[microformats-discuss] XHTML Property Lists (Mac OS X)

Ryan King ryan at technorati.com
Thu Sep 29 14:19:41 PDT 2005


On Sep 29, 2005, at 12:47 PM, Dr. Ernie Prabhakar wrote:

> Hi Kevin,
>
> On Sep 27, 2005, at 12:41 AM, Kevin Marks wrote:
>
>> note that class="string" is implied.
>> I'd keep the xml classnames, as the plist scope would clarify  
>> them. So you just get integer, real, and boolean.
>> NSDate I'd map to the <abbr> style date discussed passim.
>> array and dictionary naturally become <ol> and <dl> as per XOXO.
>> NSData is a bit of a bother - using an <a href="data:"> url (RFC  
>> 2397) might make sense.
>
> Wow, I'd never heard of RFC 2397.  Very cool.
>
> I've updated my example based on your suggestions, and posted them  
> here:
>
> http://homepage.mac.com/drernie/plist.html
>
> The only question I had was whether to use the XML-schema term  
> 'float' rather than the plist term 'real.'

The general problem here is with negotiating between two cases of  
prior art. I'm not sure that there's a clear solution for the general  
case.

In my opinion, in this specific case, I'd stick as close to plist as  
possible, so that you avoid impedance mismatches between the two.

> Which gets to Ryan's question:
>
>> I have a question: what would you use this for? What's the point?  
>> What's the problem we're trying to solve.
>
> Well, a couple things.  One, I've often wanted to view my plists as  
> HTML, so I may as well do it in a reversible fashion.  More  
> importantly, though, I've considered using plists as part of a file  
> format for cross-platform applications.  Rather than having to  
> evangelize adopting Apple's DTD, it seems simpler to just publish  
> as XOXO.   I might even file a bug suggesting the built-in 'plutil'  
> export as HTML to aid this process.

coolness. anything but the binary plist format.

> Given that usage scenario, do you think 'float' is more or less  
> useful than 'real'?

see above.
--
Ryan King
ryan at technorati.com





More information about the microformats-discuss mailing list