Edward C. Zimmermann writes:
 > Quoting Mike Taylor <[log in to unmask]>:
 >
 >
 > > So I think the request parameter should be:
 > > x-info-99-metadata=true
 > > or
 > > x-info-99-metadata=TRUE
 > > or
 > > x-info-99-metadata=t
 > > or
 > > x-info-99-metadata=yes
 > > or
 > > x-info-99-metadata=1
 > > or something similar. I am guessing that there must by now be a
 > > convention for representing "true" as a URI query parameter value --
 > > does anyone know where it is (and, better still, where it's
 > > documented?)
 >
 > To my knowledge there is (and really can't be but that's another issue) a
 > "standard". A parameter can even just be named without even a value.
 > Even in XML where we need an assignment (value) we can have #IMPLIED so
 > just naming it might be "implied" to mean its true (with the state of not
 > naming it being false). Now while we might have a syntactical model for
 > x-info-99-metadata='yes' or x-info-99-metadata='oui' or x-info-99-metadata='ja'
 > or x-info-99-metadata='1" etc. we need to assign, in our models, some meaning
 > that that's not always clear. Is x-info-99-metadata purely boolean? Is "1"
 > the same as "Oui" or "True"? What about "Jawohl"? "Yup"? Thus, I'd argue, we'd
 > be best off with the implied case! Its makes the value clear and immediately
 > also implies the states :-)

We already discussed this and decided that using a parameter with no
value, while aesthetically pleasing, would likely trip up some
toolkits.

So we've decided to use a value to mean "true" and the issue is what
value that should be. Not one of the most earth-shaking issues we've
ever had to decide, I admit :-)

 _/|_ ___________________________________________________________________
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ "Doesn't anybody live in one place any more?" -- Carole King,
"So Far Away"