CIBA: binding_message restrictions

[Takahiko Kawasaki]

It will help implementers if the description of the binding_message request parameter also includes detailed explicit restrictions on its length and valid characters as that of the client_notification_token request parameter does.

[Brian Campbell]

Based on prior discussions, I'm not sure the WG can agree on such detailed explicit restrictions. Which is a large part of the reason for the current text that is suggestive but not explicit, "... the binding_message value SHOULD be relatively short and use a limited set of plain text characters".

But we can discuss on the call today and see where it goes.

[Dave Tonge]

Agreed - it is likely to be implementation specific

[Brian Campbell]

We didn't get a chance to discuss on today's call because, well, the call time got used up on other things ...

[Dave Tonge]

I don't think this is a blocking issue to us splitting the draft. We could keep the issue open until the next call, but I'm pretty sure that we agreed in previous calls on the current wording.

[Takahiko Kawasaki]

In actual use cases, I guess that the length of a binding message will be 10 or less and their characters will be printable ASCII letters only and that implementers will reserve a larger size in their databases just in case. Anyway, thank you for your comments.

[Dave Tonge]

We discussed this on the call and decided not to change the current wording