Re: Re: eXtensible Content Syndication (the next step for OCS)

["Duynstee, Teun" <teun.duynstee@macaw.nl>]

>> 1) Lets keep really clear if the attributes on the format element should
or
>> shouldn't carry a prefix when the XCS namespace is not the default
>> namespace.
>
>I don't understand what you mean?  You are advocating *not* using attribute
>prefixes... right?

No, wrong. I have no problem with using namespaces and prefixes, but I think
that if you use them, you should use them correctly. In your example, you
use the line:

<format mimeType="" contentType="" location=""/>

Here the attributes have no prefix, but are in the namespace
http://schemas.openprivacy.org/08/10/2001/xcs, which you declared as the
default in your code. Now suppose that someone else uses another prefix for
your namespace, say xcs, it would look like:

<xcs:format xcs:mimeType="" xcs:contentType="" xcs:location=""/>

To a namespaces conforming parser, these fragments are equivalent. Now for
some reason, in many cases, developers do this:

<xcs:format mimeType="" contentType="" location=""/>

This is not equivalent.

What I tried to say was that you have (IMHO) two options:
1) state in your spec that no-one should use prefixes and that
http://schemas.openprivacy.org/08/10/2001/xcs must always be the default
namespace.

2) Show an example in the spec using prefixes, thus explaining that all of
the attributes need to be in the XCS namespace and prefixed accordingly.

I personally prefer the second option, but I could understand choosing the
first, because the format would be easier to parse (though it is beyond me
that people try to parse XML themselves).

Teun