VOEvent working draft published: version, param

[Tony Linde]

Hi Roy,

None of what I was objecting to in the <param> had to do with typing; it was
to do with the 'name' part of the name-value pair referred to by Rob. It is
informal and inconsistent naming of information that will more likely cause
downstream applications to fail than weak typing.

<<<<<
way. That is simple now. In XML-speak, we make a tiny change, from:
	<Group type="MyBrightness"> to <Group xsi:type="MyBrightness">
	The latter is now genuinely strongly typed -- it implies an object
called MyBrightness that inherits from an object called Group. The
MyBrightness type is defined in a schema, validated, etc etc.
>>>>>

Now that really has confused me. How do you get from a random group of
name-value pairs with no formal definition which therefore no software can
reliably be expected to interpret to a formal definition in a schema?
Doesn't sound like a small step to me.


Since you've both been talking about compromises as informing the makeup of
the spec, how about this:

   Add the option of extensions into the spec, explain how they make the
data both interpretable and correctly typed in order to secure the operation
of downstream applications, BUT also allow the <param> and <group> elements
as now. Then, if a provider either knows that their records are only being
sent to human endpoints or doesn't care about whether or not the downstream
apps correctly interpret the data, they can just use the base schema with
the <param>s for the specialised information, but for providers who do want
to ensure that downstream apps can properly recognise and interpret their
data, they can use the extension schema method.

Thus the downstream apps writer who is coding for human endpoints can simply
display the data as it comes in, whether it is in <param> elements or schema
extensions, and the writer of apps that are links in a chain of apps can
look for and use the formally defined data.

Does this allow us to agree a way forward?

Cheers,
Tony. 

________________________________

	From: Roy Williams [mailto:roy at cacr.caltech.edu] 
	Sent: 24 June 2005 22:27
	To: Tony Linde
	Cc: voevent at ivoa.net
	Subject: Re: VOEvent working draft published: version, param
	...