Re: [Ecrit] FW: [NENA-ltd] LoST

[Andrew Newton <andy at hxr.us>]

The tool issue is a bit of FUD, IMO.  It is still possible to use SAX  
or DOM or pull APIs just like with XML Schemas and DTDs and all the  
other types XML schema languages.  There is nothing about Relax NG  
that requires an implementation to use a special XML parser.  The  
tools people are talking about are normally code generators, which  
are usually the hallmark of sloppy programming practices (my opinion,  
but I've seen it a lot).  I certainly would hope that anyone serious  
about fielding LoST knows how XML works and does not have to rely on  
such crutches.  I've had no trouble finding tools for validation.   
And both my desktop XML editors support Relax NG, one supporting it  
much better than XML Schema.

BTW, regarding your statement about "We're not XML developers."   
Well, you better understand it.  Just like you better have an  
understanding of how TCP/IP works too.  Or would you prefer that  
people writing software be able to stand behind statements like  
"We're not Internet applications developers."

-andy

On Dec 2, 2006, at 4:52 PM, Brian Rosen wrote:

> And what is their conclusion?
>
> Andy's answer is a bit troubling.  We're not XML developers, XML is  
> a "tool"
> (distinguishing from the tools we use with XML) to us.  We get  
> disadvantaged
> when we are early adopters of new XML thingies.  The XML  
> development tools
> probably matter more than the UPA restriction, and readability is  
> in the eye
> of the beholder: to anyone who has been looking at XML schemas for  
> the last
> several years, Relax NG is definitely not more readable.  Now, I  
> wouldn't
> really use the latter as an excuse to NOT switch to a new form,  
> because that
> would keep us on obsolete tools, but it is also not a compelling  
> reason to
> switch.
>
> I had thought the development tools were okay with Relax NG.  If  
> that is not
> the case, then I question whether we should be using it.


_______________________________________________
Ecrit mailing list
Ecrit at ietf.org
https://www1.ietf.org/mailman/listinfo/ecrit