ReSpec: Intent to depracate XHTML 1.0 save support

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

ReSpec: Intent to depracate XHTML 1.0 save support

Marcos Caceres-5
Hi,
XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
we've been allowing folks to save as XHTML 1.0, but I'm wondering if
we can retire that capability?

Can anyone think of any valid reason to keep it around? It seems
harmful, as saving to a conforming XHTML 1.0 doc means we strip out
useful structuring tags and aria attributes.

Kind regards,
Marcos

Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Marcos Caceres-5


On May 16, 2016 at 12:13:39 PM, Marcos Caceres ([hidden email]) wrote:
> Hi,
> XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
> we've been allowing folks to save as XHTML 1.0, but I'm wondering if
> we can retire that capability?
>  
> Can anyone think of any valid reason to keep it around? It seems
> harmful, as saving to a conforming XHTML 1.0 doc means we strip out
> useful structuring tags and aria attributes.

Relevant bug: 
https://github.com/w3c/respec/issues/760

Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Shane McCarron-6
As I mentioned in the relevant PR, I think this is a bad idea.  There are a number of specs that were produced using XHTML 1, and if any of them need an update for errata of whatnot in the future, it will be impossible to do simply.  

On Sun, May 15, 2016 at 9:14 PM, Marcos Caceres <[hidden email]> wrote:


On May 16, 2016 at 12:13:39 PM, Marcos Caceres ([hidden email]) wrote:
> Hi,
> XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
> we've been allowing folks to save as XHTML 1.0, but I'm wondering if
> we can retire that capability?
>
> Can anyone think of any valid reason to keep it around? It seems
> harmful, as saving to a conforming XHTML 1.0 doc means we strip out
> useful structuring tags and aria attributes.

Relevant bug: 
https://github.com/w3c/respec/issues/760




--
Shane McCarron
Projects Manager, Spec-Ops
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Marcos Caceres-5


On 16 May 2016, at 9:32 PM, Shane McCarron <[hidden email]> wrote:

As I mentioned in the relevant PR, I think this is a bad idea.  There are a number of specs that were produced using XHTML 1, and if any of them need an update for errata of whatnot in the future, it will be impossible to do simply.  

Well, not impossible - but you'd have to update the markup. Or throw in a legacy version of ReSpec (so, not a big deal). 

Question: are any specs that are using xhtml 1.0 maintained today? Have they been updated in the last, say, 3-5 years? Are there examples of upcoming examples? 


On Sun, May 15, 2016 at 9:14 PM, Marcos Caceres <[hidden email]> wrote:


On May 16, 2016 at 12:13:39 PM, Marcos Caceres ([hidden email]) wrote:
> Hi,
> XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
> we've been allowing folks to save as XHTML 1.0, but I'm wondering if
> we can retire that capability?
>
> Can anyone think of any valid reason to keep it around? It seems
> harmful, as saving to a conforming XHTML 1.0 doc means we strip out
> useful structuring tags and aria attributes.

Relevant bug: 
https://github.com/w3c/respec/issues/760




--
Shane McCarron
Projects Manager, Spec-Ops
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Shane McCarron-6
Sorry - having a thread in github and here is confusing.  I am going to concentrate here.

We updated XHTML+RDFa as recently as last March.  It has an XHTML 1 version.  

I also note that the Pubrules continues to permit XHTML.  I am not saying that new specs should be authored in XHTML 1, but old specs get updated for errata or reference changes from time to time.   

On Mon, May 16, 2016 at 6:41 AM, <[hidden email]> wrote:


On 16 May 2016, at 9:32 PM, Shane McCarron <[hidden email]> wrote:

As I mentioned in the relevant PR, I think this is a bad idea.  There are a number of specs that were produced using XHTML 1, and if any of them need an update for errata of whatnot in the future, it will be impossible to do simply.  

Well, not impossible - but you'd have to update the markup. Or throw in a legacy version of ReSpec (so, not a big deal). 

Question: are any specs that are using xhtml 1.0 maintained today? Have they been updated in the last, say, 3-5 years? Are there examples of upcoming examples? 


On Sun, May 15, 2016 at 9:14 PM, Marcos Caceres <[hidden email]> wrote:


On May 16, 2016 at 12:13:39 PM, Marcos Caceres ([hidden email]) wrote:
> Hi,
> XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
> we've been allowing folks to save as XHTML 1.0, but I'm wondering if
> we can retire that capability?
>
> Can anyone think of any valid reason to keep it around? It seems
> harmful, as saving to a conforming XHTML 1.0 doc means we strip out
> useful structuring tags and aria attributes.

Relevant bug: 
https://github.com/w3c/respec/issues/760




--
Shane McCarron
Projects Manager, Spec-Ops



--
Shane McCarron
Projects Manager, Spec-Ops
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Marcos Caceres-5


On 16 May 2016, at 10:55 PM, Shane McCarron <[hidden email]> wrote:

Sorry - having a thread in github and here is confusing.  I am going to concentrate here.

Np. Here is better.


We updated XHTML+RDFa as recently as last March.  It has an XHTML 1 version.  

Ok, that's good info. Anyone else know of other specs? 


I also note that the Pubrules continues to permit XHTML.  I am not saying that new specs should be authored in XHTML 1, but old specs get updated for errata or reference changes from time to time.   

Would anything preventing future versions being published with xhtml5? 



On Mon, May 16, 2016 at 6:41 AM, <[hidden email]> wrote:


On 16 May 2016, at 9:32 PM, Shane McCarron <[hidden email]> wrote:

As I mentioned in the relevant PR, I think this is a bad idea.  There are a number of specs that were produced using XHTML 1, and if any of them need an update for errata of whatnot in the future, it will be impossible to do simply.  

Well, not impossible - but you'd have to update the markup. Or throw in a legacy version of ReSpec (so, not a big deal). 

Question: are any specs that are using xhtml 1.0 maintained today? Have they been updated in the last, say, 3-5 years? Are there examples of upcoming examples? 


On Sun, May 15, 2016 at 9:14 PM, Marcos Caceres <[hidden email]> wrote:


On May 16, 2016 at 12:13:39 PM, Marcos Caceres ([hidden email]) wrote:
> Hi,
> XHTML 1.0 has been long deprecated in favor of XHTML5. In ReSpec,
> we've been allowing folks to save as XHTML 1.0, but I'm wondering if
> we can retire that capability?
>
> Can anyone think of any valid reason to keep it around? It seems
> harmful, as saving to a conforming XHTML 1.0 doc means we strip out
> useful structuring tags and aria attributes.

Relevant bug: 
https://github.com/w3c/respec/issues/760




--
Shane McCarron
Projects Manager, Spec-Ops



--
Shane McCarron
Projects Manager, Spec-Ops
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Michiel Bijl
In reply to this post by Shane McCarron-6
On 16 May 2016, at 14:55, Shane McCarron <[hidden email]> wrote:

It has an XHTML 1 version. 

But it has another version? Would it be feasible to forward the xhtml 1 version to some other version?

—Michiel
Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Marcos Caceres-5
On June 1, 2016 at 2:17:56 AM, Michiel Bijl ([hidden email]) wrote:
> > On 16 May 2016, at 14:55, Shane McCarron wrote:
> >
> > It has an XHTML 1 version.
>
> But it has another version? Would it be feasible to forward the xhtml 1 version to some
> other version?

My (terribly limited) understanding is that RDFa is tied to xhtml 1
instead of xhtml 5, because there is no modern schema that one can
validate xhtml+rdfa against. Otherwise, it would be possible to
publish xhtml5+rdfa documents.

Shane, is that correct?

Reply | Threaded
Open this post in threaded view
|

Re: ReSpec: Intent to depracate XHTML 1.0 save support

Shane McCarron-6
Not quite.  XHTML+RDFa DEFINES a markup language - xhtml-rfda.  This is based upon XHTML 1.0.  So we need to publish a version of the specification that uses that markup language.  

On Tue, May 31, 2016 at 11:54 PM, Marcos Caceres <[hidden email]> wrote:
On June 1, 2016 at 2:17:56 AM, Michiel Bijl ([hidden email]) wrote:
> > On 16 May 2016, at 14:55, Shane McCarron wrote:
> >
> > It has an XHTML 1 version.
>
> But it has another version? Would it be feasible to forward the xhtml 1 version to some
> other version?

My (terribly limited) understanding is that RDFa is tied to xhtml 1
instead of xhtml 5, because there is no modern schema that one can
validate xhtml+rdfa against. Otherwise, it would be possible to
publish xhtml5+rdfa documents.

Shane, is that correct?



--
Shane McCarron
Projects Manager, Spec-Ops