Re: Feedback on Internet Media Types and the Web

On 11.02.2011 07:33, "Martin J. Dürst" wrote:
> Hello Julian,
>
> On 2011/02/11 0:10, Julian Reschke wrote:
>> On 10.02.2011 15:57, Chris Lilley wrote:
>
>>> For +xml, that handling is
>>> - parsing
>>> - well formedness checking
>>> - validation
>>> - looking for namespaces you recognise
>>> - handing a dom to some other process
>
>>> For +zip:
>>> - display contents of zipfile (as a tree-like filebrowser)
>>> - look for particular things inside (manifest.xml)
>>> - offer to extract some, rather than all, of the contents
>>
>> Is any of these three above something you *want* to happen when opening
>> an ePub file?
>
> Yes and no.
>
> Yes in that the Epub application definitely will look for particular
> things inside, extract some, and so on.

Understood. But the *EPub* application will do that no matter what 
format the media type has.

The question was about the case where you do *not* have an EPub reader 
installed. Would a user expect to see the contents of the archive, or a 
notification that the system doesn't have an application for that type. 
I believe the answer is "the latter" for most non-programmers.

> No in that it won't do this things only. But then again, is anything in
> the +xml list something you (only!) *want* to happen if you are opening
> an image/svg+xml file?
>
> The +foo suffixes are helpful not for when you have an application for
> the exact type, but for when you don't, and have to fall back to more
> generic processing.

Exactly; the question is what generic processing can be done.

For +xml we know about some use cases in UAs, such as XmlHttpRequest. 
For +json, it would be the same.

Best regards, Julian