Re: Syntactic sugar for options: a failed experiment?

On 5/14/07, Norman Walsh <ndw@nwalsh.com> wrote:
> With respect, I think that would be worse.
Let's see:

> It still wouldn't allow the user to specify an option that's already in a namespace.

This is not worse, it is the same.

> It would allow the user to specify a value for options with names
> that clash with other step attributes

This is an argument in factor of using a namespace.

> having to declare (and remember to use) a different
> namespace for this purpose seems to outweigh any benefits of the
> shortcut.

This is arguable. Namespace declarations a little bit like "imports"
in Java: editors can help you add the relevant namespace declarations
and users often copy-paste those from other files. Also the namespace
declaration is done only once per file, while the full syntax for
options can occur multiple times. And at the end of the day, if
pipeline authors think they are better off using the full syntax, they
can do so: nothing is lost by providing the option of using a shorter
syntax.

But I agree: it is easy to argue both ways on this one. My pick overall: abort.

Alex
-- 
Orbeon Forms - Web 2.0 Forms for the Enterprise
http://www.orbeon.com/

Received on Monday, 14 May 2007 12:32:37 UTC