On Feb 17, 2013 3:17 PM, "Mukul Gandhi" <gandhi.mukul@gmail.com> wrote:
>
> On Sun, Feb 17, 2013 at 4:20 PM, Alex Russell <slightlyoff@google.com>
wrote:
>>
>> How would it have helped? Did you control the source of the document?
Could you in some way guarantee well-formedness from the publisher?
>
>
> For the application I was working with, I had no control on the source of
document since it was produced by a legacy process. I cannot guarantee
well-formedness from the publisher.
>>
>> Polyglot only helps those who know or control the parseability of the
documents they process.
>
> I agree.
>
> But I do feel, having a polyglot document structure improves
interoperability of information represented by them. As I thought in my
previous mail within this thread, I think the following processing would
help HTML/XHTML/XML users,
How?
> legacy HTML -> polyglot converter -> polyglot markup
>
> This functionality can be present outside the browser environment, or/and
within the browser.
You may be confused. A "polyglot converter" is simply an html-to-xml (or
back) conversion. Many tools enable this today. The term "polyglot", in
this discussion, refers to a property of documents, not their processing
pipelines. Exporting to a polyglot subset might be a feature of these
tools, but would not appear to have aided you in the problem you initially
described.