RE: XInclude

[Carlos Tejo Alonso <ca...@fundacionctic.org>]

> >>> 18. - [code] Migrate to a decent schema language, primarily 
> >> so that we can
> >>> use namespaces in XML docs, allowing things like XInclude, 
> >> in-line metadata,
> >>> in-line SVG, Jelly snippets, or anything else users can 
> >> make a Transformer
> >>> for. → open
> >>>
> >>> So it seems that is not done. I don't know the best way 
> >> forward on this, we
> >>> can add it to our DTD, create a new one and add it to our 
> >> schema, point to
> >>> something better existing or .. ??
> >>>
> >>> The patch itself seems flawless in its application, 
> matching the W3C
> >>> examples [1] pretty closely, so I don’t have a problem with 
> >> the way they
> >>> have been applied as such, just need to work out the best 
> >> approach to
> >>> approve the method and get it validated against our tests.
> > 
> > Are you planning to change the xdoc intermediate language? 
> Maybe change to xhtml 1.1? I will be waiting for news in 
> order to do my bit.
> 
> For a very long time we have intending to move to a subset of 
> XHTML 2. 
> But planning and doing are two different things. We need 
> someone with a 
> suitably strong need to solve the problems posed by XDoc to 
> actually do it.

Maybe, we should start to think to do it. Could somebody show the steps to move from xdoc to xhtml? I will try to do my bit.
 
> Why XHTML2? For a full answer see the mail archives but in 
> short it is 
> because it is modular and therefore allows us to strip out 
> all the bits 
> that we don't want. That is all the bits that make it useless as an 
> intermediate language.

As xhtml2 is still a working draft ... Why not move to xhtml1.1? Is it not suitable for the required needs?

Cheers,

Carlos Tejo Alonso
R&D+I Deparment - CTIC Foundation [Asturias, Spain]
www.fundacionctic.org 

Re: XHTML2 as forrest internal

[Ross Gardler <rg...@apache.org>]

Tim Williams wrote:
> On Wed, Aug 6, 2008 at 10:07 PM, David Crossley <cr...@apache.org> wrote:
>> Ross Gardler wrote:
>>> Carlos Tejo Alonso wrote:
>>>
>>>>> For a very long time we have intending to move to a subset of XHTML
>>>>> 2. But planning and doing are two different things. We need someone
>>>>> with a suitably strong need to solve the problems posed by XDoc to
>>>>> actually do it.
>>>> Maybe, we should start to think to do it. Could somebody show the
>>>> steps to move from xdoc to xhtml? I will try to do my bit.
>>> If it were that simple we would have done it by now ;-)
>>>
>>> There have been a number of false starts on this. My own effort was
>>> aborted because a significant number of people in the community
>>> disagreed with my approach.
>> Wow. That is not my recollection. I thought that we had a
>> little try, confirmed that it would take a co-ordinated
>> approach, and left it for another day. I am talking about
>> the re-working of all sitemaps and stylesheets (including
>> plugins) to deal with XHTML2 internally, leaving everything
>> else (e.g. cocoon) as it already is.
>>
>> Are you talking about something else, or do you recall
>> it differently?
> 
> <blush> I reckon he's mostly referring to me:(
> 
>>> Gav made a start on an XHTML2 plugin (in whiteboard) but that also stalled.
>> IIRC, the initial "internal.xhtml2" and "input.xdoc" were
>> the result of Ross, Tim, and others around some of our
>> Friday IRC sessions. I thought that was a fantastic step forward.
>>
>> I thought that it was just waiting for some group of people
>> to get itchy again.
> 
> The details are forgotten with time, but I believe the sticking point
> was an issue of "scope" of what it means to transition.  There were
> two pieces of work to be done:
> o) Data format - schema, updating input plugins, create xdoc->xhtml2
> plugin, etc.
> o) Pipeline refining - update pipelines to take advantage of xhtml2 + views
> 
> IIRC, I wanted them to be addressed separately to allow baby steps.
> Ross preferred more holistic approach.  Part of my motivation at the
> time was that views were still very much a moving target.  In the end,
> no substantive steps were taken and I probably gave way too much merit
> to how much any given approach really matters in the grand scheme.
> Perhaps the only good news is that in the meantime the dispatcher has
> matured, making the timing ripe for someone to re-engage.

As you say time blurs the recollection - but all the discussions are in 
the mailing lists and the IRC session logs, in particular:

http://svn.apache.org/repos/asf/forrest/events/forrest-friday/20050906-log.txt

and

http://svn.apache.org/repos/asf/forrest/events/forrest-friday/20050918-log.txt

I think Tim's recollection is close enough. At the time some people felt 
that views were not mature enough for this early XHTML2 experimental work.

Today views (now Dispatcher) is reasonably stable, but it still has not 
graduated from the whiteboard.

At the time of doing our experiments with the XHTML2 work I used views 
*because* it was a moving target. At that time it was an opportunity to 
rewrite significant portions of Forrest a piece at a time.

Once that was done we could decide to retrofit that to skins or to move 
the work into the main dispatcher (nee views) pkugin. However, we never 
got that far because we couldn't agree on the right approach and I 
didn't want to waste my time on an approach that was not going to be 
accepted by the community.

In that respect I think Davids recollections are also true. There is a 
great deal of work to do to make XHTML2 happen and without the full 
support of the community it isn't going to happen.

We are agreed that we need to do it. We are not agreed with how to do 
it. Personally I think anyone who has the time to devote to this should 
be free to just get on with it and we, as a community, should provide 
whatever support we can.

It is the end result that matters not the road we take.

So, if someone is working in dispatcher on their home sites and their 
itch says they need XHTML2 support then go for it with the dispatcher. 
If you are using skins, then go for it using skins.

Ross



> 
> If my memory has served me wrong, I apologize in advance, it's
> definitely not my intent to be a revisionist historian:)
> 
> --tim

Re: XHTML2 as forrest internal

[Ross Gardler <rg...@apache.org>]

Tim Williams wrote:
> On Wed, Aug 6, 2008 at 10:07 PM, David Crossley <cr...@apache.org> wrote:
>> Ross Gardler wrote:
>>> Carlos Tejo Alonso wrote:
>>>
>>>>> For a very long time we have intending to move to a subset of XHTML
>>>>> 2. But planning and doing are two different things. We need someone
>>>>> with a suitably strong need to solve the problems posed by XDoc to
>>>>> actually do it.
>>>> Maybe, we should start to think to do it. Could somebody show the
>>>> steps to move from xdoc to xhtml? I will try to do my bit.
>>> If it were that simple we would have done it by now ;-)
>>>
>>> There have been a number of false starts on this. My own effort was
>>> aborted because a significant number of people in the community
>>> disagreed with my approach.
>> Wow. That is not my recollection. I thought that we had a
>> little try, confirmed that it would take a co-ordinated
>> approach, and left it for another day. I am talking about
>> the re-working of all sitemaps and stylesheets (including
>> plugins) to deal with XHTML2 internally, leaving everything
>> else (e.g. cocoon) as it already is.
>>
>> Are you talking about something else, or do you recall
>> it differently?
> 
> <blush> I reckon he's mostly referring to me:(

Not at all, there was considerable support for your position.

My comment is not a complaint, merely an observation that the approach 
taken in my work was not universally accepted. I was trying to prevent 
us going down the same road again without discussion first.

I'll get back to this thread when I have time to contribute to the 
discussion (within 24 hours I hope)

Ross

Ross

Re: XHTML2 as forrest internal

[Tim Williams <wi...@gmail.com>]

On Wed, Aug 6, 2008 at 10:07 PM, David Crossley <cr...@apache.org> wrote:
> Ross Gardler wrote:
>> Carlos Tejo Alonso wrote:
>>
>> >>For a very long time we have intending to move to a subset of XHTML
>> >>2. But planning and doing are two different things. We need someone
>> >>with a suitably strong need to solve the problems posed by XDoc to
>> >> actually do it.
>> >
>> >Maybe, we should start to think to do it. Could somebody show the
>> >steps to move from xdoc to xhtml? I will try to do my bit.
>>
>> If it were that simple we would have done it by now ;-)
>>
>> There have been a number of false starts on this. My own effort was
>> aborted because a significant number of people in the community
>> disagreed with my approach.
>
> Wow. That is not my recollection. I thought that we had a
> little try, confirmed that it would take a co-ordinated
> approach, and left it for another day. I am talking about
> the re-working of all sitemaps and stylesheets (including
> plugins) to deal with XHTML2 internally, leaving everything
> else (e.g. cocoon) as it already is.
>
> Are you talking about something else, or do you recall
> it differently?

<blush> I reckon he's mostly referring to me:(

>> Gav made a start on an XHTML2 plugin (in whiteboard) but that also stalled.
>
> IIRC, the initial "internal.xhtml2" and "input.xdoc" were
> the result of Ross, Tim, and others around some of our
> Friday IRC sessions. I thought that was a fantastic step forward.
>
> I thought that it was just waiting for some group of people
> to get itchy again.

The details are forgotten with time, but I believe the sticking point
was an issue of "scope" of what it means to transition.  There were
two pieces of work to be done:
o) Data format - schema, updating input plugins, create xdoc->xhtml2
plugin, etc.
o) Pipeline refining - update pipelines to take advantage of xhtml2 + views

IIRC, I wanted them to be addressed separately to allow baby steps.
Ross preferred more holistic approach.  Part of my motivation at the
time was that views were still very much a moving target.  In the end,
no substantive steps were taken and I probably gave way too much merit
to how much any given approach really matters in the grand scheme.
Perhaps the only good news is that in the meantime the dispatcher has
matured, making the timing ripe for someone to re-engage.

If my memory has served me wrong, I apologize in advance, it's
definitely not my intent to be a revisionist historian:)

--tim

XHTML2 as forrest internal

[David Crossley <cr...@apache.org>]

Ross Gardler wrote:
> Carlos Tejo Alonso wrote:
> 
> >>For a very long time we have intending to move to a subset of XHTML
> >>2. But planning and doing are two different things. We need someone
> >>with a suitably strong need to solve the problems posed by XDoc to
> >> actually do it.
> >
> >Maybe, we should start to think to do it. Could somebody show the
> >steps to move from xdoc to xhtml? I will try to do my bit.
> 
> If it were that simple we would have done it by now ;-)
> 
> There have been a number of false starts on this. My own effort was 
> aborted because a significant number of people in the community 
> disagreed with my approach.

Wow. That is not my recollection. I thought that we had a
little try, confirmed that it would take a co-ordinated
approach, and left it for another day. I am talking about
the re-working of all sitemaps and stylesheets (including
plugins) to deal with XHTML2 internally, leaving everything
else (e.g. cocoon) as it already is.

Are you talking about something else, or do you recall
it differently?

> Gav made a start on an XHTML2 plugin (in whiteboard) but that also stalled.

IIRC, the initial "internal.xhtml2" and "input.xdoc" were
the result of Ross, Tim, and others around some of our
Friday IRC sessions. I thought that was a fantastic step forward.

I thought that it was just waiting for some group of people
to get itchy again.

> Status is discussed in 
> http://markmail.org/message/f4shq54nmqebki5o?q=list:forrest-dev+xhtml2#query:list%3Aforrest-dev 
> xhtml2+page:1+mid:ttv65i23quvrkzks+state:results
> 
> Like any open source project our archives and issue tracker are our 
> memory and information pook. Look in JIRA for mentions of XHTML2, in 
> particular the following issue identifies the outline steps involved in 
> making this move and tracks activity to date (including links into the 
> mail arcives):

Yeah, and i recommend that all Forrest developers use
the http://forrest.apache.org/tools/forrestbar.html for Firefox.
It is fantastic for a quick directed search of our mail archives,
and issue tracker, and svn commits, etc.

-David

> http://issues.apache.org/jira/browse/FOR-184
> 
> You should also conduct a search of the archives to see if there is 
> anything missing from those issues (and please add anything you do find).
> 
> By reading those (and possibly other) threads you will be able to throw 
> your own thoughts into the mix. Of course, we'd be happy to answer 
> questions that arise as a result of reading those threads.
> 
> >>Why XHTML2? For a full answer see the mail archives but in short it
> >>is because it is modular and therefore allows us to strip out all
> >>the bits that we don't want. That is all the bits that make it
> >>useless as an intermediate language.
> >
> >As xhtml2 is still a working draft ... Why not move to xhtml1.1? Is
> >it not suitable for the required needs?
> 
> No it is not suitable. XHTML1.1 is monolithic. Forrest has to use a 
> clean markup so that no style inforamtion is included in the source (or 
> intermediate format). We also need clearly structured documents that 
> prevent the user from doing something like:
> 
> <h1>Heading type 1</h1>
> 
> <h3>Heading type 3</h3>
> 
> <h2> Heading type 2</h2>
> 
> Failure to do this will result in unpredictable behaviour at the output 
> stage.
> 
> XHTML2 is modular and allows us to select the markup we want to be legal 
> and also provides proper structuring of the source.
> 
> For more see the archives, e.g. 
> http://marc.info/?l=forrest-dev&m=102884176431540&w=2
> 
> Ross

Re: XInclude

[Ross Gardler <rg...@apache.org>]

Carlos Tejo Alonso wrote:

>> For a very long time we have intending to move to a subset of XHTML
>> 2. But planning and doing are two different things. We need someone
>> with a suitably strong need to solve the problems posed by XDoc to
>>  actually do it.
> 
> Maybe, we should start to think to do it. Could somebody show the
> steps to move from xdoc to xhtml? I will try to do my bit.

If it were that simple we would have done it by now ;-)

There have been a number of false starts on this. My own effort was 
aborted because a significant number of people in the community 
disagreed with my approach.

Gav made a start on an XHTML2 plugin (in whiteboard) but that also stalled.

Status is discussed in 
http://markmail.org/message/f4shq54nmqebki5o?q=list:forrest-dev+xhtml2#query:list%3Aforrest-dev 
xhtml2+page:1+mid:ttv65i23quvrkzks+state:results

Like any open source project our archives and issue tracker are our 
memory and information pook. Look in JIRA for mentions of XHTML2, in 
particular the following issue identifies the outline steps involved in 
making this move and tracks activity to date (including links into the 
mail arcives):

http://issues.apache.org/jira/browse/FOR-184

You should also conduct a search of the archives to see if there is 
anything missing from those issues (and please add anything you do find).

By reading those (and possibly other) threads you will be able to throw 
your own thoughts into the mix. Of course, we'd be happy to answer 
questions that arise as a result of reading those threads.

>> Why XHTML2? For a full answer see the mail archives but in short it
>> is because it is modular and therefore allows us to strip out all
>> the bits that we don't want. That is all the bits that make it
>> useless as an intermediate language.
> 
> As xhtml2 is still a working draft ... Why not move to xhtml1.1? Is
> it not suitable for the required needs?

No it is not suitable. XHTML1.1 is monolithic. Forrest has to use a 
clean markup so that no style inforamtion is included in the source (or 
intermediate format). We also need clearly structured documents that 
prevent the user from doing something like:

<h1>Heading type 1</h1>

<h3>Heading type 3</h3>

<h2> Heading type 2</h2>

Failure to do this will result in unpredictable behaviour at the output 
stage.

XHTML2 is modular and allows us to select the markup we want to be legal 
and also provides proper structuring of the source.

For more see the archives, e.g. 
http://marc.info/?l=forrest-dev&m=102884176431540&w=2

Ross