Jari,

>> I'm guessing that Jari's comment is because of the notion that documents
>> published as Experimental RFCs should include (unless it's totally
>> obvious) some discussion of how their success or failure will be
>> evaluated.
>
> That's part of it.
>
>> That doesn't IMHO mean that the WG is actually responsible
>> for the experiment.
>>   
>
> That is indeed possible. What I am saying that the work on experiments  
> is required. We could decide that the work happens elsewhere. However, I  
> think we've already demonstrated that its too easy to focus on protocol  
> bits in this space. I'm concerned that we'll forget the other parts. An  
> analogy would be promising to develop security later/elsewhere...
>
> That being said, I'm not necessarily opposed to agreeing that we need  
> steps a through z in this space and then ensuring that we have different  
> groups responsible for the different steps, particularly if the key  
> external expectations were explained in the charter. I still think that  
> some steps other than the pure protocol specifications should fall on  
> the proposed group. Its not just all about fun writing specifications,  
> the group also needs to take a real share of the more difficult tasks :-)
>
>> It's difficult in this case because the goal is to make the Internet
>> scale much bigger than it is today, and that would be quite an
>> experiment ;-)
>>   
> Brian, you know very well that we can test and evaluate new technology  
> in various ways before its deployed.

        No one is suggesting that the acquisition of further
        knowledge and understanding of all of this isn't a good
        thing (in fact, that is what I was trying to do with 
        draft-meyer-loc-id-implications-00.txt). So quite
        obviously increased understanding is a good thing, and no
        one is arguing that.

        The problem here, and the one that I and others are
        pushing back on, is the conflating the experimentation
        (etc) with what IETF WGs are generally chartered to do,
        namely, build protocol specs. What is being proposed in
        the LISP WG charter is tightly scoped WG to finish up the
        LISP specs, and publish them as EXPERIMENTAL (this is all
        documented on http://www.1-4-5.net/~dmm/lisp.charter).
        You will also note from the charter that it is an
        explicit non-goal to provide architectural surveys or
        analyzes, or to provide experimental design for future
        experimental studies. 

        That said, additional studies (e.g., analysis, taxonomy,
        experimentation of various varieties, etc) is will naturally
        occur,  whether it be in the I*TF or elsewhere. Further,
        you might also note that some of this work is already
        occurring. See for example
        
http://scholar.google.com/scholar?q=LISP+Locator%2FID&hl=en&lr=&btnG=Search 


        What I find truely odd about this discussion is that in
        the 5 cases when I have had WGs chartered, my ADs have
        asked me to tighten the scope of the charter rather than
        expand it in the open-ended ways have been suggested
        here.

        Dave

Attachment: signature.asc
Description: Digital signature

_______________________________________________
Int-area mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/int-area

Reply via email to