National Bodies:
No, that's not the way I would look at it.  If you're talking about the 
specification document itself being in simplified Chinese and 
non-Chinese-speaking developers cannot read them, it is a non-issue.  If 
a Chinese literature as cryptic as the Art of War gets translated 
multiple times into English and other languages just by its sheer 
content value, translation of the UOF specification is not going to be 
the main problem if UOF is important enough to be understood by 
non-Chinese-speaking users.  The UOF spec, in its present form, might be 
in simplified Chinese due to the need by many Chinese standards 
participants.  There's no saying that it cannot be translated into 
English and/or other languages.

Singapore's usage of Mandarin, electronically, socially and officially, 
follows those of China.  Singapore's choice is for practical and proper 
reasons such as trade, culture, history, etc.  Linguistically, China 
defines the Chinese language.  So it makes sense that if China adopts 
UOF, Singapore is likely to follow.  And no, you can't say 
English-language names are more practical for Singapore, because 
Mandarin is an official language in the country.

Usage of Mandarin in other Chinese cities take cue from China.  If UOF 
becomes China's standard, we can expect its influence on other 
Chinese-speaking cities.  Note that I'm talking about the format usage, 
not the translation of UOF as this is not an issue as per what I 
described above.


> consider that OOXML only got in because of its market dominance, 
> overcoming the general desire not to multiply standards.
Good point.  ISO has proven by action that its objective is not to "not 
multiply standards".  Furthermore,  a standards body's objective cannot 
be such as to seek "not to multiply standard", for otherwise who would 
allow a subset of a richer, more descriptive language to be extracted as 
a separate standard?  Yes, I'm talking about XML from SGML.

> Usually, I would say it would be more realistic to expect that UOF's 
> Chinese requirements would get added to ODF. However, the plans for 
> ODF 1.3 (or ODF 2.0) have not been made yet, so that could well be 4 
> or 5 years in the future.
Seeking to address user community requirements would be more relevant 
than an all-inclusive ODF.

Let UOF become the ISO definitive for all electronic representation of 
Chinese cultural and linguistic content, and let  ODF seek its open 
source objectives.  Tweaking next generation of ODF to include 
requirements  from China (and not other countries) and in the process 
ironing out the relevance of UOF more and more is simply diluting ODF's 
focus and killing China's well-mannered effort in UOF serves no useful 
purpose to anybody.


> Until then, I think it would be very productive for a translation of 
> UOF to made into an ISO Technical Report: this is less than a standard 
> (and could lead the way to becoming an international standard), it 
> gets the information out there, will be less contentious than a full 
> standards process, and can be done fairly quickly. I think that would 
> be really useful: there is almost no material on UOF in English. I am 
> told a lot of it is based on an early fork of ODF, so probably  much 
> of the ODF material could be re-used, which might help translators!
Translation of UOF to other languages so more people  could understand 
it, yes, I think it's a good call.  But awarding it a second prize as a 
"goodwill gesture" of not awarding it the official status it deserves is 
not needed, because as I said, ISO has proven its capability to fast 
track 6,000+ pages of content which apparently proposes to duplicate an 
ISO specification.

UOF is a deserving and serious candidate.  Let it mature, and then let 
the community of users and developers decide.


regards,
Chin Chee-Kai