Registries, IVO ids, and Data Set Identifiers (fwd)

[Arnold Rots]

----- Forwarded message from Arnold Rots -----

Return-path: <arots at head-cfa.harvard.edu>
Received: from xebec.cfa.harvard.edu (xebec [131.142.52.100])
	by head-cfa.cfa.harvard.edu (8.12.8p1/8.12.8) with ESMTP id h8NGF2NT028426;
	Tue, 23 Sep 2003 12:15:02 -0400 (EDT)
From: Arnold Rots <arots at head-cfa.harvard.edu>
Received: (from arots at localhost)
	by xebec.cfa.harvard.edu (8.12.8p1/8.12.8) id h8NGEwZq014161;
	Tue, 23 Sep 2003 12:14:58 -0400 (EDT)
Message-ID: <200309231614.h8NGEwZq014161 at xebec.cfa.harvard.edu>
Subject: Re: Registries, IVO ids, and Data Set Identifiers
In-Reply-To: <3F6F0A85.9080200 at nasa.gov>
To: Tom McGlynn <Thomas.A.McGlynn at nasa.gov>
Date: Tue, 23 Sep 2003 12:14:58 -0400 (EDT)
cc: aaccomazzi at cfa.harvard.edu, borne at gsfc.nasa.gov, thomas at adc.gsfc.nasa.gov, 
	  mazz at ipac.caltech.edu, jcg at ipac.caltech.edu, gbb at ipac.caltech.edu, 
	  kimball at stsci.edu, arots at head-cfa.cfa.harvard.edu, 
	  paul at milkyway.gsfc.nasa.gov, tam at lheapop.gsfc.nasa.gov, 
	  jchavez at ipac.caltech.edu, registry at ivo.org
X-Mailer: ELM [version 2.4ME+ PL99 (25)]

Tom has nicely summarized the issues; there is some merit to being cut
off from th rest of the world and having nothing else to do but to
think and reflect.

A few comments:

I would argue (and have done so) that there are really only two
significant parts to the iedntifier - at least for the literature-data
link: an Authority Id and a Dataset Id (or fragment specifier).
For this particular use of the identifier there is nothing gained by
introducing the middle part - URI string or Resource Key.
The reason is simply that the registry needs to be able to provide a
URL translation for the authority.id-uri.string combination and the
distinction between authority.id and uri.string is meaningless in this
context.  Hence my preference for Tom's option 2: everything before
the first slash is to be looked up in the registry.
I would envision that the following actually happens:

	ivo://authority.id/dataset.id
	|________________| |________|
	         |             |
	     root.url   literal.string

The first part gets translated to a proper URL (or possibly another
ivo: identifier) and the second part is just appended.  For example:

	ivo://sa.cxo/2000

ivo://sa.cxo gets translated to
"http://cda.harvard.edu:9011/chaser/ocatList.do?obsid=", resulting in:

	http://cda.harvard.edu:9011/chaser/ocatList.do?obsid=2000


I would caution against using the last slash as the separator, since
it puts a constraint on what the authority can designate as dataset Id.

We might at some point decide to register sa.cxo.ftp and issue dataset
ids like 2000/primary/image.fits, so that:

	ivo://sa.cxo.ftp/2000/primary/image.fits

translates into:

	ftp://cdaftp.harvard.edu/pub/science/2000/primary/image.fits


I would also caution against allowing Identifiers that imply the
entire collection represented by the authority Id by omitting the
dataset Id.  I think it is far to be preferred if there is an actual
dataset id defined that explicitly refers to the entire collection.
I.e., use ivo://sa.cxo/all, rather than ivo://sa.cxo


I sympathize with Bob's comment about the list of authority Ids.
On the other hand, they are just strings and they are unique - and
that is really the only requirement.  I'm sure one could come up with
a better or more meaningful list, but I don't see that happen anytime
soon, so we might as well stick with what we have.

I will be e-mailing the ITWG list about some of the discussions that
Alberto and I had about ADS-data center services.

Cheers,

  - Arnold

Tom McGlynn wrote:
> Since I was cut off from civilization (or electricity at least) for a bit,
> I've been thinking about the issues that have been raised regarding IVO
> identifiers, registries and data set ids.  Below is
> a review of  the issues and a suggested synthesis.  I'm sending this to both
> the ADEC ITWG mailing list and the IVO registry group...
> 
> 	Tom
> 
> ...
> Formats that have been discussed include
> 
>     1. ivo://sa.rosat/x/rh300001n00
>     2. ivo://sa.rosat.x/rh300001n00
>     3. ivo://sa.rosat/x#rh300001n00
>     4. ivo://sa.rosat.x/#rh300001n00
> 
> ...
> 
--------------------------------------------------------------------------
Arnold H. Rots                                Chandra X-ray Science Center
Smithsonian Astrophysical Observatory                tel:  +1 617 496 7701
60 Garden Street, MS 67                              fax:  +1 617 495 7356
Cambridge, MA 02138                             arots at head-cfa.harvard.edu
USA                                     http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------

----- End of forwarded message from Arnold Rots -----
--------------------------------------------------------------------------
Arnold H. Rots                                Chandra X-ray Science Center
Smithsonian Astrophysical Observatory                tel:  +1 617 496 7701
60 Garden Street, MS 67                              fax:  +1 617 495 7356
Cambridge, MA 02138                             arots at head-cfa.harvard.edu
USA                                     http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------