RE: [xml-dev] "Introducing MicroXML, Part 1: Explore the basic principle

[Pete Cordell <petexmldev@codalogic.com>]

On 3 July 2012 16:43, David Lee <dlee@calldei.com> wrote:
> I still find this a tantalizing discussion.
> My question is this:
> Suppose we identify a really useful set of attributes for "doing X" 
> and want to include it in the xml namespace.  ( note: for MicroXML 
> thats the ONLY namespace ???) ...
> And suppose we really believe this is useful stuff.
>
> Question: How sure are we that we have it right ?  Once things get 
> baked in they are pretty baked.   I believe a lot of XMLish things 
> failed because the were not quite baked before being pulled out of 
> the oven.
> Even today I am not quite convinced of the validity of a URL vs a URI.
> That is, what *is* the guarantee that a URL actually can be indirected 
> to a resource, and that resource  is what we think it is ?
> Of course that hasn't stopped HTML href's ... they sometimes work ok :)
>
> So if we propose adding something useful to a new standard ... how 
> sure are we its the right thing ?  I can accept that its "optional" 
> but if it's not well defined and not quite right then that is worse 
> than nothing not better.   I am not convinced, even looking over 
> the shoulders of giants, that the link problem has been solved properly.

I'm no expert in this either, but I suggest just doing xml:href.  If we're primarily interested in discovery then that gives us the 'where'.  Search engines are rarely interested in the 'why' and they can find out the 'what' when they get there.  (Completing the quintet, 'when' is probably irrelevant and 'how' is probably HTTP!)

It seems to me that the lesson from initiatives like XLink is that going any further than that is a voyage into semantics and we know that's a hard problem.  If a reader really cares about the 'why' then they have the option to fully interpret the relevant XML vocabulary.

I do feel that there might be a need for a MIME-like Content-Disposition that says whether the target is intended to be a separate entity or inline, but we can add that when people shout for it.  (Although I would probably implement this by doing something like xml:ihref for an "inline href" rather than "xml:href='x' xml:disposition='inline'".)

Pete Cordell
Codalogic Ltd
Interface XML to C++ the easy way using XML C++
data binding to convert XSD schemas to C++ classes.
Visit http://codalogic.com/lmx/ or http://www.xml2cpp.com
for more info