This is an extended, revised version of . Findability
and interoperability of some PIDs and their compliance with the FAIR data principles
are explored, where ARKs were added in this version. It is suggested that the wide
distribution and findability (e.g. by simple 'googling') on the internet
may be as important for the usefulness of PIDs, as the resolvability of PID
URIs. This version also includes new reasoning about the failure to use PIDs such as
DOIs for citation. The prevalence of phenomena such as link rot implies
that the persistence of URIs cannot be trusted. By contrast, the well distributed,
but seldom directly resolvable ISBN identifier has proved remarkably resilient, with
far-reaching persistence, inherent structural meaning and good
validatability, by means of fixed string-length, pattern-recognition,
restricted character set and check digit. Examples of regular expressions used for
validation of PIDs are supplied or referenced. The suggestion to add
context and meaning to PIDs, making them "identify themselves", through
namespace prefixes and object types is more elaborate in this version. Meaning can
also be conferred by means of structural elements, such as well defined, restricted
string patterns, that at the same time make PIDs more "validatable". Concluding this
version is a generic, refined model for a PID with these properties, in which
namespaces are instrumental as custodians, meaning-givers and validation schema
providers. A draft example of a Schematron schema for validation of "new" PIDs in
accordance with the proposed model is provided.
Introduction: Identifiers in science
Identifiers in science may refer to digital or physical objects, or concepts.
They may also refer to persons, researchers,
as ORCiD IDs , or to research organizations, as the
recently launched RORs . This paper will focus on
persistent identifiers, PIDs, for research outputs, 'things' such as
articles, datasets, samples, concepts etc. But, as suggested in section 7, ORCiDs or
RORs may be an optional part of a modular, integrated identifier for research
outputs. PIDs may be general or domain-specific.
Among the more prevalent general PID-types are ARK, DOI, Handle and UUID. There are
also old, bibliographic identifiers like ISBN. Created in the 1960's and -70's of
the print era, how come they survived into this digital age? Some reasons might be:
they are well distributed across the internet, and widely used by stake-holders
(libraries, publishers, readers). They have a semantic structure, identifying
well-defined objects, and a fairly precise validation mechanism through fixed
string-lengths, limited character-set and check digits. Some of these properties of
good identifiers are shared by ARKs, DOIs, Handles and UUIDs, or other more domain
specific identifiers used for scholarly data, but seldom all of them simultaneously.
The focus here is on findability and 'validatability' of PIDs of different
types.
Identifiers - why do we need them?
The general purpose of identifiers is to serve as references to the objects
that they are supposed to identify. This requires identifiers to indicate,
preferably in and by themselves, what type of objects they are meant to
identify. Far from all identifiers fulfil this requirement. It is often left to the
names of things to describe the objects identified, providing
context and meaning. Context may be added
by means of location within an hierarchical system, e.g. as in Linnéan taxonomy,
where scientific names situate a species within a genus, sometimes also containing
the provenance of that name, serving to disambiguate beween names of species
belonging to widely different genera, e.g. Asterina gibbosa Gaillard 1897
(a fungus), and Asterina gibbosa(Pennant, 1777) (an echinoderm, a
starfish). For the same purpose, it also happens that 'things', objects are renamed
later, as with the preceding fungus species now having the accepted scientific name
Asterolibertia gibbosa (Gaillard) Hansf. 1949 and / or are assigned an
identifier:
urn:lsid:catalogueoflife.org:taxon:02af8238-ac8f-11e3-805d-020044200006:col20150401.. However, even if a PID may well serve the need for
disambiguation by uniquely identifying an object, it may still be no better -
sometimes, as in this case, perhaps even worse at giving access to said object, or
at least a landing page with metadata about it. The identifier assigned above is
neither directly resolvable nor 'googlable', while the scientific name is at least
easily findable via a search engine.
While scientific names are often useful for describing objects,
they have other drawbacks compared to PIDs, some of which were identified by . For example, homonymy and disambiguation should be
no problem for globally unique identifiers. And while concatenations or
abbreviations may be problematic in the use of names for identification,
string-length and pattern restrictions are useful for validation of identifiers.
Missing or added characters, all types of misspellings are easier to detect and
validate in standardized identifiers of fixed string-length or well-defined
character patterns. Inconsistent encoding should also not be a problem in good PIDs,
with limited character sets. However, these assets of some identifiers may conflict
with the interest in having also transparent, meaningful PIDs that at least in part
"speak for themselves". The result of a compromise between these two interests may
be seen in the Handle system (below).
FAIR principles
The FAIR guiding principles aim to make (meta)data Findable, Accessible,
Interoperable, and Re-usable. As
such they concern also PIDs, as is seen from some of the principles:
The FAIR data principles.
The FAIR principles clearly need interpretation to become fully operational, and such
work is also well in progress. Further explications of some of the principles
are also available in . Figuring prominently in the
explications of all these principles, particularly for interoperability, is
the requirement that metadata should be machine actionable, a conditio sine
qua non for FAIRness.] This is also used by fairmetrics.org as a measure of
Findability.
However, the FAIR principles do not say anything explicitly about
validation. Particularly for the principles of
Interoperability and Re-usability, it is crucial that metadata
can be properly validated against a schema, as adhering to an accepted metadata
standard. It has been remarked that this is already
implied by the FAIR principle R1.3 above, but if so, only indirectly and still open
to interpretation. There are several cases where general data repositories,
professing to be FAIR and adhere to accepted metadata standards both for their
default output and export formats, nevertheless fail to validate against schemas of
these same standards.Fairmetrics.org explicates R1.3, as measuring a "Certification
from a recognized body, of the resource meeting community standards", by means of a
valid electronic signature, such as a verisign signature.
However, one might ask whether general data repositories such as Harvar'ds Dataverse,Figshare or Zenodo, qualify as "recognized bodies" in this respect, all being part of
the test reported in "Evaluation_Of_Metrics/Supplementary Information_ FM Evaluation
Results.pdf" , but none of which could be
evaluated on R1.3. This comes as no surprise, since there is already a comment on
R1.3 saying that "Such certification services may not exist, but this principle
serves to encourage the community to create both the standard(s) and the
verification services for those standards." True, in the
rationale for FM_R1.3 there is mention of validation: "... As such, data should be
(individually) certified as being compliant, likely through some automated process
(e.g. submitting the data to the community's online validation service)". But it remains unclear what the "community" here refers to
a general metadata standard, or a repository, using it's own standard and validation
service? Some output metadata files from repositories even lack a
schemaLocation reference, making it difficult to validate them, or, the
schemaLocation given might be erroneous, as observed in one case. We cannot just wait for repositories themselves to
provide verification of compliance with standards. We must use available validation
tools, testing to what extent they are keeping their promises of compliance
compliance. This concerns metadata in general, but
naturally includes also identifiers. We must be sure that they are of the type or
format they claim to be, even if they cannot be resolved to a dedicated
landing-page. Failed validation, e.g. due to simple typos or wrong namespace, may
even be one way of checking why an identifier or URI does not resolve as expected.
Validation is also important for the possibility to export metadata to another
format, thereby promoting the re-use of data, without exporting also potential
errors. Resistance to transcription errors, e.g. by means of a
restricted character set, using base32 for encoding, and fixed string-length (suffix
has 2 times 4 characters, separated by a hyphen), has been promoted as an advantage
of so-called "cool DOIs". These are precisely the kind
of properties that make PIDs eminently "validatable" and thereby
machine-actionable, a crucial requirement for FAIRness.
Although transformation or harvesting of metadata might be
possible even without validation, trust in the results and quality as well as the
eventual findability of the data (and so again the re-usability) might be seriously
affected. For enhanced findability, it is also important that standard, widely
distributed identifiers are used.
Validation of an identifier means ensuring that it is true to its proclaimed type,
for example, making sure that what is flagged as an ISBN is not in fact an ISSN
(real use case), or that the string-length and check-sum is compliant with its type.
An advantage of promptly validatable identifiers is that validation can often be
performed also off-line, by means of a pattern expressed by a
regular expression, a piece of script (JavaScript, Python, etc.), an HTML
form, or a schema (e.g. XSD or Schematron) and a piece of software
such as an XML-editor.
Resolvable or findable?
In the present FAIR principles the focus is very much on resolvability of
identifiers, despite the general awareness of phenomena like 'link rot' and
'reference rot'. It has even been
suggested to put up digital tombstones over disappeared resources, with metadata
from their last known whereabouts serving as epitaphs. A
2013 study in BMC Bioinformatics analyzed nearly 15,000 links in abstracts from
Thomson Reuters’ Web of Science citation index and found that the median lifespan of
web pages was 9.3 years, and just 62% were archived.
This happens although there is an understanding that [u]nique identifiers, and
metadata describing the data, and its disposition, should persist -- even beyond
the lifespan of the data they describe. A
recent study of some 40 research data repositories found that only one of these (3%)
was compliant with the FAIR principle of Accessibility requiring a clear policy
statement (or various examples of data this has actually happened to) indicating
that metadata is still available even if the data is removed. The argument here is not that resolvable, persistent
URIs should be avoided as identifiers; most often they serve their purpose of more
persistently giving access to a metadata source than "ordinary", plain URLs. But, as
has been eloquently remarked, "persistent URIs must be used to be persistent". Resolvable URIs as PIDs work by decoupling the
location and the identification functions of URIs.
The custodian of a web resource maintains the correspondence between the
identifying URI and the locating URI in the resolver’s look-up table as the
resource’s location changes over time. ... The solution comes at a price because
it requires operating a resolver infrastructure and maintaining the look-up
table that powers it.
This is true of ARKs, DOIs, as well as Handles, PURLs and URNs. There are in fact
numerous cases when the lookup-table is not maintained and updated as required. That
is why it may be wise not to rely on a single 'custodian' for the resolution of
identifiers and access to associated metadata. Note that we
are not talking here about simply having more than one proxy server acting as
resolvers of the same PIDs. We already have that; provided the lookup-table is
managed properly, the three different DOI-URIs from three different proxy-servers
all resolve to the same landing-page location: https://doi.org/10.1007/978-3-319-53637-8_11, https://hdl.handle.net/10.1007/978-3-319-53637-8_11 and https://identifiers.org/doi:10.1007/978-3-319-53637-8_11. ARKs (Archival
Resource Keys) are resolved by identifiers.org and n2t.net, as
well as by their "mother institutions", e.g. n2t.net/ark:/67531/metapth346793/, identifiers.org/ark:/67531/metapth346793/ and digital.library.unt.edu/ark:/67531/metapth346793/ resolve the same content.
It is rather the distribution and use
of identifiers - whether resolvable or not - that is important here. It seems not
even the authors of are true to their own
principles, since three of their references that actually have DOIs are cited
without them: . So, despite having DOIs
or other PIDs assigned, documents are often not cited by those PIDs. One possible
reason might be that the PID is not clearly displayed in the landing page with
metadata, or in the document itself. In the case of
above it takes an extra click on a link 'Cite as' to actually have the DOI
displayed. But that should hardly be the reason why it was not used for citation in
, since the citation there is actually much
more verbose and complex, than it would have been to just copy-paste from the 'Cite
as' page above. Another, slightly ironic case concerns , the founding paper of the FAIR principles, where you either have to download
the citation with the DOI from the landing page, where it is not displayed, or find
it at the bottom of each page in the actual paper, but not prominently marked. This
may partly explain why a recent paper on software sustainability and
reproducibility, while arguing that one of the ways to make software more
reproducible is to "use a persistent identifier such as a Digital Object Identifier
(DOI) to help find and cite code" , failed itself to
use the DOI when citing . Another reason, gathered
from one of the authors by personal communication, but which I believe could be
generalized, is that inclusion of the DOI (or other PID) was not part of the
citation style of the publisher. In fact, it sometimes happens that publishers
impose their own citation formats or standards, excluding the use of PIDs.
Again, PIDs must be used and cited to persist.
Citations help in achieving wide distribution, depending on good 'validatability',
in order not to multiply errors and 'non-resolution' as a result.
Again, going back to the question of resolvability, the relationship between
identifiers such as DOIs and URIs/IRIs is not always straightforward, and sometimes
involves a chain of redirects ('303s'), before reaching a destination holding also
the appropriate metadata..
Another reason resolvability may not be sufficient, even if the metadata is somehow
in place, is that the file on the destination page resolved to is behind a paywall.
In a case from December 2016, public domain content more than 110 years old was
hidden behind a DOI-resolver charging 50$ for release of the content.
When someone in an ensuing Twitter conversation complained about this, an answering
tweet seemed to mean, that was the price we have to pay for something as useful as
DOIs. There was also a remark to the effect that every object gets one and only one
DOI. But this is not necessarily true. It is possible to mint several DOIs for the
same resource by different agents, such as Dataverse, Figshare, Zenodo etc. If not,
that would give commercial publishers the opportunity to pro-actively seize whatever
public domain content there is out there on the internet, quickly mint and assign
their DOI to it and then lock it up behind paywalls. In any event, it would
certainly not promote the use of PIDs instead of simple, ephemeral URLs for
citation.
The DOI in question then was: 10.1080/00222930908692639. The remedy proposed to "jump over" the
paywall was by means of what was then oadoi.org, which still worked quite well. In several steps it eventually led, via an
API, to an XML-file with a link to the freely accessible fulltext at http://www.biodiversitylibrary.org/part/60220. However, things have
changed. When tried again in November 2018, the replacement unpaywall.org and oadoi-API for 10.1080/00222930908692639 was not working anymore for this DOI; the
response received was: best_oa_location: null. But the resource sought,
free from paywall, although no longer detected by unpaywall.org, could
still be found e.g. at biodiversitylibrary.org. However, often you will not find a free
replacement copy of paywalled resources. An earlier "error message" (December, 2016)
from oadoi.org for , said this holds for around
80% of scholarly articles. Maybe the situation has changed
for the better since then. In January 2018, the unpaywall.org FAQ-page stated: We find fulltext for 50-85% of
articles, depending on their topic and year of publication. But, this
message is no longer found there (in November 2018). And as we have seen in the
example above, apparently in some cases unpaywall.org no longer finds open
access fulltext alternatives, that was previously found by the oadoi.org
application used then.
If unpaywall.org or an oadoi-API (e.g. https://api.oadoi.org/10.3897/biss.2.25805?email=p@x - the 'email' argument
as given here may well be "fake", as long as it holds the @-character) fails to find
a freely accessible publication (as with https://api.oadoi.org/10.1007%2Fs10654-018-0449-x?email=n@y), an
alternative might be to try the identifiers.org SPARQL endpoint . But, it
does not necessarily give us an open access URI in return. And it only works if the
potential corresponding URIs have been assigned the property owl:sameAs
just as the submitted subject URI. Unfortunately, in neither of our cases above
these conditions are met.
Assuming we have finally found a single seemingly reliable custodian of our PIDs and
URIs, promising 24/7 resolution and top quality metadata, should we rest content
with that? Most serious lawyers and journalists probably would agree: it is wise not
to judge by the testimony of a single witness, a single source alone. The evidence
of at least two, mutually independent witnesses is generally preferred. Multiple
resolution of any PID by several different proxy servers, as we already
know, still means single parenthood, single custodianship of that lookup-table that
has to be managed and updated in order for the PID to resolve as expected. Clark
describes it as representing a stage in the evolution of PIDs, that will eventually
be surpassed by a more mature age when we supply also data types to come
with the PIDs, in order to make them more machine actionable . But we want more than that. We want backup custodians, relatives or friends
as caretakers when the poor single parent fails duty. Providing multiple
access to, or identification of resources through PIDs, that
are capable of serving as trustworthy, competent, valid independent witnesses from
different moments in time, at different sites, in different places is a good idea.
Thus, we accept that an object may have multiple PIDs. Ideally these multiple
PIDs should get to "know about" each other as a way towards interoperability. This can be achieved already, e.g. by means of linked
open data (LOD), sameAs-relationships and tools provided by n2t.net,
unpaywall.org and the identifiers.org SPARQL endpoint referred
to above. Multiple identifiers from different namespaces for the same object may
even be desirable in order to ensure interoperability in different environments.. It is also in line with the principle of the semantic
web known as the NUNA, Non-Unique Naming Assumption, implying that
things described in RDF data can have more than one name and any object
may be identified by more than one URI, serving in RDF as 'names' of things.
However, this does not imply that any identifier, any PID is as good as the other. In
fact, there are significant differences in quality between identifiers, particularly
in terms of 'validatability' and 'meaningfulness'. We are
getting there a bit later.
But first, having referred to linked data and sameAs-relationships as a possible
solution to achieving interoperability, what about long-term sustainability? Are
LOD, relying heavily on URIs, fit for survival? Archival records for long-term
preservation need to be self-sustained, carrying meaning within themselves, while
the references may no longer be resolvable. In e-archives compliant with the
OAIS-model and Trustworthy Digital Repositories standards for self-sustenance, this
means that URIs lacking an inherently meaningful structure will often serve only as
another set of dumb identifiers. Unless they can import some meaning from outside,
through resolution or sameAs links, such opaque, non-resolvable URIs should
henceforth rather be described as "non-semantic".
Which identifiers are FAIR enough?
We must ask about PIDs, Persistent Identifiers, just how persistent
they are really? Even if not always resolvable, are they in general still
'findable', well distributed over the internet in time and space? Are they
'validatable' (e.g. through fixed string-length, pattern-recognition, restricted
character set, built-in checksum, built-in type?) Are they FAIR?
Findability: Beginning with the F for findability, for comparison we
go back in time to 'old-fashioned' ISBNs, Internaional Standard Book Numbers.
Publicly declaring what type of objects they are meant to identify, ISBNs are rarely
directly resolvable. But they are widely distributed, they have good
findability in terms of precision hits, as seen by simple
'googling', with good survival rate, longer than the median age of web-pages 9.3
years. For example, look at ISBN 0-14-029161-X: The Diversity of Life / Edward
O. Wilson (2001). Simple googling of 014029161X, unprefixed and without
hyphens results in 57/57 precision hits (date: 2017-01-30). ISBNs could also be
searched in library catalogs, the most comprehensive of which is probably the Karlsruhe Virtual Catalog – KVK
worldwide. Result of the query '014029161X', with the same unprefixed ISBN
without hyphens yields 123/123 precision hits, recall being difficult to compute
since in 55 of 72 catalogs the search could not be successfully processed or no
records were found. To counteract the possibly unfair bias with a modern classic
like this, we try instead an even older, and presumably less well-known example:
ISBN: 2130381030. L'Identité : séminaire interdisciplinaire dirigé par Claude
Lévi-Strauss, 1974-1975 (Paris: PUF, 1983). Googling without prefix
(2130381030) the precision is between 14/39 and 22/50; with prefix (ISBN2130381030)
it reaches as high as 17/18 (date: 2017-01-30).
Accessibility: Data and (digital) objects are accessible only in so
far as identifiers are findable or resolvable preferably to open access landing
pages with either direct availability of resources, or sufficient metadata to direct
the user to such an access point. In this respect DOIs are often, but not always, as
good as or sometimes better than ISBNs (for obvious reasons regarding print only
material), while gni-UUIDs as described above are all but useless.
Interoperability and Re-usability are both
intimately associated with 'validatability', as argued above. We will look more into
detail at the performance of different PIDs regarding this below.
Archival Resource Key (ARK) Identifiers: ARKs have a well defined
syntax :
[http://NMA/]ark:/NAAN/Name[Qualifier] , where NMA is a
(changeable) Name Mapping Authority, a "host" or proxy resolving agent.
This is not part of the ARK proper, as marked out by the encompassing brackets. The
NAAN is the Name Assigning Authority Number, corresponding to
the prefix starting with '10.dddd' in a DOI, and serving as a namespace for the
following /Name. The NMA-supported [Qualifier] is not further defined in , but an example is given by the suffix
s3/f8.05v.tiff, including also a file extension as we can see. As
examples below, none of them having a qualifier, we find ARKs giving direct access
to digital fulltext of Buffon’s Histoire naturelle at BnF and a 20th
Century Guide for mixing fancy drinks in Internet Archive, here with
identifiers.org as resolving NMA. In the third case, we see a resource
with a prominent feature of ARKs, their possible inflections, here
represented by the two '??' at the end of the URI, giving both metadata for the
'document' itself, a photo of the Dallas Police Department from 1963, and the name
and location of the collection holding it. This inflection property of ARKs might
serve as a response to the requirement ensuing from the FAIR principle of
accessibility (A2), that "metadata are accessible, even when the data are no longer
available" , i.e. resources that have for some reason
been removed from the net should at least leave a gravestone with metadata of last
known whereabouts behind to the afterworld. Another interesting, perhaps
unintentional feature of this particular case is that when the resolving agent, the
NMA is changed from texashistory.unt.edu to either identifiers.org
or n2t.net, the very same ARK does not resolve to the same landing page,
the same location.
What about the "validatability" of ARKs, then? As obvious already from the examples
above, apart from the specific structure, there is no specific pattern or definite
string-length of an ARK. The only restrictions on the Name and Qualifier parts "as
strings of visible ASCII characters" is that they "should be less than 128 bytes in
length" using "letters, digits, or any of these six [Sic!] characters: = # * + @ _
$", allowing also four more characters with reserved meaning: % - . / These general restrictions are hardly sufficient for
efficient validation by means of a regular expression. It is possible to impose
restrictions defining a specific pattern and string-length within a namespace, for a
particular NAAN, using for example a NOID template , but
this is not something that is required by the ARK specification.
More heavily incumbent on an ARK seems to be the demand that, "[i]f longevity is the
goal, it is important to keep the prefixes free of recognizable semantics". This, I believe, is a misconception. It is not the
absence of semantic content that guarantees longevity; it is rather the continued
use of the identifier that enhances its persistence, as observed again
by above, and this use may well be promoted by at
least some semantic content in the identifier, allowing a user to recognize it as an
identifier of precisely that particular object that it is supposed to identify. The
Findability by simple 'googling' and current Accessibility of
the example ARKs above presently (Nov. 2018) still seems quite good. At least the
first of these three examples seems to be well distributed, producing an impressive
precision score of 25/25 by simple googling of '12148/bpt6k97497t' (each page in the
hitlist actually containing a reference to the same document by Buffon in the
Gallica collection). The second example apparently has a narrower distribution, but
the few items found still display good precision, 4/4. The third example, without
inflection, has been used extensively as a paradigmatic case, so should perhaps be
considered outside competition here, but anyway also shows good precision. And,
given the limited validatability, the degree to which the resources they
identify will also be Interoperable and Re-usable will very much
depend on the difficult to predict long-term sustainability of these two first FAIR
properties of these ARKS.
DOI: DOIs can look just like anything. Here are some real cases, all
at the time of writing resolvable and with multiple Findability also by simple
googling, some of them pretty 'old', although they got their DOIs assigned fairly
recently. One is even from 1977 (doi: 10.1177/030631277700700112), but it still
produces an impressive precision score of 68/68 (date: 2018-11-07), mostly due to it
quite high citation rate, yielding hits for all the citing sources.
Now, following are two old DOIs from Wiley Online Library 1996 and Springer 2001 that
still do not seem to resolve properly (neither on 2017-01-31, nor on
2018-11-11):
However, the two following old DOIs, again from Wiley Online 1996 and 1998, that were
similarly unresolvable at the same date (2017-01-31), are proof that some PIDs might
regain there resolvability later.
Obviously, all these DOIs, whether resolvable or not, vary substantially in
string-length, from just 17 to over 60 characters, some involving abbreviations of
journals or organisations, one an ISBN, and some containing characters in need of
special XML-encoding, different from URI. Note that although the two last items in
the first group are from the same journal, Scientometrics, they are quite
different in structure. Anyway, all the above DOI examples are valid in
accordance with the best we can offer as a regular expression restriction, with only
partial pattern recognition: ^10[.][0-9]{4,}\/\S+$
meaning that any valid DOI must start by '10.' followed by a minimum of 4 digits,
before the slash '/' and then a suffix of any length or characters, but no spaces in
between.
But then, according to the same partial restriction, defined by the regex
above, this entirely fake DOI is equally valid:
10.99999999/xxxxxxxx/x(y)x\:-{=?%%@@@@@
To be sure, there are other regular expression restrictions suggested for DOIs, those
that are even more permissive (as DataCite 4.1, with the pattern value for doiType
set to "10\..+/.+" , apart from not being PHP or
JavaScript compliant, allowing also inline spaces, or the pattern registered for
DOIs at identifiers.org as "^(doi\:)?\d{2}\.\d{4}.*$" MIR:00000019, both of which also allow for the fake DOI above as valid,
when tested in regex101.com).
There are other patterns that are more restrictive,
but then obviously not catching all the now prevalent and permitted DOIs by one
singular regular expression. , Thus, unlike ISBNs, DOIs are difficult to validate properly. Or rather, it is
hard to find sufficiently discriminatory criteria to distinguish proper DOIs from
fake ones. They have no fixed string-length, and very little of character set
restrictions. All we can have is a partial pattern recognition; the more restrictive
the validation rule or regular expression it is based on, the more actually existing
DOIs it will leave out.
Handle: The Handle identifier system, of which DOIs are only a
special case, seems fairly easy and handy at first glance. It comes in two different
flavors. One is the semantically opaque, which has the structure:
Prefix/noid (10079/sqv9sf1), where the NOID-part (for Nice Opaque
Identifier ) is a short alphanumeric string from the
restricted character set "0123456789bcdfghjkmnpqrstvwxz", with random minting
order. The other flavor is the semantically
transparent, which could be of three different types: the URL handle:
Prefix/local-PID(10079/bibid/123456) , the
user handle: Prefix/netid/netid(10079/netid/guoxinji),
which as demonstrated here seems to be less persistent, as people tend to move, and
the simpler group handle: Prefix/group(10079/ISPS). While those of the
second flavor might be more instantly "meaningful", providing context, how are
Handles fairing regading Findability and Accessibility? The findability by googling
will again, as for other PIDs, largely depend on the use and citation rate of items,
while the accessibility again rests largely on the maintenance of the lookup-table
by the custodian. Even so, as just demonstrated, Handles may not always resolve to
the page expected, especially when used as context dependent identifiers of
individuals. In these cases, ORCiD ids should be preferred. What about the
Interoperability and Re-usability of Handles then? Those of the NOID type, with a
restricted character set, will in principle at least be effectively "validatable",
to the extent that the "namespace" or minting agent restricts the string-length, as
e.g. 2077/36687 – Gothenburg
university: 4/5 characters, and 10079/31zcrtn – Yale university: 5/7 characters. Those of the second,
"semantic" flavor will apparently prove less "validatable" in the sense that there
is no longer any fixed string-length or restricted character-set.
UUID: UUIDs v5 are used within the field of biodiversity taxonomy,
as a complement to scientific names. They were
introduced to the field in 2015 by the Global Names Architecture - GNA . The arguments for using them instead of name strings for
certain functions are that they save space as index keys in databases, they have a
fixed string length (36 characters, including the dashes) while scientific names are
of different length. UUIDs do not suffer, as names sometimes do, from encoding
problems that are difficult to detect and they are more easily distinguishable one
from the other than name strings for closely related species variants. Specifically,
it is argued that UUIDs v5 ... can be generated independently by anybody and
still be the same to the same name string... Same ID can be generated in any
popular language following well-defined algorithm.
Note, however, that it is actually the specific name string that is
identified here, not the object, the organism, the 'thing itself'. Thus, the
resulting UUID is completely dependent upon the particular name string (with its
encoding), it cannot be used as a bridge between different name forms for the same
organism, telling us that they are naming the same object. This is due to the fact
that it is generated by hashing a namespace identifier and name. As a result, UUIDs generated in this way by the gna name
resolver, e.g. "707f84e1-e5b8-5063-8256-369ba9d72e13" for Antiaris
toxicaria are next to useless as instruments of Findability, often yielding
0 hits by simple googling, all the while a search on the scientific name
alone will give plenty of precision hits for the sought after organism, providing
rich metadata for the 'thing' itself. Likewise, the same UUID is seldom or never
Accessible, by being resolvable on its own. As an example, consider one of the most
well studied organisms of all, the fruitfly Drosophila melanogaster. Using
the Global Names Resolver to get a UUID v. 5 for
Drosophila melanogaster,
<gni-uuid>1bc2f359-47e4-5da6-a748-74676b7c8c5d</gni-uuid>, googling it
either unprefixed or prefixed gives a zero result (0 recall, 0 precision, date:
2017-01-30). Trying instead the same UUID in a general search of all databases of
NCBI, the US National Center for Biotechnology Information), we get 0 hits
(2018-11-15): 1bc2f359-47e4-5da6-a748-74676b7c8c5d. Most notably, we get 0 hits in the NCBI Taxonomy database, that on the face of it would seem to be the most
relevant to our search.
By contrast, the UUID v5 is eminently "validatable", with a character set restricted
to digits and lower case [a-f], and a fixed string length, 36 characters including
hyphens, in a recognizable, precise pattern: "8-4-4-4-12", allowing for validation
by a regular expression such as
^[a-f\d]{8}-[a-f\d]{4}-[a-f\d]{4}-[a-f\d]{4}-[a-f\d]{12}?$, or by
means of an online
validator. On the other hand, given the fact that these generated UUIDs are
seldom or ever used for citation, and are not "fed back" to the source databases,
where the corresponding scientfic names were found, it is doubtful whether this
"validatability" is also sufficient to make them qualify for
Interoperability and Re-usability. To improve its findability
and re-usability the UUIDs v5 issued by the Global Name Resolver could "ping-back"
and assign themselves to the records in the biodiversity database sources they were
drawn from and further use schema.org markup to get incoming links and a
better ranking by search engines.
Why context?
Generally speaking, although it is preferable that identifiers be findable and
identifiable also in their unprefixed, pure form, typed identifiers give context by
means of namespace prefixes of a metadata standard, a vocabulary or ontology. A
typed identifier "intoduces itself", telling us what kind of identifier it is, and
what type of objects it is used for. Most importantly the namespace tells us what
schema(s), which rules should be used for its validation.
Page claimed that e.g. "dc:title" is adding "unnecessary
complexity (why do we need to know that it's a "dc" title?)" in the JSON
expression:
{ "@context": { "dc:title": "http://purl.org/dc/terms/title" },
"dc:title": "Darwin Core: An Evolving Community-Developed Biodiversity Data
Standard" }
A simple answer is that namespaces are important to retain meaning from context,
serving as a key to interpretation for the future. Self-sustained long-term
preservation should ideally mean in a case like this that the dc specification and
schemas valid at the time be archived together with the records, or at least that there is provenance metadata
including timestamps and namespace of terms used. Metadatafiles in XML usually have
a xsi:schemaLocation indicating which schema to validate against. This information,
together with timestamped metadata elements such as 'dateIssued' should be
sufficient to provide context. For JSON metadata there are name/value pairs such as
{ "protocol": "doi", ... "createTime": "2017-01-12T10:49:03Z", ...}
that could fill the same function. And then, context is just as important for
validation of records also in the present.
A "new" contextual, integrated, validatable PID?
As seen in the case of Handle above, validatability sometimes comes at a cost:
transparency lost. Are we forced to make a choice between the two? Can we create
identifiers that are both fully validatable and at the same time more meaningful,
providing context? So, here we suggest a model for a "new" PID, with a
limited character set, at least for the object id part, defined by namespace
specifications and schemas.
Model: [namespacePrefix].[objectType].[objectId: 10 positions].[issuedDate: YYYY-MM-DD].[registrant: ORCID or ROR-ID]
Example (expression of this paper): fabio.PositionPaper.pp1255qv43.2018-11-12.0000-0001-5699-994X
It is a model of a structured, contextual, modular, validatable identifier. To make
it easier to implement, and more generalizable, there is no character set or
string-length restrictions for the first two modules, except that they should not
contain the dot (.), which is reserved as a module separator. Nevertheless, this
means already existing namespaces and object types could already be used to create a
PID in accordance with this model.
The third module, the objectId (local ID) has a limited character set, selected to
avoid ambiguous interpretations and, to avoid making local uniqueness case-dependent
, restricted to lower case letter characters
and digits. The full stop or dot (.) was chosen as module separator, since it works
well in both xml- and http-environments, without encoding, and is not subject to
confusion as sometimes hyphens and dashes (en-dash and em-dash) can be. It also
works for tokenization of strings. The object type identified in the second module
should belong to the initial namespace prefix. Every namespace can have as many
object types as needed. Namespace schemas could also define valid data
types for their different object types, thus supplying PIDs with data
types, in order to make them even more machine actionable.
The scalability of this model will mainly depend on the 10 character objectId and
the size of the permitted character set. An objectId limited to the proposed
character set [a-kmnp-z0-9] will have 3410 permutations within each
namespace (and possibly objectType), still better than e.g. a 7 character Handle
with NOID.
The objectId module, thus, could be validated separately by a regular expression
restricted to ^[a-kmnp-z0-9]{10}$. It may also be part of a more
comprehensive validation schema, involving a random or pattern based minting
algorithm, preferably including a checksum, with different rules invoked for
different namespace contexts, checking also for example the correspondence between
namespace (module 1) and objectType (module 2) as in this still crude Schematron
schema:
<schema xmlns="http://purl.oclc.org/dsdl/schematron" queryBinding="xslt2">
<ns prefix="rdf" uri="http://www.w3.org/1999/02/22-rdf-syntax-ns#"/>
<ns prefix="fabio" uri="https://w3id.org/spar/fabio"/>
<ns prefix="local" uri="local"/>
<pattern>
<rule id="newPid-rule" context="local:newPid">
<let name="objectType" value="for $i in (.) return tokenize($i,'\.')[2]"/>
<let name="objectId" value="for $i in (.) return tokenize($i,'\.')[3]"/>
<let name="x" value="'https://w3id.org/spar/fabio'"/>
<let name="objectTypeList" value="(for $i in (doc($x)//rdf:type[@rdf:resource='http://www.w3.org/2002/07/owl#Class']/parent::rdf:Description/@rdf:about)
return substring-after($i,'fabio/'))"/>
<let name="objectTypeString" value="string-join($objectTypeList,',')"/>
<assert test="matches($objectId,'^[a-kmnp-z0-9]{10}$')">
An identifier of type 'newPid' must have as its third module a namespace unique objectId of 10 characters from the set [a-kmnp-z0-9].</assert>
<assert test="matches($objectTypeString,$objectType)">The objectType, the second module of the newPid must belong to the namespace of the first module.
</assert>
</rule>
</pattern>
<schema>
One might consider generalizing such a validation schema to the extent possible, so
that the namespace URI in $objectTypeList, from which the $objectType should be
drawn, was automatically construed based on the namespace prefix (module 1) of the
newPid instance to be validated. This could be achieved by having the
namespacePrefix (module 1) expressed as a link with namespace URI, e.g. as fabio in our case above. But that
would also make the validation schema a bit more complicated, notably the
tokenization and separation of modules.
It is also conceivable, to allow for integration of already existing identifier
schemes, that a namespace sets its own character set and string-length restrictions,
to be declared by the validation rules of that namespace. For "narrow" namespaces,
lacking defined different object types, possibly since they comprise basically only
one type of object (as for e.g. ISBNs and ISSNs) we suggest as default second module
value 'NOT' = No Object Type. So we could have an IGSN, International
Geo Sample Number, with string-length of objectID set to 9, expressed in
this model:
The identifier should be fully validatable, as a whole or in part (modules), in the
corresponding namespace(s). The last two modules are optional, but they are meant to
offer built in data provenance. For organisation identifiers, we hope that the
recently launched ROR-IDs will become a global standard, like the ORCID for persons.
Then we could replace the last module in the IGSN-PID above with "05h2dda38".
The resulting PIDs should be minted within the corresponding namespaces, who would
also be the 'custodians' and resolving authorities of their PIDs, responsible for
their uniqueness within their namespace. Another task would be to monitor and assign
sameAs-properties to PIDs that refer to the same 'thing' in other
namespaces.
It has been suggested that in order to build more connected, cross-linked and
digitially accessible Internet content it is necessary to assign
recognizable, persistent, globally unique, stable identifiers to ... data
objects.. The model proposed here aims to
make "new" PIDs fully recognizable, universally unique, stable, but always in a
well-known context, meaningful, and with a good potential for backup.
Conclusions
The purpose of this paper was to analyse some of the more prevalent general PIDs used
in scholarly communication, identify some of their shortcomings and find out how
PIDs could be made more FAIR. Real examples of PIDs were analysed to find out what
additional requirements there might be to make them fully Findable, Accessible,
Interoperable and Re-usable - FAIR. The "novelty" of the paper, if any, is then the
"widening" of the FAIR principles to include also Findability as rate of
distribution or dissemination (e.g. as measured by means of 'googling') and
Interoperability or Re-usability to include also 'validatability'. Further, as
against earlier insistence on the opaqueness of PIDs as a warrant for persistence,
we argued for the importance of adding meaning to PIDs, through
namespace prefixes and object types, as possibly enhancing their future use,
distribution, findability and interpretability and safeguarding against failed
resolvability. The custodianship and minting of PIDs, we suggested to be the
responsibility of namespaces, as these are already assuming
the administration of specifications, validation schemas, vocabularies or
ontologies, and should be well qualified for the task. The minting algorithm, the
patterns for PID-recognition, restriction in character set, string-length (with
possible checksum) of objectId module should all
be part of the validation schemas. These namespaces should then be able to register
their schemes with n2t.net or identifiers.org, as already happens.
And there might be several services such as the SPARQL endpoint of
identifiers.org for registering sameAs-links. To create, maintain and
make our PIDs truly persistent, widely used and FAIR should be a cooperative effort
of the whole scholarly community.
Clark, J. (2016).
PIDvasive:_What's possible when everything has a persistent
identifier? PIDapalooza, November 10, 2016. Retrieved Jan 16, 2017.
https://doi.org/10.6084/m9.figshare.4233839.v1
Coyle, K. et al.(2014). How
Semantic Web differs from traditional data processing. RDF
Validation in the Cultural Heritage Community. International Conference on
Dublin Core and Metadata Applications, Austin, Oct. 2014. Date accessed: 24
Mar. 2017. http://dcevents.dublincore.org/IntConf/dc-2014/paper/view/311
Cruz, M., Kurapati, S., &
Turkyilmaz-van der Velden, Y. (2018). The Role of Data Stewardship in
Software Sustainability and Reproducibility. Zenodo. 2018-09-14. https://doi.org/10.5281/zenodo.1419085
Doorn, P., Dillo, I. (2017).
Assessing the FAIRness of Datasets in Trustworthy Digital Repositories:
A Proposal. IDCC Edinburgh, 22 February 2017. http://www.dcc.ac.uk/webfm_send/2481
Duerr, R.E. et al. (2011).
(2011). On the utility of identification schemes for digital earth
science data: an assessment and recommendations . Earth Science
Informatics 4:139. ISSN: 1865-0473 (Print) 1865-0481 (Online) https://doi.org/10.1007/s12145-011-0083-6
Dunning, A., de Smaele, M., Böhmer,
J. (2017). Are the FAIR Data Principles fair? Practice Paper. 12th
International Digital Curation Conference (IDCC 2017), Edinburgh, Scotland,
20 - 23 February 2017. https://doi.org/10.5281/zenodo.321423
Force11 (2016b). Guiding
Principles for Findable, Accessible, Interoperable and Re-usable Data
Publishing version B1.0.https://www.force11.org/fairprinciples
Force11 (2016). Guiding
Principles for Findable, Accessible, Interoperable and Re-usable Data
Publishing version b1.0 San Diego, CA: FORCE11 https://www.force11.org/node/6062/#Annex6-9
Gertler, A., Bullock, J.
(2017). Reference Rot: An Emerging Threat to Transparency in
Political Science. The Profession. http://doi.org/10.1017/S1049096516002353
Guralnick, R. et al. (2015).
Community Next Steps for Making Globally Unique Identifiers Work for
Biocollections Data. ZooKeys 494: 133–154. https://doi.org/10.3897/zookeys.494.9352
Haak, L. et al. (2012). ORCID:
a system to uniquely identify researchers. Learned Publishing, 25:
259–264. https://doi.org/10.1087/20120404
Hennessey, J., Xijin Ge, S.
(2013). A Cross Disciplinary Study of Link Decay and the
Effectiveness of Mitigation Techniques. Proceedings of the Tenth
Annual MCBIOS Conference. BMC Bioinformatics, 14(Suppl 14):S5. https://doi.org/10.1186/1471-2105-14-S14-S5
Jones, SM., Van de Sompel, H.,
Shankar, H., Klein, M., Tobin, R., Grover, C. (2016). Scholarly Context
Adrift: Three out of Four URI References Lead to Changed Content.
PLoSONE 11(12): e0167475. https://doi.org/10.1371/journal.pone.016747
Klein, M., Van de Sompel, H.,
Sanderson, R., Shankar, H., Balakireva L., Zhou, K., Tobin, R. (2014).
Scholarly Context Not Found: One in Five Articles Suffers from Reference
Rot. PLoS ONE 9(12): e115253. https://doi.org/10.1371/journal.pone.0115253
Li, C.& Sugimoto, S.
(2014). Provenance Description of Metadata using PROV with PREMIS
for Long-term Use of Metadata. Proceedings of the International
Conference on Dublin Core and Metadata Applications (Austin TX, 2014). http://dcpapers.dublincore.org/pubs/article/view/3709
McMurry, JA. et al. (2017).
Identifiers for the 21st century: How to design, provision, and reuse
persistent identifiers to maximize utility and impact of life science
data. PLoSBiol 15(6):e2001414.. http://doi.org/10.1371/journal.pbio.2001414
Page, R. (2016). Towards a
biodiversity knowledge graph. Research Ideas and Outcomes 2: e8767
(07 Apr 2016). http://doi.org/10.3897/rio.2.e8767
Patterson, D. et al. (2016).
Challenges with using names to link digital biodiversity
information. Biodiversity Data Journal 4: e8080 (25 May 2016). https://doi.org/10.3897/BDJ.4.e8080
Philipson, J. (2019). The Red Queen in
the Repository: metadata quality in an ever-changing
environment.IDCC 2019. (In press). https://doi.org/10.5281/zenodo.2276777
Wass, J. (2016). When PIDs
aren't there. Tales from Crossref Event Data. PIDapalooza,
Reykjavik, November 2016. Retrieved: 11:57, Mar 20, 2017 (GMT). https://doi.org/10.6084/m9.figshare.4220580.v1
Wilkinson, M. et al. (2016).
The FAIR Guiding Principles for scientific data management and
stewardship. Scientific Data 3:160018. http://doi.org/10.1038/sdata.2016.18
Wilkinson, M., Schultes, E.,
Bonino, L., Sansone, S., Doorn, P. & Dumontier, M. (2018, July 4).
FAIRMetrics/Metrics: FAIR Metrics, Evaluation results, and initial
release of automated evaluator code. Scientific Data. Zenodo. http://doi.org/zenodo.1305060
Zhou, K. et al. (2015). No More 404s:
Predicting Referenced Link Rot in Scholarly Articles for Pro-Active
Archiving. In: Proceedings of the 15th ACM/IEEE-CE on Joint
Conference on Digital Libraries. JCDL '15, p. 233-236. http://doi.org/10.1145/2756406.2756940
