Agreed.

On Thu, Jan 31, 2013 at 6:29 AM, Wendy Roome <[email protected]>wrote:

> I'm in favor of returning the complete map, rather than generating a error.
>
> Why?  Because errors are annoying to implement, and they're annoying to
> test in a bake-off. So whenever practical, I prefer to legislate errors
> out of existence by using reasonable defaults.
>
>         - Wendy Roome
>
>
> >Date: Mon, 28 Jan 2013 23:03:56 +0000
> >From: "Reinaldo Penno (repenno)" <[email protected]>
> >To: "[email protected]" <[email protected]>
> >Subject: [alto] ALTO Protocol Outstanding Issue I: Specify behavior of
> >       degenerated map filtering service
> >Message-ID:
> >       <[email protected]>
> >Content-Type: text/plain; charset="us-ascii"
> >
> >Discussion I:  Specify behaviors of degenerated map filtering service
> >
> >Do we enforce non-empty PID/AddressType? If protocol allows empty there
> >are two possible design choices: default to complete map or reporting an
> >Error
> >
> >Proposal going forward: Return complete Map.
> >
> >
> >Thanks,
> >
> >Reinaldo
>
>
> _______________________________________________
> alto mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/alto
>
_______________________________________________
alto mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/alto

Reply via email to