Summary of OID and URI discussion and steps forward

Dear all:

After reading the messages exchanged dealing with
URIs and OIDs, I send this message to try to summarize
the current status of the discussions and to 
give my opinion on what should be the way forward
concerning to the document we are producing for
ETSI.
The first conclusion that comes to my mind is that,
as it was said in the ETSI meeting, this is a 
very general topic where different groups have
something to say (as it has been shown by the
suggestions on lists where this discussion should
also take place). This means that a definitive solution
in an inminent period of time is likely to be given.

Now, this is my recollection of the discussions. I have
identified different topics on discussion. I will 
present them and summarize the positions of the different
people involved. Please, feel free to correct me if I
misunderstood something:

----> LISTS WHERE THE DISCUSSION COULD TAKE PLACE.
 The following people have suggested to expand the
discussion forum to the lists mentioned below:

.Joseph Reagle and Martin Druest suggest the URI list in the W3C: 
	uri@w3.org
	(http://lists.w3c.org/Archives/Public/uri/)  (W3C side)

. Martin also adds the possibility of looking what the URI
Planning Interest Group is doing:

	http://www.w3.org/Addressing/Activity#current

. Michael Mealling has cossposted the messages to the
URN Working group list :

	urn-ietf@lists.netsol.com

.O. Dubuisson suggests asn1@oss.com  (ASN.1 side)


CONCLUSION: we have a good collection of places where
we can discuss on the topic. I think that to have
both sides of the issue (ie ASN.1 people from 
asn1@oss.com and URI's people from urn and uri lists
would lead to a solution resulting from an eventual
good consensus). So, I propose to send the messages
on this subject to the four lists listed above and 
the ETSI list (if
people of XML digsig think that this is not 
a specific topic for them to discuss, then perhaps
Joseph could notify it and the we should restrict
ourselves to the other three).

----> OID REPRESENTATION AS AN URI: WITH NAMES
OR WITHOUT NAMES?

A controversial point, as far as I have seen is 
that of coding the OID as an URI including the names
or not. 

People in favour of including names: Karl argues
the legibility of the OID if inclussion of the names
is done, and that synonynms would not represent
a problem.

People in favour of NOT including names: 
O. Dubuisson argues that synonyms could 
appear. 
Leslie Daigle argues that uncoherences between name
and digits can appear when writting....
Besides that, the initial text of the
RFC draft dealing with the representation of
OID using URIs envisaged only the inclusion of
the digits. 
It seems that now the point has raised a new
discussion on the people working in the
production of this draft, as shows the
message from Michael Mealling (by the way,
Michael, have you got any additional reaction
to the question by some other people apart
from Leslie? what is your personal opinion?.

I would say, summarizing, that additional
discussions should be carried also including
people from the asn1 list (they can give 
us their view on the importance of the 
text -in the end, OIDs were developed by
those people)...
Concerning to the document that we are producing,
I would say that if there are possibilities of 
having a RFC for incorporating the OID represented
as URI, we could envisage to allign our proposal 
with this solution.

But remembering what was said in the ETSI meeting, 
people saw as an advantage that for representing
OIDs as URIs that not administrative process should
be performed, whereas the allocation of URNs falls
below the umbrella of IANA and that this means 
to follow such administrative process.
Would not be this an overload?


----> OID REPRESENTATION AS AN URI: WITH DUPLICATION
OR WITHOUT DUPLICATION?

The other controversial point is the inclussion or not
of the organization specification of the OID in a URI
as a subordinate of the domain name (ie the repetition
of information that would point to the organization 
appearing in the OID).
This topic is related with the umbrella under which
the URIs would come... In the URN draft a complete
schema is shown and probably in its definitive version
it will say something about it.


Concerning to the way forward. What I will do 
is to send a message to the asn1, urn (do you 
agree Michael ?) and uri list 
presenting the discussion and in this way to see
if the discussion leads to some consensus  .....

Regards and thank you for your comments.

Juan Carlos Cruellas

Received on Wednesday, 29 November 2000 07:57:40 UTC