Re: More comments on draft-aboba-sg-experiment-03.txt

[Hannes Tschofenig <Hannes.Tschofenig at gmx.net>]

Hi Bernhard,

~snip~
> > For example, at IETF#69 we had an adhoc meeting on SPIT prevention. In short,
> > it was a disaster.
> >
> > We have a few solution approaches, we can envision a problem but we obviously
> > do not have a lot of SPIT to determine whether the proposed mechanisms would
> > help and which mechanism is more likely to be successful. 
> > This is a quite practical example and I would like to better understand
> > how the proposals in draft-aboba-sg-experiment-03.txt could help me to make
> > progress.
> >     
>
> >From your description, it seems that there is not yet enough data to 
> develop a problem statement, let alone to evaluate solutions.

Right. As you know from the Email spam the problem is only that waiting 
longer makes the problem substantially harder to solve later.
In some sense this is similar to security in general.

To make it more complicated it might not provide you a lot of additional 
insight when you know how the current attacks look like. The future ones 
will be different  anyway.

>   In such a 
> situation, it seems that the most that could be accomplished would be to 
> do a literature review,
We did this already. Most existing documents just say what has been 
stated in various IETF drafts already.

>  summarizing the state of the art and perhaps 
> identifying problems for further research.

I am not sure whether it is such a big research topic. If you, for 
example, look at XMPP then these folks have already published a lot of 
their specifications to deal with this subject. They are far ahead of 
the IETF crowd. They just do it and then figure out whether it helps or 
not. The worst thing that can happen is that it does not get implemented 
when people don't think it isn't useful.


>   I'm not clear that a SG would 
> be the best avenue for that, due to the short duration.  An IRTF RG might 
> make more sense. 
>   
I am not so sure about the value of IRTF groups.

>   
> > I could also come up with a simpler example, based on the early warning adhoc
> > meeting from IETF#69. It is a simpler case since we had constructive
> > discussions and there was a lot of support for the work. 
> >     
>
> Of of the top of my head, this sounds like it could potentially be a 
> better understood problem, for whichthe SG might be more relevant. 
>   
Good to know. Maybe we have a potential candidate for a SG experiment here.

I guess I would also like to see the outcome of this experiment. I 
wonder what you think about my other suggestions that go more along the 
lines of "providing help from experienced people"? Do you think that 
this will automatically happen as soon as a SG is formed? I don't think so.

Ciao
Hannes


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