OASIS Darwin Information Typing Architecture (DITA) TC

I think this part of page 78 could be improved too:

"Using markup to identify language is strongly recommended to make the
document as portable as possible. The
marked-up document can be read and understood by humans. Because the
boundaries of each language are
clear, it is much easier for the author to update the document."

If the second and third sentences are currently saying anything, they
aren't providing any rationale in support of these assertions. I think
we should remove these sentences unless someone can suggest how to
clarify them.

Su-Laine



                                                

					
                                                    

        
                                                
				
                                                

                                                    
                                                
                                                
                                            

I don't see any point (or value) in the entire paragraph. I think we can delete the whole paragraph. (My personal opinion is that using markup to identify language is not strongly recommended, *it's a requirement!*)

Cheers,
Gershon




                                                

					
                                                    

        
                                                
				
                                                

                                                    
                                                        
                                                        Original Message
                                                
                                                Original Message-----
From: Su-Laine Yeo [mailto:su-laine.yeo@justsystems.com] 
Sent: Friday, August 27, 2010 3:40 AM
To: dita@lists.oasis-open.org
Subject: RE: [dita] Clarification needed on whether xml:lang is required

I think this part of page 78 could be improved too:

"Using markup to identify language is strongly recommended to make the
document as portable as possible. The
marked-up document can be read and understood by humans. Because the
boundaries of each language are
clear, it is much easier for the author to update the document."

If the second and third sentences are currently saying anything, they
aren't providing any rationale in support of these assertions. I think
we should remove these sentences unless someone can suggest how to
clarify them.

Su-Laine


                                            

Concur on both points. This might have been written before the ATC got
started.  

The attribute description at 3.4.1.6 "localization-atts attribute group"
punts to the XML spec. The XML spec doesn't talk about this as either
advice or requirement. It does give some interesting examples that DITA
specializers could consider, and that the ATC might elaborate.

Elsewhere in 2.1.3.9.1 "The @xml:lang attribute" we say that the top
element "should" set this attribute, and that applications "should"
ensure that this is done. Should we beef this up, Gershon? I think we
punted the "must" talk to the ATC.

	/B

> 

                                                

					
                                                    

        
                                                
				
                                                

                                                    
                                                        
                                                        Original Message
                                                
                                                Original Message-----
> From: Gershon Joseph (gerjosep) 
> Sent: Sunday, August 29, 2010 1:54 PM
> To: Su-Laine Yeo; dita@lists.oasis-open.org
> Subject: RE: [dita] Clarification needed on whether xml:lang 
> is required
> 
> I don't see any point (or value) in the entire paragraph. I 
> think we can delete the whole paragraph. (My personal opinion 
> is that using markup to identify language is not strongly 
> recommended, *it's a requirement!*)
> 
> Cheers,
> Gershon
> 
> 
> 
Original Message-----
> From: Su-Laine Yeo [mailto:su-laine.yeo@justsystems.com]
> Sent: Friday, August 27, 2010 3:40 AM
> To: dita@lists.oasis-open.org
> Subject: RE: [dita] Clarification needed on whether xml:lang 
> is required
> 
> I think this part of page 78 could be improved too:
> 
> "Using markup to identify language is strongly recommended to 
> make the document as portable as possible. The marked-up 
> document can be read and understood by humans. Because the 
> boundaries of each language are clear, it is much easier for 
> the author to update the document."
> 
> If the second and third sentences are currently saying 
> anything, they aren't providing any rationale in support of 
> these assertions. I think we should remove these sentences 
> unless someone can suggest how to clarify them.
> 
> Su-Laine
> 
> 
                                            

Hi Bruce,

I think we can leave the DITA spec as-is. The DITA spec cannot make this a "must" requirement. On the adoption TC we can strongly recommend it.

Cheers,
Gershon




                                                

					
                                                    

        
                                                
				
                                                

                                                    
                                                        
                                                        Original Message
                                                
                                                Original Message-----
From: Bruce Nevin (bnevin) 
Sent: Wednesday, September 01, 2010 7:15 PM
To: dita@lists.oasis-open.org
Subject: RE: [dita] Clarification needed on whether xml:lang is required

Concur on both points. This might have been written before the ATC got
started.  

The attribute description at 3.4.1.6 "localization-atts attribute group"
punts to the XML spec. The XML spec doesn't talk about this as either
advice or requirement. It does give some interesting examples that DITA
specializers could consider, and that the ATC might elaborate.

Elsewhere in 2.1.3.9.1 "The @xml:lang attribute" we say that the top
element "should" set this attribute, and that applications "should"
ensure that this is done. Should we beef this up, Gershon? I think we
punted the "must" talk to the ATC.

	/B

> 
Original Message-----
> From: Gershon Joseph (gerjosep) 
> Sent: Sunday, August 29, 2010 1:54 PM
> To: Su-Laine Yeo; dita@lists.oasis-open.org
> Subject: RE: [dita] Clarification needed on whether xml:lang 
> is required
> 
> I don't see any point (or value) in the entire paragraph. I 
> think we can delete the whole paragraph. (My personal opinion 
> is that using markup to identify language is not strongly 
> recommended, *it's a requirement!*)
> 
> Cheers,
> Gershon
> 
> 
> 
Original Message-----
> From: Su-Laine Yeo [mailto:su-laine.yeo@justsystems.com]
> Sent: Friday, August 27, 2010 3:40 AM
> To: dita@lists.oasis-open.org
> Subject: RE: [dita] Clarification needed on whether xml:lang 
> is required
> 
> I think this part of page 78 could be improved too:
> 
> "Using markup to identify language is strongly recommended to 
> make the document as portable as possible. The marked-up 
> document can be read and understood by humans. Because the 
> boundaries of each language are clear, it is much easier for 
> the author to update the document."
> 
> If the second and third sentences are currently saying 
> anything, they aren't providing any rationale in support of 
> these assertions. I think we should remove these sentences 
> unless someone can suggest how to clarify them.
> 
> Su-Laine
> 
>