Re: NU’s polyglot possibilities (Was: The non-polyglot elephant in the room)

Smylers, Sat, 26 Jan 2013 11:33:47 +0000:
> Alex Russell writes:
> 
>> I find myself asking (without an obvious answer): who benefits from
>> the creation of polyglot documents?
> 
> I can see an advantage of the creating a snippet as polyglot HTML/XHTML.
> For example if you want to distribute a badge or banner or status widget
> or search form or something that others can paste into their sites, it's
> simpler if you can just offer a single snippet which will work fine when
> pasted into either an HTML or an XHTML document.

I agree. And, in fact, I have reported bugs about tools based on that 
motivation. Remember: Both <H1>foo</H1> and <h1>foo</h1> are conforming 
per HTML5. So, based on what motivation should a developer change that 
to <h1>foo</h1>? I see no other motivation than compatibility with XML 
for such a change. But which tool vendor would not like to expand their 
tools' utility?

> However:
> 
> * The number of sites served as XHTML is so small that this may not be
>   worth bothering with.

Like Sam, I wonder if it is worth wondering about. Just do it - and 
move on to more important things.

> * This scenario doesn't involve an entire document being polyglot. So if
>   we wanted to support it we'd actually need a spec and validator for
>   polyglot snippets, rather than entire documents.

The "old" validator, at http://validator.w3.org, supports snipped 
validation. But else, it is not very difficult - as long as the text 
field of the validators provides polyglot dummy code, the user can 
paste the snippet where it fits.

> * You can't quite get this right in all detail anyway. For example,
>   suppose your snippet includes a <table>, and you want it to be styled
>   by an enclosing site's style-sheet to match other <table>-s on the
>   site. Do you include <tbody> tags or not? If the enclosing site is
>   XHTML then its styles may depend on either the presence of or the
>   absence of <tbody> elements. In order to match your snippet needs to
>   do the same -- but obviously you can't produce a single snippet which
>   both includes and omits <tbody> tags, so you can't cover both
>   possibilities anyway.

Polyglot XHTML includes <tbody> - so says the spec. Obviously, polyglot 
markup cannot solve the problem of pages that relies on the code *not* 
being polyglot.
-- 
leif halvard silli

Received on Sunday, 27 January 2013 22:21:17 UTC