Re: "Last Look" at the RADIUS Design Guidelines document

[Alan DeKok <aland@deployingradius.com>]

Romascanu, Dan (Dan) wrote:
> Would it be possible for Alan and Avi to take it from this point of the
> discussion and try to write and agree on a few paragraphs of text which
> describe what this BCP covers and what is not covered and left out
> possibly for future work? 

  The document already says what it covers.  To summarize:

1) RADIUS has traditionally had a few simple data types which are used
   by most attributes (90% or more)

2) A small percentage of attributes have "complex" types that
   are not one of the "simple" types defined in (1)

3) The BCP recommends continuing practice (1) where possible

4) the BCP explicitly acknowledges that there are situations where
   practice (2) is acceptable.

5) the BCP does not forbid anyone from engaging in practice (2)


  That's really it.  I'm not sure how to reach a common ground that
excludes any of the above points.

  All of the discussion of dictionaries and server capabilities is
secondary to the above points.

  We *could* add text about "modern" servers.  It would look like:

---
  Many servers can define attributes via more powerful methods than the
traditional RADIUS dictionaries.  However, this capability does not
change the above recommendation to prefer simple types over complex ones.
---

  That's the point of contention.  We could *remove* all of the text in
the document about traditional dictionaries, and insert pages of text
describing how wonderful "modern"" systems are.  It doesn't *matter* how
capable or how incapable the implementations are.

  The BCP will still recommend simple designs over complex ones.

  Alan DeKok.

--
to unsubscribe send a message to radiusext-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/radiusext/>