On Wed, Apr 7, 2010 at 11:13 AM, Andrew Welch <> wrote:
> On 7 April 2010 13:49, Simon St.Laurent <> wrote:
>> On 4/5/10 11:59 PM, Liam R E Quin wrote:
>>>
>>> If you don't have an ambiguity problem, the complexity of namespaces
>>> is simply unlikely to give you any benefit.
>>
>> I'll agree with Liam, to the extent that I'd like to see that emblazoned on
>> a brass plaque somewhere in the XML Hall of Fame.
>>
>> The opening "if" is important, of course.
>
> True. However, can you be sure that somewhere down the line some
> other consumer of your markup won't come along? The pain from
> retrofitting namespaces is much, much worse than coding them in from
> the start.
Yes; I always expect my XML to be combined in unexpected ways with
other dialects, and I am rarely disappointed. This is actually one of
the thing I deeply appreciate about XML (and thus, the existence of
namespacing technology). As a result, every dialect I develop ends up
in its own namespace.
> My suggestion would be to use a namespace, as it brands your markup
> and prevents problems down the line, and definitely use a prefix.
>
> Unprefixed namespaced xml can be a pain, but prefixed xml is no problem at all.
And here is where I get confused. To a namespace-aware XML processor,
the existence of prefixes (or lack thereof) in the source is
irrelevant. I don't feel any pressure to use namespace prefixes until
multiple dialects get combined, and then I tend to award the majority
dialect with prefix-freedom. (A notable exception is XSLT, where I
like to emphasize the majority *output* dialect by placing it in the
default namespace.)
Take care,
John L. Clark
--
PLEASE NOTE that this message is not digitally signed. As a result,
you have no strong evidence that this message was actually sent by me.
Upon request I can provide a digitally signed receipt for this
message or other evidence validating its contents if you need such
evidence.
