Re: [e2md] RFC 5507 & RFC 5395

[Jim Reid <[e-mail]>]

On 16 May 2010, at 13:35, Bernie Hoeneisen wrote:

> To address the various objections I recommend everyone to read at  
> least the former (RFC 5507), as it explains quite some arguments  
> grey-beards and others have expressed before and during the BoF.

Bernie, while these links are useful they don't really help us make  
progress. In the context of the discussion on this list RFC5507 pretty  
much says there are two choices open to us: a new RRtype or a new  
subtype of an existing RRtype (=> NAPTR). If it's the latter, careful  
attention to the size of the returned RRset is necessary. This is not  
news.

So we're still running around in circles. Without clear use cases and/ 
or requirements, it's not possible to decide which of the two  
approaches above makes more sense. I am frustrated and disappointed  
that this information is still to be presented here (other than in  
vague anecdotal terms). It's pointless to continue a technical  
discussion unless there is an generally accepted problem statement.

With that in mind, I make the following suggestion. Could whoever is  
in charge of this BOF ask those who have use cases and requirements to  
submit them and set a deadline for that input? If nothing is  
forthcoming, then there isn't a problem to work on and we can all go  
home. And if we get that info, there would at least be a basis for a  
problem statement that could then be analysed and perhaps worked on.