Wait, Eric, if you're saying that you beleive the COinS spec ALREADY
means that you must make your COinS data payload a proper valid URL
string (that is, you must URL-encode : and / as well as &; and further,
you CAN URL-encode any other char you want)---then I think we all (at
least everyone who has commented on this thread so far) agree.

But that disagrees with your earlier posts on the topic. When you said
you thought it was important to let COinS producers have the option of
not producing Strings that were valid URLs if they wanted. So I'm
confused. (I've started using the cumbersome phrase "Strings that are
valid URLs" in recognition of the point Eric made that "URL-encoded" is
a somewhat ambiguous phrase, since any char CAN be URL-escaped in a
URL, but only certain chars MUST be.)

The suggestion:
1. A proper COinS context object string must be a valid URL. Or it is
not a proper COinS string. That means that / and : as well as & must be
encoded. (Any other chars?)
2. A COinS context object string MAY include URL-style-encoding on ANY
char, as in a URL. So processors/activators should be prepared to deal
with this.

It kind of seems like everyone agrees with this at this point, so I
don't see what people are arguing about? If both Ross and Eric agree
with these things, then they certainly do understand each other's
positions enough to resolve this particular question(s).

The only other thing is:
3. The spec document should clarify both these things.

If everyone agrees with #1 and #2, I can't see why there would be
opposition to #3.  Because there HAS been confusion on this point,
various COinS producers and consumers HAVE been in disagreement about
these things, so it seems pretty indisputable to me that it should be
clarified in the document.

On Dec 12, 5:19 pm, Eric Hellman <[EMAIL PROTECTED]> wrote:
> since Ross and I still don't understand each other's positions, I
> think it is premature to "vote" on anything.
>


--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups 
"gcs-pcs-list" group.
To post to this group, send email to gcs-pcs-list@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/gcs-pcs-list?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to