Re: [Ecrit] comments on LoST

[Henning Schulzrinne <hgs at cs.columbia.edu>]

I generally agree with Andrew, with some additional comments:


> > * for dialstrings, suggest perhaps the KPML syntax for digit maps
>
> Aren't returned dialstrings to be stuck into tel or sip URIs?  If so, 
> would KPML be compatible?

From my view, KPML mainly adds the ability to express dial patterns. 
I'm not sure that capability is all that useful for services. Would you 
want to say "dialing any number starting with 9 gets you emergency 
services?" (Obviously, this also increases the mischief potential: "Dial 
[any number] to get to PizzaHouse pizza!")


> > * It'd be nice to be able, in a single lost query, ask for multiple 
> > services for the same location
>
> Given the conversation on multiple locations, I'd prefer we hold off on 
> this.  It is nice to have, but keeping complexity down is also a nice to 
> have.

To add to this, these services may also have different service areas, so 
you may not save a whole lot of bytes by doing that. Plus, you now have 
much more complicated failures: "Sorry, the third service you specified 
did not exist".  Note that different top-level services can also go to 
different LoST servers, so we now have to worry about people putting 
unrelated services in one query, creating yet another error case.



> Being able to have that would be nice, but server side authentication 
> will not work world wide.  I suppose it would depend on the TLS library, 
> but some would require a requery with different options if the first 
> query failed due to bad authentication.

I don't see client authentication as particularly helpful in general, 
given that the mapping information is generally public. Server 
authentication would probably generally follow the HTTP(S) model, where 
you might have the same information available in both modes.


> -andy
>
> _______________________________________________
> Ecrit mailing list
> Ecrit at ietf.org
> https://www1.ietf.org/mailman/listinfo/ecrit


_______________________________________________
Ecrit mailing list
Ecrit at ietf.org
https://www1.ietf.org/mailman/listinfo/ecrit