W3C home > Mailing lists > Public > www-tag@w3.org > March 2009

Re: @rel syntax in RDFa (relevant to ISSUE-60 discussion), was: Using XMLNS in link/@rel

From: Mark Nottingham <mnot@mnot.net>
Date: Sun, 1 Mar 2009 22:48:02 +1100
Cc: Ben Adida <ben@adida.net>, Manu Sporny <msporny@digitalbazaar.com>, "www-tag@w3.org WG" <www-tag@w3.org>, HTMLWG WG <public-html@w3.org>, RDFa mailing list <public-rdf-in-xhtml-tf@w3.org>, public-xhtml2@w3.org
Message-Id: <DD756F24-417D-4042-983F-7B6829BB7D35@mnot.net>
To: Sam Ruby <rubys@intertwingly.net>

Yes, absolutely; users will cherry-pick features from different  
versions of HTML to suit their desires, and plant them into whatever  
version of the format they like / their CMS supports / is the latest  
fashion. They'll expect browsers to seamlessly handle this.

This only underlines the importance making the syntax and  
disambiguation of @rel compatible among the different versions of  
HTML; that would serve these users well. Using a different version  
attribute is not an excuse to throw backwards compatibility out of the  
window, and it's not like RDFa is an old specification, so ignorance  
of the issues surrounding versioning and compatibility can't be claimed.

It certainly doesn't justify putting the subset of users who want to  
do the right thing and have valid, unambiguous markup into a place  
where they can't, because the features they need are spread out among  
those incompatible versions of HTML. Not every consumer of HTML is a  
browser.

> Users who are attracted to RDFa today are likely to have been  
> influenced either directly or indirectly by Zeldman and his  
> brethren.  They include an XHTML DOCTYPE and try to be careful about  
> quotes.  The few that actually read specs will see that XHTML 1.0  
> Transitional allows the use of the text/html MIME type.

Sorry, RDFa just became mainstream, when the Creative Commons started  
showing people how to use it by example. Speaking of attraction, I'm  
very much reminded of the US legal concept of an attractive nuisance  
here. Anyway...

 From where I'm sitting, RDFa should not have gone out the door as it  
is, and because it did we have some damage to contain. Likely it's not  
too bad, owing to the bad state of @rel in HTML anyway, but it has  
effectively created one more thing to sniff in HTML -- "what rel  
convention is in use here?" -- with all of the ambiguity and issues  
that entails.

So, I have a fair idea of what I'm going to write in the next Link  
draft now (see recent messages to Ben for a rough idea). What I really  
want to know -- and this is why the TAG is still on the CC list here  
-- is what's going to be done to prevent this from happening the next  
time.

Cheers,

P.S. Sam, I'm confused; you've brought up whitehouse.gov in this  
context a few times, but AFAICT they don't serve RDFa on their front  
page. Yes, they're serving XHTML with a text/html media type, but  
that's very wide and understood practice. Please explain?


On 01/03/2009, at 9:46 PM, Sam Ruby wrote:

> Mark Nottingham wrote:
>> On 01/03/2009, at 10:33 AM, Ben Adida wrote:
>>> Mark Nottingham wrote:
>>>> Are people using RDFa in HTML using the profile mechanism, or  
>>>> xmlns, or
>>>> both? Do they flag the use of RDFa in any way (like @version does  
>>>> for
>>>> XHTML+RDFa)?
>>>
>>> We recommend using the doctype with @version. We've determined that,
>>> while @profile is the right approach for interpreting new values of
>>> @rel, it doesn't cover new attributes, so it's not the right way  
>>> to flag
>>> RDFa.
>> Sorry, I should have been more clear; I'm talking about in HTML4  
>> (since CC doesn't require or recommend XHTML for CC+).
> [snip]
>> But an HTML4 parser has absolutely no business knowing about xmlns,  
>> period.
>
> I think we use these terms differently than users do.
>
> $ curl --silent --head http://www.whitehouse.gov/ | grep -i content- 
> type
> Content-Type: text/html; charset=utf-8
> $ curl --silent http://www.whitehouse.gov/ | head -3 | tail -1
> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd 
> ">
>
> From a user perspective features are things like canvas or RDFa.   
> Add a little markup, include a bit of javascript and you have  
> awesome graphics or delectable metadata.
>
> From a HTTP perspective, HTML is unversioned... there is no way to  
> indicate via a MIME type whether something is HTML4 vs HTML5, just a  
> way to indicate something is HTML vs XHTML; and that is done is a  
> way that is beyond what most users can control and is at variance  
> with what users understand XHTML to be and is entirely unsupported  
> by the dominant browser.  And triggers unforgiving error recovery  
> strategies.
>
> From a browser perspective, HTML is unversioned.  No browser that  
> supports canvas will refuse to do so simply because that tag is  
> included in a page which includes the HTML4 DOCTYPE.  Firefox (to  
> pick an concrete example) has no plans to have a separate HTML5  
> parser. Javascript libraries which will extract RDFa data will do so  
> even in quirks mode.  Even if the attributes happen to be named  
> starting with the characters 'x', 'm', and 'l'.
>
> Users who are attracted to RDFa today are likely to have been  
> influenced either directly or indirectly by Zeldman and his  
> brethren.  They include an XHTML DOCTYPE and try to be careful about  
> quotes.  The few that actually read specs will see that XHTML 1.0  
> Transitional allows the use of the text/html MIME type.
>
> I am biased.  I believe that we should cater to these users.  They  
> outnumber us.
>
> - Sam Ruby


--
Mark Nottingham     http://www.mnot.net/
Received on Sunday, 1 March 2009 11:48:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:48:13 GMT