Hello Henry,
You are quite correct the OMG has not yet registered the LSID NID with urn.arpa yet. I believe that as required by the LSID spec, the reference software implementations that are in use at this time default to the (b) NAPTR rule which substitutes in the authority portion of an LSID and queries that domain name for a SRV record indicating the whereabouts of the resolution server for that authority. As far as I know nobody has yet requested splitting the DNS domain from their LSID authority, which is no doubt why there has been no urgency to wrap this up, although things being as things are, I guess it is only a matter of time.
I happened to communicated with Dr Richard Soley (OMG Chair) a few weeks back on this very topic. In the BioRDF group I was requested to see if I could get a prototype working of the LSID -> HTTP Arc style resolver that you suggested and which I think will generally be perceived as a good idea although with the LS LSID crowd one never really knows! At least I think it is a good idea and am happy to prototype it since the software is 98% written already. Anyway, Dr Soley is strongly supportive of this new effort and has arranged for me to work with the OMG administrators to register an appropriate domain for the prototype etc. It seems he is also interested to see an update to the LSID spec that would reflect this addition (I am assuming if the prototype works well and there is interest ) in due course. I noted the LSID NID registration status and asked about the OMG's intentions and he reiterated his previous position that the OMG would register it and asked me to work with various people there to get it done. So I am on the hook.
Kindest regards, Sean
--
Sean Martin
IBM Corp
| [EMAIL PROTECTED] (Henry
S. Thompson)
09/26/2006 12:09 PM |
|
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 31 July 2006, Dan Connolly wrote:
> In today's call, I learned a bit about how LSID lookup
> works:
>
> (1) try to find an SRV record using the domain in the LSID
> (2) fall back to a DDDS lookup and NAPTR and such
>
> Option (2) depends on some NAPTR stuff in the "arpa" parts
> of DNS space that OMG is expected to install but hasn't
> gotten around to, largely for lack of urgent need, I gather.
Indeed I've spent a fair amount of time exploring the foundations of
DDDS for use in URN resolution, as proposed for LSID, which are
supposed to be located in NAPTR records for subdomains of urn.arpa.
As far as I can tell after extensive search, there are _no_ such
subdomains behaving as advertised. Indeed the _only_ active use of
DDDS/NAPTR I can find is e164.arpa, the root authority for the ENUM
story about DNS-resolvable telephone numbers.
There is only one URN namespace registered as mandated by DDDS, namely
PIN [1], but following the DDDS algorithm for it runs into the sand:
> dig pin.urn.arpa naptr
. . .
;; ANSWER SECTION:
pin.urn.arpa. 21600 IN NAPTR 100 100 "" "" "" pin.verisignlabs.com.
> dig pin.verisignlabs.com naptr
. . .
status: SERVFAIL
> dig @ns1.verisignlabs.com pin.verisignlabs.com naptr
. . .
status: NOERROR, . . . ANSWER: 0
So Verisign appears to have abandoned PINs -- if anyone knows of _any_
working subdomains of urn.arpa, please let us know!
This is all about (2) in your original list above. Could someone
point to the basis in the LSID spec [2] for your (1) -- I can't spot
it. . .
Thanks,
ht
[1] http://www.iana.org/list-archives/register-urn/msg00008.html
[2] http://www.omg.org/cgi-bin/apps/doc?formal/04-12-01.pdf
- --
Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
Half-time member of W3C Team
2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
Fax: (44) 131 650-4587, e-mail: [EMAIL PROTECTED]
URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)
iD8DBQFFGVDXkjnJixAXWBoRAjszAJ4ixuAfKRkJCR4nPWrkOquAlsHrkQCfXcx+
ZGzYEMKFvoX2AC200rYxL3w=
=YpH9
-----END PGP SIGNATURE-----
