[whatwg] The problems with namespaces in text/html

[James Graham]

Elliotte Harold wrote:
> Henri Sivonen wrote:
> 
>>> http://cafe.elharo.com/web/mokka/
>>
>> http://hsivonen.iki.fi/validator/?doc=http%3A%2F%2Fcafe.elharo.com%2Fweb%2Fmokka%2F&parser=xml&laxtype=yes 
>>
>>
>> I think that makes my point for me.
> 
> Not really. That's not the system I was talking about. The article at 
> the URL I referenced describes the old system as well the reasons I 
> switched to WordPress, but that had nothing to do with well-formedness.

But everything to do with the actual reasons that people will choose one 
system over another -- ease of authoring, attractiveness of final 
output, ease of maintainance; /not/ welformedness of the final HTML. 
Insisting on wellformed HTML can only make tools harder to use, because, 
it requires all templates and all /content/ must be well formed. 
Templates may be written by geeks but content is typically written by 
users don't really understand well-formedness. Therefore avoiding error 
messages that are read as "can't publish because the blah wibble foo on 
line 80 frobulates the kaniki bar on line 71" is impossible (several 
popular blogs that use XHTML have exactly this problem. Fortunately in 
those cases the audience tend to be web geeks or string-theory geeks so 
it's not a critical problem.). Needless to say systems which regularly 
spew out such apparent nonsense will not be popular with their users.

However, let's assume that we have a tool that can guarantee well-formed 
markup. As you note the pieces for building such tools do exist 
(although, as you fail to note, they are typically both slower and 
harder to use than the pieces for building sites based on simple string 
interpolation). One example is turbogears [1], which comes with the Kid 
  template system by default [2]. Under the covers the Kid template is 
turned into an XML ElementTree [3] and page rendering involves mutation 
of that tree. Given that, Turbogears will produce sites are good to be 
sent over the wire as application xhtml+xml to supporting browsers (of 
course it is almost certainly possible to break this property if you try 
hard enough). Despite this, _none_ of the sites listed on the turbogears 
front page are sent as anything other than text/html. Apparently authors 
desire the browser support and error-handling of HTML over the simpler 
parsing of XHTML even in situations where they have a real choice.

Can we please give up on the notion that XML is just as easy for authors 
as HTML? It's not, and wanting it to be doesn't make it so.

[1] http://www.turbogears.org
[2] http://kid-templating.org/
[3] http://effbot.org/zone/element-index.htm

-- 
"The universe doesn't care what you believe. The wonderful thing about 
science is that it doesn't ask for your faith, it just asks for your 
eyes" --- http://xkcd.com/c154.html