What do we mean by GUID?

Matt Jones jones at NCEAS.UCSB.EDU
Wed Oct 12 08:09:39 CEST 2005

 From my perspective several of these issues are pretty clear.  Let me
attempt answers below and see if we're all in agreement.

Chuck Miller wrote:
> I have been pondering this question of what exactly is meant by a GUID
> since Donald's first call for input.
The term GUID is one we started using in SEEK when looking for a
solution to the identity and resolution problems that we saw looming for
the Taxonomic COncept Standard.  Dave Thau's presentation on this
(linked on the GUID wiki) defines this pretty well and explores the issues.

> First, is it correct that GUID stands for Globally Unique Identifier?
> Second,
> What do we mean by Globally?
> What do we mean by Unique?
"globally unique" means simply that an identifier that is issued can
only have one valid interpretation across all possible systems.
Regardless of the mechanism used to resolve the identifier, the object
that the id 'identifies' will be bit-for-bit identical.  However, note
that a given object can legitimately have more than one identifier.

> What do we mean by Identifier?  Or, specifically what are we identifying?
This is a bit trickier.  We will clearly be identifying several
different types of objects, some of which are physical (e.g.,
specimens), and some of which are digital (e.g., observation data).  On
the specimen side, resolution of the identifier and retrieving the
'data' makes no sense because the 'data' is a physical object that
cannot be electronically transported.   On the 'digital' side, it makes
sense to resolve and retrieve the data.  There are some tricky issues
dealing with granularity of the identifier for digital data (does the
identifier point at a tuple in an entity, or at a whole entity, or at
multiple entities).  In addition you still have the very thorny issue of
what is data and what is metadata.  I'll write another note regarding
this issue.


> I believe we are far from consensus on all three of those definitions,
> even the meaning of globally.  And, I believe we will be going around in
> circles on LSIDs, ARNs, and such until we get the expectation more
> clearly defined.
> What I hear in most of the discussions so far are descriptions of a GLID
> - Globally Locatable IDentifier.  In the Internet world, GLIDs started
> with the URL - Universal Resource Locator which has evolved to the URI -
> Universal Resource Identifier concept.  Another form of URI is the URN -
> Uniform Resource Name which enables a persistent name, independent of
> server location.  This is the kind of thing I think we want and are
> discussing in this GUID thread.
> I think we should draw a distinction between GUID and GLID.
> An identifier of a thing can be globally unique without stating its
> location.  But, again, it raises the question of what the definition of
> unique is.  An ISBN number identifies a book "uniquely", but there may
> be millions of "unique" copies of it.  Similarly, duplicate sheets of a
> collected plant specimen are all from the same "unique" organism and may
> each even be referred to by the same "unique" collector and number. But,
> each sheet itself is also unique.  We need a clear definition.
> An identifier of a place can also be globally unique, like a URL.  But,
> being able to go to that place requires a global infrastructure to
> handle the addressing.
> Where it gets really messy is when we want an identifier of a thing that
> is unique but can move around to different places, like a URN.  The
> addressing has to work like an administrative assistant who keeps tabs
> on where the staff is currently located so she can direct phone calls to
> them.  Without the administrative assistant, people who move around
> can't be contacted. It looks like a lot of what LSID, ARN, and such seem
> to about is "administrative assistant" addressing schemes, how to
> navigate to the entity through layers of address abstraction.  But, in
> each case it raises the issue of who/where is the administrative
> assistant, on top of the question of the addressing scheme itself.
> Shouldn't we get these definitions and expectations nailed down first?
> Then look at solutions?
> Chuck Miller
> Chief Information Officer
> Missouri Botanical Garden
> 4344 Shaw Boulevard
> Saint Louis, Missouri 63119
> Phone: 1-314-577-9419
> Cell: 1-314-614-6952

Matt Jones
jones at nceas.ucsb.edu                         Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)
UC Santa Barbara     http://www.nceas.ucsb.edu/ecoinformatics

More information about the tdwg-tag mailing list