Re: [dita] Re: Discuss list processing expectations

I respect your right to instruct your
users on best practices - please respect my right to do the same. 

I'm genuinely confused by the sticking
point here, which seems to be that translation tools cannot recognize that
the occurrence of a block element in a para signals the end of the current
segment. My suggestion that we update the segmentation rules is not meant
to be inflammatory - I thought it was the easiest solution. If it's not,
then let's look at other solutions, although I continue to see it as a
tool limitation. 

Perhaps the mismatch is more fundamental
- if segmentation rules assume that the property inline or block is inherent
to the element, and XML assumes that the element role is determined by
its container's content model, then we are always going to have edge cases,
and perhaps this is one of them.

If necessary, we could perhaps introduce
a preprocessing step for translation tools - to make the implicit segments
explicit. I'm not sure of how the details would work, but it's an approach
that seemed to resolve an equivalent impasse with conref (in the case of
conreffing proper nouns). In other words, if we want flexibility for the
author, but predictability for the translators, let's see if we can't find
a solution that provides both.

Michael Priestley

Lead IBM DITA Architect 



http://dita.xml.org/blog/25

"Gershon L Joseph
\[Yahoo\]" <> 

04/15/2008 09:12 AM

To

Michael Priestley/Toronto/IBM@IBMCA

cc

DITA TC List <>,
JoAnn Hackos <>, Robert D Anderson <>,
Rob Frankland <>

Subject

Re: [dita] Re: Discuss list processing
expectations

It is very questionable whether HTML can
be considered a good markup language. I suggest we ignore HTML for the
purposes of this discussion.

 

I know we've had lots of discussions in the
DocBook TC about allowing things like lists inside paragraphs, and I think
many of the TC members were against this and many didn't care, which is
how it got in.

 

I encourage my users to mark up in a way
that is clean, yields predictable results, and works with today's tools.
I don't see any point in having them mark up in a way that costs them down
the line. Blame the tools if you like, but I prefer successful implementations
rather than having to blame the tools for failed projects.

 

Gershon

----- Original Message ----

From: Michael Priestley <>

To: Gershon L Joseph [Yahoo] <>

Cc: DITA TC List <>; JoAnn Hackos <>;
Robert D Anderson <>; Rob Frankland <>

Sent: Tuesday, April 15, 2008 3:51:32 PM

Subject: Re: [dita] Re: Discuss list processing expectations

DocBook allows list inside paragraph, just like DITA. 

HTML allows a mix of blocks and paragraphs in other contexts like inside
a list item. 

So this is not unique to DITA. 

If the current segmentation rules are at odds with the rules of many major
markup languages and a large body of users disagrees with them, whose interests
are being served by protecting the existing segmentation rules?

Which is easier to update, several major authoring markup languages plus
all authoring community best practices, or XLIFF?

Note: I would say *all* the major markup languages, but I only checked
two besides DITA. 

Michael Priestley