[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Restrictions on existence of attributes?

• To: "xml-dev@l... Developers" <xml-dev@l...>
• Subject: Re: Restrictions on existence of attributes?
• From: David Carver <d_a_carver@y...>
• Date: Fri, 30 Jun 2006 14:23:23 -0400
• Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Received:Message-ID:Date:From:Reply-To:User-Agent:MIME-Version:To:Subject:References:In-Reply-To:Content-Type:Content-Transfer-Encoding; b=YvuPUUDO77QfXVGJCbp5EjxqiGi/IAzMM0m5EDxjmNlHxMFH73hhtcX6sQ77MTnGvulwKvgqzpVXic3EFooWytYlD4NO39Qr8AWYzc0HZbzgxhi/bAVHqwfbLojbCokniB/FMDWqFn/54N62I/zgytGfN4POvxoCefAY/cQ4eYs= ;
• In-reply-to: <1151683981.2960.5.camel@marge>
• References: <7411F30464DC9C479FB14CFD348D71D97BBA91@U...> <ABAEF25D-573A-44BA-B8B9-7540FA85463D@e...> <1151683981.2960.5.camel@marge>
• Reply-to: d_a_carver@y...
• User-agent: Thunderbird 1.5.0.4 (Windows/20060516)

Here is the thing, as I've been following this debate for the last two 
days, and truthfully, it is going no where.   XSD is here to stay for 
the foreseeable future, it isn't go to go away any time soon.   While it 
does have it's issues, the vast majority of XML Schema that are produced 
today, and the foreseable future is going to be XSD.   UNCEFACT ATG2 
NDRs currently only support W3C Schema as their output for XML for Core 
Components.    The big B2B standards like UBL, OAGIS, HR-XML, Global 
Justice etc are going to be using XSD because that is what their 
membership is asking.

RelaxNG may be the greatest thing since sliced bread, but until it 
reaches it's way into main stream, it will be a niche area.  XSD 
currently does the 80/20 rule fairly well, and for about 90% of the 
people and implementations, it works well, and that 80% is 
interoperable.   I work on creating XML Schema everyday, and make sure 
that I validate the schema against the validtors that are most in use 
(spy, xerces, ms xml .net).

Data binding tools have their own issues, as even they don't necessarily 
produce XML that validates against a schema even though the objects were 
created from it.  JAXB 1.0 was notrious for this, and JAXB 2.0 still has 
issues.

So, basically, neither side is going to convince the other, and Relax NG 
is going to fit those needs where W3C schemas don't work.

Dave

Dave Pawson wrote:
> On Fri, 2006-06-30 at 18:03 +0200, Robin Berjon wrote:
>
>   
>> But as a user confronted with the choice between a number of systems,  
>> given a choice between using a solution that offer RelaxNG and one  
>> that offers XML Schema, I'll pick the former with my eyes closed as I  
>> know it'll cut my costs (and having jumped through the hoops of using  
>> XML Schema, I have a notion of what those costs are).
>>     
> And as cost conscious managers employ cheaper labour....
>    Relaxng is a no brainer.
> It's readable.
> It's documentable.
> It's generatable. (Sorry Len the English is going)
> It's designed by two people I|We(??) have a lot of respect for.
> It's a winner.
>
> The only thing xsd is good for is showing how not to do it.
>
>
>   

• Follow-Ups:
  • RE: Restrictions on existence of attributes?
    • From: "Michael Kay" <mike@s...>

• References:
  • RE: Restrictions on existence of attributes?
    • From: "Bullard, Claude L \(Len\)" <len.bullard@i...>
  • Re: Restrictions on existence of attributes?
    • From: Robin Berjon <robin.berjon@e...>
  • Re: Restrictions on existence of attributes?
    • From: Dave Pawson <davep@d...>

• Prev by Date: RE: Restrictions on existence of attributes?
• Next by Date: RE: Restrictions on existence of attributes?
• Previous by thread: Re: Restrictions on existence of attributes?
• Next by thread: RE: Restrictions on existence of attributes?
• Index(es):
  • Date
  • Thread