Metadata Matters

Category: Vocabularies

  • Onward!

    One of the things Jon and I still talk about (despite our putative retirement), is how to pass on to the newer generation what we’ve learned over our years working in the vocabulary, data maintenance and data aggregation spaces. Of course, we’ve written or co–written quite a few papers and articles, and most continue to be read and cited (or so ResearchGate tells me), and as well we have significant numbers of presentations and slides available freely.

    Jon sometimes tries to convince me that in the age of AI and waning interest in metadata and vocabularies, particularly in non-library research venues, our time of relevance has passed. He may be right in some respects, but I’m not convinced that’s true in libraries. And the Open Metadata Registry, still alive and quietly kicking, is still being used by an interesting variety of vocabulary nerds. I checked today and noted that there’s a NASA Name Authority vocabulary still in active development as of last week (I haven’t checked to see whether the developers have been active during the partial government shutdown or not).

    Recently, I worked on the NISO effort to describe the state of the vocabulary world and one of our main discoveries was how few of the tools developed over the past few years are still functioning. Most were developed using research funding, notoriously time limited (one or two years is common). What’s left is an array of commercial products (difficult to determine how effective they are, though they’re certainly costly), and the OMR, which we’ve been supporting since our NSF funding dried up in 2008. We had funding from ALA Publishing for some of that time, and we’re still working on a beta version of the OMR.

    We’d like to find out whether there’s any interest out there in figuring out what can be done, in the non-commercial realm, to determine what the needs are–in libraries particularly, but also in other institutions–for continuing a service like the OMR, and what might be strategies to continue it (or something similar) into the next decade or so.

    Ultimately an important question is whether formal RDF vocabulary management matters in a data world dominated by Machine Learning, fragmented by the low bar set for publishing ‘Linked Data’ and the ongoing ‘format wars’ in the library community. Vocabularies matter when definitions are clearly expressed and mechanisms are provided to align heterogeneous metadata across divergent datasets. Definitions and alignments have to be maintained over time — all data has a lifespan of utility. Too often data is released into the wild of the internet without any thought to how it will be consumed and how it will be maintained. Ongoing vocabulary maintenance is crucial to future-proofing public datasets. Or is it all just too hard, given the pressures of funding and the more immediate worry of whether you’ll still have a job next week?

    Given the burgeoning of podcasts on anything and everything, we’ve considered whether that might be a way to get some interesting conversations going (maybe even interviews, which are my favorite kind to listen to). Or we might just up the effort to continue blogging (thus cementing our reputation for persistence!) Do mention some topics that interest you, whether for the blog or (possibly) a podcast. Comments welcome in the comment section of this blog, or on Twitter, at @Metadata_Maven or @jonphipps.

    Discuss!

  • Remembering Christine Frodl

    I can’t remember when I first met Christine, but I certainly knew her well enough to make an informal (and more than a bit subversive) proposal to her at DC-2008 in Berlin. Jon & I had been discussing how RDA might develop translations — this was about the time when we had added the Properties and subProperties, based on the RDA relationships originally published in the Appendices, into the OMR. Many of the readers of this blog may remember that the 2008-2013 versions of the RDA Vocabularies were never formally published by the JSC, but that changed in 2014 when the RDA Registry was unveiled.

    In Berlin we decided to approach Christine about working with us to demonstrate how the translations might work. We knew the German librarians were very interested in participating in RDA and had been pushing at the JSC’s door for some time. We thought she would at least hear us out, but we were in no way convinced that she’d be very interested. We knew that catalog librarians in particular were very imprinted on rules, and we suspected that those from Germany might be even more so. How wrong we were!

    We made sure that Christine knew that we hadn’t asked permission, and there might be repercussions from this subversive activity, even if it were only a demonstration. To our surprise, she said yes, with a visible twinkle in her eye at the thought of such an outrageous project. The repercussions I was worried about stemmed from some discomfort JSC had about us at that point. They didn’t understand what we were trying to do, and someone accused us of stealing their work (I couldn’t make that up).

    So Christine didn’t just say yes, she followed through, appointing Veronika Leibrecht from her shop, who spoke both German and English. Veronika did a fantastic job, and after that we rarely missed a chance to show folks looking at the OMR that there was a German translation. RDA now has (or is working on) translations in the RDA Registry that include Catalan, Chinese, Danish, Dutch, Finnish, French, German, Greek, Hebrew, Hungarian, Italian, Norwegian, Polish, Portuguese, Slovak, Spanish, Swedish, Ukrainian, and Vietnamese.

    This story came to mind immediately when I heard that Christine had died last week. It was clear to us that in addition to the many professional and personal qualities mentioned by those who knew her, she had a vision of a multi-lingual future for RDA that she had the courage to help get moving. That’s a legacy to be celebrated.

  • Vocabulary Instability, or, Junk in Space

    I hope many of you have delved into the NISO Technical Report on Issues in Vocabulary Management, released this past summer. In my opinion, one of the most important insights contained in that report came about as the group reviewed vocabulary management tools and development projects over the past few years, most of them still cited as if they were alive and kicking. But for the most part, they weren’t.

    “Finally, there is the issue of sustainability of the vocabularies selected for use in data. A sustainable vocabulary is protected by organizational or institutional commitments, policies that make clear who makes those commitments and what they mean, as well as a record of responsible maintenance and growth. A vocabulary without those commitments may not be sustainable over time and may be a questionable investment for organizations seeking to use the vocabulary in their data.”

    Why is this important? A key concept in the notion of sustainability of vocabularies is expressed in the use of use of the term ‘investment’. A data provider who chooses to use a particular vocabulary for their data has (whether they recognize it or not) an investment in that vocabulary, thus an important interest in its sustainability. Consider the challenge of keeping those data instances relevant over time in an open world (which librarians have been doing for decades in a very closed world). Doing so requires a combination of sustainably managed vocabularies, and tools to provide change notification and other services to the maintainers of data containing those vocabularies. Inexplicably, there has been very little discussion of this, despite the fact that just about every model for future metadata development presumes the increased use of public vocabularies.

    The NISO group spent some time on these issues:

    “In some respects, the problem of vocabulary discovery and availability for general use is not easily separated from issues around the loss of funding for projects building vocabulary development or management tools, almost all of which were initially developed in time limited circumstances. This report cites several projects that were initially funded in whole or part to address issues around vocabulary provision in particular research or practice communities, but that have not received funding to extend or maintain their tools or their vocabularies. This is a significant problem, as without the ability to support the structural or conceptual vocabularies required to describe resources being aggregated or distributed, there is no such thing as useful distribution or maintenance of any data instances using those vocabularies. The many projects now being funded to consider the problems of ‘big data,’ particularly scientific research data, all depend on stable metadata vocabularies.

    This concern is not that different from what we hear regularly about the crumbling physical infrastructure supporting our transportation systems (although without the scary safety implications). Many on the academic side of these questions take it for granted that funding comes and goes, and may not consider the longer-term implications of these ebbs and flows. But the reason to talk about vocabulary sustainability is that there are long-term implications of depending on funded projects to build and maintain the infrastructure around vocabularies used for linked open data, not to mention the vocabularies themselves.”

    So what’s the solution for this set of problems? Is it to convince funding agencies to include sustainability plans as a requirement for receiving support (and if they’re not going to fund projects for more than two years how effective would that be)? Should we cede the development of vocabularies to institutional entities who can afford to maintain them over time (ex., the Getty)? Would a more stable funding model cover tools as well?

    The current funding models have left a legacy of junk in space–many projects on the net and cited through the literature but no longer functioning, which cannot provide an infrastructure for vocabulary management we can all build upon. I wonder whether the discussion of these issues will become urgent–I hope before I retire from the scene!

  • “Understanding Metadata”: a sort of review

    While checking email during a moment of inattention at the recent ALA Midwinter meeting in Atlanta, I spotted an announcement of the new NISO publication “Understanding Metadata”, by Jenn Riley. It was a surprise, I’d not known that it was in preparation, but I hoped that such a primer would be helpful in raising the level of understanding around metadata in the general library community.

    So I downloaded the publication, and as is my wont, started with the Table of Contents to get a sense of what was being covered. To my absolute astonishment, neither the list of Notable Metadata Languages: Examples in Broad Use nor the one of Notable Metadata Languages: Examples from the Cultural Heritage Sector included RDA: Resource Description and Access. In disbelief, I searched the document for “RDA” and found only a footnote describing RDA as the successor to AACR2 and categorized as a ‘content standard’. I then tried searching for “FRBR”, and there was nothing at all to be found.

    Full disclosure: I’m part of the RDA Development Team and have worked on the RDA Vocabularies since late 2007, after the RDA group and DCMI got together to build them. Granted, it took until January 2014 to publish them formally and bring up the RDA Registry, but we were not hiding under a rock for those years.

    There’s a long trail of articles and blog posts documenting the trajectory of RDA (only some of them with my name on them), so it’s inexplicable to me how the author of this primer, the estimable Jenn Riley, could still subscribe to the notion that RDA is all about instructions for populating element sets. That was certainly true before 2008, but I know Jenn has not been hiding under a rock, either. Particularly when efforts outside North America are considered, RDA is a force to be reckoned with in the metadata space, and interest in Europe and other places outside of North America is enthusiastic and growing.

    None of the IFLA standards show up in this publication either, which suggests a somewhat US-centric view–inexplicable to me given the increasing internationalization of metadata efforts overall. Several of those missing IFLA standards, including FRBR, are being republished after extensive reviews, and at least in RDA-land, the IFLA Library Reference Model (LRM) is a significant topic of conversation where library metadata people hang out.

    Then I checked out Jenn’s wonderful image categorizing metadata standards “Seeing Standards: A Visualization of the Metadata Universe”, and noted that it also included RDA as a content standard and successor to AACR2. Then I saw that the image was copyrighted 2009-2010, and the most obvious answer for the gap became clearer. [I should note that I have a large poster sized version of the image, which I had printed up so I could study the image better, back when I was doing more teaching]. I was pretty relieved by this realization, to be honest, but it occurred to me that the revolutionary change represented by the RDA Vocabularies seems to have been overlooked by Jenn (and probably others) who think they know what RDA is, but haven’t looked at it lately (or at least since 2008, when the vocabularies began to appear in the Open Metadata Registry). I guess as one involved in metadata standards for well over 3 decades, it’s been my experience that with few exceptions, metadata standards change continually, and the assumption that they are fixed in time and space is a dangerous one.

    That said, there is much to be applauded in this primer. The explanation of storage and sharing is an excellent overview, clear and well illustrated. It was good to see a section covering standardization issues, but I would argue with this statement in the ‘Standardization’ section: “Formal standardization is not the norm for most RDF vocabularies; instead, communities tend to build vocabularies that are useful to them, then promote their use through a combination of documentation and open sharing of data that uses these vocabularies.” This statement absolutely does not apply to RDA, nor to the majority of the examples listed in Notable Metadata Languages: Examples in Broad Use. But in a sense it explains the emphasis on XML that seemed a bit out of date as I first read through the document

    I would also argue with the following statement in the Controlled Vocabularies section: “Those developed and maintained by the cultural heritage community, such as LCSH, tend to be among the most robust controlled vocabularies in common use.” I’m not sure what definition of ‘robust’ is being used here, but I posted on Metadata Matters in 2013 about the significant problems with LCSH in http://id.loc.gov, which i think makes a strong argument against the ‘robustness’ of the LCSH services. (https://managemetadata.com/blog/2013/07/23/versions-and-services-pt-2/). Not much has changed since that post, sadly.

    I guess, on the whole, I’d have to say I’m disappointed with this primer, despite some strong sections. Although I’m no fan of peer review, it strikes me that this is one situation where it might have been useful.

  • The Conundrum of Research Funding

    I think I mentioned somewhere that one of my current projects is a part of the NISO Bibliographic Roadmap effort to develop vocabulary best practices in the areas of use/reuse, documentation and preservation. The project itself builds on a full-day vocabulary workshop brought together at the DCMI conference in 2011.

    I share co-chair duties on the Use/Reuse subgroup with Daniel Lovins of NYU, but as we’ve all moved along on this project, the overlaps between the three subgroups tasked with writing the best practices documentation (the other groups are tasked specifically with documentation and preservation) becomes more and more apparent. We’ve determined that our final recommendations will be made together in one document, rather than spend our time trying to untie the Gordian Knot that is vocabulary issues in toto.

    At some point I fielded a question from one of the group members about the difference between vocabulary maintenance and vocabulary preservation. It was a great question, and one that I hadn’t really thought much about. To some extent each topic has a basic assumption: for a vocabulary being actively maintained, preservation isn’t much of an issue. But when, for whatever reason, the vocabulary ceases being actively maintained, preservation becomes the issue. One question the Roadmap project is trying to address is how to tell the difference, given the dearth of data generally provided about the who, how and when of ownership and management of vocabularies, much less what policies and practices are being followed. There’s a whole range of issues around preservation: how it should be done and by whom, and what kind of context to wrap around the supine resource.

    In some respects, the problem of vocabulary preservation is not easily separated from issues around the loss of funding for projects building vocabularies as well as vocabulary development or management tools, almost all of which were initially developed by funded projects. This suggests we have a preservation problem built on a sustainability problem. The report draft as it stands cites several projects which were initially funded in whole or part to address issues around vocabulary provision, in particular research or practice communities, but have not received funding to build out or maintain their tools or their resources. This is a significant issue, given that without the ability to maintain the infrastructure supporting the structural or conceptual vocabularies required to describe resources being aggregated or distributed, there is no such thing as proper distribution or maintenance of any data resources. Consider how many projects are now being funded to consider the problems of ‘big data’, particularly scientific research data, all of whose solutions depend to some extent on metadata vocabularies.

    This concern is not that different than what we hear constantly about the physical infrastructure supporting our transportation systems, crumbling in ways not dissimilar to our data distribution systems (although without the scary safety implications). Many of us on the academic side of these questions take it for granted that funding comes and goes, and consider not the longer-term implications of these ebbs and flows and how funding agency priorities (generally focused on innovation rather than maintenance) affect the vocabulary environment. But it remains clear that a big reason we talk about vocabulary preservation is because there are long-term implications of depending on funded projects to build and maintain the infrastructure around vocabularies as well as the vocabularies themselves.

    I speak from experience on this issue. The Open Metadata Registry (OMR) was built on funding from the US National Science Foundation (NSF), and when its funding ended in 2007 should have died the death of most such project based tools. That it didn’t is due almost entirely to the fact that it was built by two very stubborn people who were pretty hooked by the usefulness of vocabularies. For several years we searched in vain for additional funding, but at that time funding was scarce, and what little there was went to much sexier undertakings. For much of that time the OMR survived on a ‘too cheap to fail’ strategy, where early grant funding had paid ahead for long-term costs of server space and basic technical upkeep.

    As far as I know, the OMR is one of the only free general-purpose vocabulary development and maintenance tools that has survived past its initial funding. At this point a large percentage of the library worlds’ vocabularies are maintained using the OMR, and although it is built to enable all of them to migrate at any time to another repository or tool, there is not currently much of anything available for them to migrate to. This suggests that sustainable funding is a far more critical issue than we’ve yet considered.

    At present the OMR developers are setting up a proposal for funding a sustainability plan for the OMR, to spend the next 2-3 years bringing it up-to-date as a viable, community-driven Free Open Source Software project with long-term institutional support for its infrastructure, so it can be maintained after our retirement. If the plan works, the OMR will remain a place where vocabularies will be safe, secure, and maintainable, with room for maps and application profiles using those vocabularies. Other commercial and research-based options are still there, but not necessarily friendly to limited budgets or organizations with limited technical support.

    During the later days of the National Science Digital Library (NSDL)–under whose aegis the OMR was developed–the NSF began to include requirements for sustainability into their grant process. I don’t know how many other granting agencies mandate those kinds of requirements, but judged by huge amounts of the abandoned detritus of decades of research funding, certainly more should.

  • Why Do We Undervalue Maintenance?

    I ran across a really interesting article last week, and the points it makes have been rocketing around my head as I consider what’s broken (and why) in the small world many of us live in, not to mention how we can fix those broken things. I’d really recommend taking a look at it: “Hail the maintainers: Capitalism excels at innovation but is failing at maintenance, and for most lives it is maintenance that matters more”, by Lee Vinsel & Andrew Russell.

    The gist of the authors’ point–that we overvalue innovation and undervalue the maintenance of our existing technologies–is one that ought to resonate particularly with librarians, especially those of us who have toiled in the world of catalogs and the maintenance of same.

    “The most unappreciated and undervalued forms of technological labour are also the most ordinary: those who repair and maintain technologies that already exist, that were ‘innovated’ long ago. This shift in emphasis involves focusing on the constant processes of entropy and un-doing – which the media scholar Steven Jackson calls ‘broken world thinking’ – and the work we do to slow or halt them, rather than on the introduction of novel things.”

    One of the things that struck me in this article was the connection the authors made between ‘innovation’–associated with high tech activities primarily seen as the male domain, and ‘maintenance’, the ‘women’s work’ of the age of technology.

    “We can think of labour that goes into maintenance and repair as the work of the maintainers, those individuals whose work keeps ordinary existence going rather than introducing novel things. Brief reflection demonstrates that the vast majority of human labour, from laundry and trash removal to janitorial work and food preparation, is of this type: upkeep. This realisation has significant implications for gender relations in and around technology. Feminist theorists have long argued that obsessions with technological novelty obscures all of the labour, including housework, that women, disproportionately, do to keep life on track.”

    The authors make the point that these skewed values resonate throughout our society. Notice how boring most people consider the maintenance of our physical infrastructure–roads, bridges, pipelines, etc. Of course, most of us agree that there needs to be more of it, and more resources aimed in that direction, but ho hum, talk about it in any detail? Not on your life.

    For those of us considering the data infrastructure to support the rebuilding and expansion of the data sharing that libraries have done for many decades, we see the same thing. The Open Metadata Registry (OMR to those of us who prefer to speak in acronyms), now too often dismissed as ‘old technology’, is a case in point. Back in 2004 when we received funding from the National Science Foundation, there weren’t many people thinking about data ‘infrastructure’–particularly for vocabulary data, but we were. Back in the day, we often talked about the OMR in the context of the plumbing beneath sidewalks and buildings. Few of us who are not plumbers think about plumbing — we all assume that it’s there, but is rarely a topic in non-plumber gatherings until something critical breaks, or a community is systematically poisoned.

    And even plumbers innovate, in ways that few of us understand, much less think about. The water debacle in Flint, Mich. has brought up the problems of old pipes, but not many discussions go beyond that. In the world of vocabulary management, the OMR has been building and maintaining vocabularies for well over a decade now, and we know a lot about it. In particular we know where improvements should be made, and how OMR services can best support the activities needed by the folks building instance data. We’re also thinking about how to better sustain the effort over time.

    But the OMR was built to work within an environment where there are many such services, supporting other communities looking at the same changes ahead that we see in the library data environment. For lots of reasons, that array of services hasn’t come about, though we see activity in the structural metadata world–new formats, old formats morphing, etc.–effort almost entirely devoted to semantic churning, not services. Much effort is being put into creating and publishing Linked Open Data, but very little consideration is being given to how that data will need to change over time and how the entire plumbing infrastructure of publishing and consuming that data must support the need for ongoing maintenance of both the data and the public and private vocabularies that describe that data.

    [This is the first of two posts, the second proposing some ways to move forward]

  • Denying the Non-English Speaking World

    Not long ago I encountered the analysis of BibFrame published by Rob Sanderson with contributions by a group of well-known librarians. It’s a pretty impressive document–well organized and clearly referenced. But in fact there’s also a significant amount of personal opinion in it, the nature of which is somewhat masked by the references to others holding the same opinion.

    I have a real concern about some of those points where an assertion of ‘best practices’ are particularly arguable. The one that sticks in my craw particularly shows up in section 2.2.5:

    2.2.5 Use Natural Keys in URIs
    References: [manning], [ldbook], [gld-bp], [cooluris]

    Although the client must treat URIs as opaque strings, it is good practice to construct URIs in a systematic and human readable fashion for both instances and ontology terms. A natural key is one that appears in the information about the resource, such as some unique identifier for the resource, or the label of the property for ontology terms. While the machine does not care about structure, memorability or readability of URIs, the developers that write the code do. Completely random URIs introduce difficult to detect semantic and algorithmic errors in both publication and consumption of the data.

    Analysis:

    The use of natural keys is a strength of BIBFRAME, compared to similarly scoped efforts in similar communities such as the RDA and CIDOC-CRM vocabularies which use completely opaque numbers such as P10001 (hasRespondent) or E33 (Linguistic Entity). RDA further misses the target in this area by going on to define multiple URIs for each term with language tagged labels in the URI, such as rda:hasRespondent.en mapping to P10001. This is a different predicate from the numerical version, and using owl:sameAs to connect the two just makes everyone’s lives more difficult unnecessarily. In general, labels for the predicates and classes should be provided in the ontology document, along with thorough and understandable descriptions in multiple languages, not in the URI structure.

    This sounds fine so long as you accept the idea that ‘natural’ means English, because, of course, all developers, no matter their first language, must be fluent enough in English to work with English-only standards and applications. This mis-use of ‘natural’ reminds me of other problematic usages, such as the former practice in the adoption community (of which I have been a part for 40 years) where ‘natural’ was routinely used to refer to birth parents, thus relegating adoptive parents to the ‘un-natural’ realm. So in this case, if ‘natural’ means English, are all other languages inherently un-natural in the world of development? The library world has been dominated by the ‘Anglo-American’ notions of standard practice for a very long time, and happily, RDA is leading away from that, both in governance and in development of vocabularies and tools.

    The Multilingual strategy adopted by RDA is based on the following points:

    1. More than a decade of managing vocabularies has convinced us that opaque identifiers are extremely valuable for managing URIs, because they need not be changed as labels change (only as definitions change). The kinds of ‘churn’ we saw in the original version of RDA (2008-2013) convinced us that label-based URIs were a significant problem (and cost) that became worse as the vocabularies grew over time.
    2. We get the argument that opaque URIs are often difficult for humans to use–but the tools we’re building (the RDA Registry as case in point) are intended to give human developers what they want for their tasks (human readable URIs, in a variety of languages) but ensure that the URIs for properties and values are set up based on what machines need. In this way, changes in the lexical URIs (human-readable) can be maintained properly without costly change in the canonical URIs that travel with the data content itself.
    3. The multiple language translations (and distributed translation management by language communities) also enable humans to build discovery and display mechanisms for users that are speakers of a variety of languages. This has been a particularly important value for national libraries outside the US, but also potentially for libraries in the US meeting the needs of non-English language communities closer to home.

    It’s too easy for the English-first library development community to insist that URIs be readable in English and to turn a blind eye to the degree that this imposes understanding of the English language and Anglo-American library culture on the rest of the world. This is not automatically the intellectual gift that the distributors of that culture assume it to be. It shouldn’t be necessary for non-Anglo-American catalogers to learn and understand Anglo-American language and culture in order to express metadata for a non-Anglo audience. This is the rough equivalent of the Philadelphia cheese steak vendor who put up a sign reading “This is America. When ordering speak in English”.

    We understand that for English-speaking developers http://bibframe.org/vocab/title is initially easier to use than http://rdaregistry.info/Elements/w/P10088 or even (heaven forefend!) “130_0#$a” (in RDF: http://marc21rdf.info/elements/1XX/M1300_a). That’s why RDA provides http://rdaregistry.info/Elements/w/titleOfTheWork.en but also, eventually, http://rdaregistry.info/Elements/w/拥有该作品的标题.ch and http://rdaregistry.info/Elements/w/tieneTítuloDeLaObra.es, et al (you do understand Latin of course). These ‘unnatural’ Lexical Aliases will be provided by the ‘native’ language speakers of their respective national library communities.

    As one of the many thousands of librarians who ‘speak’ MARC to one another–despite our language differences–I am loathe to give up that international language to an English-only world. That seems like a step backwards.

  • Review of: DRAFT Principles for Evaluating Metadata Standards

    Metadata standards is a huge topic and evaluation a difficult task, one I’ve been involved in for quite a while. So I was pretty excited when I saw the link for “DRAFT Principles for Evaluating Metadata Standards”, but after reading it? Not so much. If we’re talking about “principles” in the sense of ‘stating-the-obvious-as-a-first-step’, well, okay—but I’m still not very excited. I do note that the earlier version link uses the title ‘draft checklist’, and I certainly think that’s a bit more real than ‘draft principles’ for this effort. But even taken as a draft, the text manages to use lots of terms without defining them—not a good thing in an environment where semantics is so important. Let’s start with a review of the document itself, then maybe I can suggest some alternative paths forward.

    First off, I have a problem with the preamble: “These principles are intended for use by libraries, archives and museum (LAM) communities for the development, maintenance, governance, selection, use and assessment of metadata standards. They apply to metadata structures (field lists, property definitions, etc.), but can also be used with content standards and value vocabularies”. Those tasks (“development, maintenance, governance, selection, use and assessment” are pretty all encompassing, but yet the connection between those tasks and the overall “evaluation” is unclear. And, of course, without definitions, it’s difficult to understand how ‘evaluation’ relates to ‘assessment’ in this context—are they they same thing?

    Moving on to the second part about what kind of metadata standards that might be evaluated, we have a very general term, ‘metadata structures’, with what look to be examples of such structures (field lists, property definitions, etc.). Some would argue (including me) that a field list is not a structure without a notion of connections between the fields; and although property definitions may be part of a ‘structure’ (as I understand it, at least), they are not a structure, per se. And what is meant by the term ‘content standards’, and how is that different from ‘metadata structures’? The term ’value vocabularies’ goes by many names, and is not something that can go without a definition. I say this as an author/co-author of a lot of papers that use this term, and we always define it within the context of the paper for just that reason.

    There are many more places in the text where fuzziness in terminology is a problem (maybe not a problem for a checklist, but certainly for principles). Some examples:

    1. What is meant by ’network’? There are many different kinds, and if you mean to refer to the Internet, for goodness sakes say so. ‘Things’ rather than ‘strings’ is good, but it will take a while to make it happen in legacy data, which we’ll be dealing with for some time, most likely forever. Prospectively created data is a bit easier, but still not a cakewalk — if the ‘network’ is the global Internet, then “leveraging ‘by-reference’ models” present yet-to-be-solved problems of network latency, caching, provenance, security, persistence, and most importantly: stability. Metadata models for both properties and controlled values are an essential part of LAM systems and simply saying that metadata is “most efficient when connected with the broader network” doesn’t necessarily make it so.

    2. ‘Open’ can mean many things. Are we talking specific kinds of licenses, or the lack of a license? What kind of re-use are you talking about? Extension? Wholesale adoption with namespace substitution? How does semantic mapping fit into this? (In lieu of a definition, see the paper at (1) below)

    3. This principle seems to imply that “metadata creation” is the sole province of human practitioners and seriously muddies the meaning of the word creation by drawing a distinction between passive system-created metadata and human-created metadata. Metadata is metadata and standards apply regardless. What do you mean by ‘benefit user communities’? Whose communities? Please define what is meant by ‘value’ in this context? How would metadata practitioners ‘dictate the level of description provided based on the situation at hand’?

    4. As an evaluative ‘principle’ this seems overly vague. How would you evaluate a metadata standard’s ability to ‘easily’ support ‘emerging’ research? What is meant by ‘exchange/access methods’ and what do they have to do with metadata standards for new kinds of research?

    5. I agree totally with the sentence “Metadata standards are only as valuable and current as their communities of practice,” but the one following makes little sense to me. “ … metadata in LAM institutions have been very stable over the last 40 years …” Really? It could easily be argued that the reason for that perceived stability is the continual inability of implementers to “be a driving force for change” within a governance model that has at the same time been resistant to change. The existence of the DCMI usage board, MARBI, the various boards advising the RDA Steering Committee, all speak to the involvement of ‘implementers’. Yet there’s an implication in this ‘principle’ that stability is liable to no longer be the case and that implementers ‘driving’ will somehow make that inevitable lack of stability palatable. I would submit that stability of the standard should be the guiding principle rather than the democracy of its governance.

    6. “Extensible, embeddable, and interoperable” sounds good, but each is more complex than this triumvirate seems. Interoperability in particular is something that we should all keep in mind, but although admirable, interoperability rarely succeeds in practice because of the practical incompatibility of different models. DC, MARC21, BibFrame, RDA, and Schema.org are examples of this — despite their ‘modularity’ they generally can’t simply be used as ‘modules’ because of differences in the thinking behind the model and their respective audiences.

    I would also argue that ‘lite style implementations’ make sense only if ‘lite’ means a dumbed-down core that can be mapped to by more detailed metadata. But stressing the ‘lite implementations’ as a specified part of an overall standard gives too much power to the creator of the standard, rather than the creator of the data. Instead we should encourage the use of application profiles, so that the particular choices and usages of the creating entity are well documented, and others can use the data in full or in part according to their needs. I predict that lossy data transfer will be less acceptable in the reality than it is in the abstract, and would suggest that dumb data is more expensive over the longer term (and certainly doesn’t support ‘new research methods’ at all). “Incorporation into local systems” really can only be accomplished by building local systems that adhere to their own local metadata model and are able to map that model in/out to more global models. Extensible and embeddable are very different from interoperable in that context.

    7. The last section, after the inarguable first sentence, describes what the DCMI ‘dumb-down’ principle defined nearly twenty years ago, and that strategy still makes sense in a lot of situations. But ‘graceful degradation’ and ‘supporting new and unexpected uses’ requires smart data to start with. ‘Lite’ implementation choices (as in #6 above) preclude either of those options, IMO, and ‘adding value’ of any kind (much less by using ‘ontological inferencing’) is in no way easily achievable.

    I intend to be present at the session in Boston [9:00-10:00 Boston Conference and Exhibition Center, 107AB] and since I’ve asked most of my questions here I intend not to talk much. Let’s see how successful I can be at that!

    It may well be that a document this short and generalized isn’t yet ready to be a useful tool for metadata practitioners (especially without definitions!). That doesn’t mean that the topics that it’s trying to address aren’t important, just that the comprehensive goals in the preamble are not yet being met in this document.

    There are efforts going on in other arenas–the NISO Bibliography Roadmap work, for instance, that should have an important impact on many of these issues, which suggests that it might be wise for the Committee to pause and take another look around. Maybe a good glossary would be a important step?

    Dunsire, Gordon, et al. “A Reconsideration of Mapping in a Semantic World”, paper presented at International Conference on Dublin Core and Metadata Applications, The Hague, 2011. Available at: http://dcpapers.dublincore.org/pubs/article/view/3622/1848

  • Separating ideology, politics and utility

    Those of you who pay attention to politics (no matter where you are) are very likely to be shaking your head over candidates, results or policy. It’s a never ending source of frustration and/or entertainment here in the U.S., and I’ve noticed that the commentators seem to be focusing in on issues of ideology and faith, particularly where it bumps up against politics. The visit of Pope Francis seemed to be taking everyone’s attention while he was here, but though this event has added some ‘green’ to the discussion, it hasn’t pushed much off the political plate.

    Politics and faith bump up against each other in the metadata world, too. What with traditionalists still thinking in MARC tags and AACR2, to the technical types rolling their eyes at any mention of MARC and trying to push the conversation towards RDA, RDF, BibFrame, schema.org, etc., there are plenty of metadata politics available to flavor the discussion.

    The good news for us is that the conflicts and differences we confront in the metadata world are much more amenable to useful solution than the politics crowding our news feeds. I remember well the days when the choice of metadata schema was critical to projects and libraries. Unfortunately, we’re all still behaving as if the proliferation of ‘new’ schemas makes the whole business more complicated–that’s because we’re still thinking we need to choose one or another, ignoring the commonality at the core of the new metadata effort.

    But times have changed, and we don’t all need to use the same schema to be interoperable (just like we don’t all need to speak English or Esperanto to communicate). But what we do need to think about is what the needs of our organization are at all stages of the workflow: from creating, publishing, consuming, through integrating our metadata to make it useful in the various efforts in which we engage.

    One thing we do need to consider as we talk about creating new metadata is whether it will need to work with other data that already exists in our institution. If MARC is what we have, then one requirement may be to be able to maintain the level of richness we’ve built up in the past and still move that rich data forward with us. This suggests to me that RDA, which RIMMF has demonstrated can be losslessly mapped to and from MARC, might be the best choice for the creation of new metadata.

    Back in the day, when Dublin Core was the shiny new thing, the notion of ‘dumb-down’ was hatched, and though not an elegantly named principle, it still works. It says that rich metadata can be mapped fairly easily into a less-rich schema (‘dumbed down’), but once transformed in a lossy way, it can’t easily be ‘smartened up’. But in a world of many publishers of linked data, and many consumers of that data, the notion of transforming rich metadata into any number of other schemas and letting the consumer chose what they want, is fairly straightforward, and does not require firm knowledge (or correct assumptions) of what the consumers actually need. This is a strategy well-tested with OAI-PMH which established a floor of Simple Dublin Core but encouraged the provision of any number of other formats as well, including MARC.

    As consumers, libraries and other cultural institutions are also better served by choices. Depending on the services they’re trying to support, they can choose what flavor of data meets their needs best, instead of being offered only what the provider assumes they want. This strategy leaves open the possibility of serving MARC as one of the choices, allowing those institutions still nursing an aged ILS to continue to participate.

    Of course, the consumers of data need to think about how they aggregate and integrate the data they consume, how to improve that data, and how to make their data services coherent. That’s the part of the new create, publish, consume, integrate cycle that scares many librarians, but it shouldn’t–really!

    So, it’s not about choosing the ‘right’ metadata format, it’s about having a fuller and more expansive notion about sharing data and learning some new skills. Let’s kiss the politics goodbye, and get on with it.

  • Semantic Versioning and Vocabularies

    A decade ago, when the Open Metadata Registry (OMR) was just being developed as the NSDL Registry, the vocabulary world was a very different place than it is today. At that point we were tightly focussed on SKOS (not fully cooked at that point, but Jon was on the WG that was developing it, so we felt pretty secure diving in).

    But we were thinking about versioning in the Open World of RDF even then. The NSDL Registry kept careful track of all changes to a vocabulary (who, what, when) and the only way to get data in was through the user interface. We ran an early experiment in making versions based on dynamic, timestamp-based snapshots (we called them ‘time slices’, Git calls them ‘commit snapshots’) available for value vocabularies, but this failed to gain any traction. This seemed to be partly because, well, it was a decade ago for one, and while it attempted to solve an Open World problem with versioned URIs, it created a new set of problems for Closed World experimenters. Ultimately, we left the versions issue to sit and stew for a bit (6 years!).

    All that started to change in 2008 as we started working with RDA, and needed to move past value vocabularies into properties and classes, and beyond that into issues around uploading data into the OMR. Lately, Git and GitHub have started taking off and provide a way for us to make some important jumps in functionality that have culminated in the OMR/GitHub-based RDA Registry. Sounds easy and intuitive now, but it sure wasn’t at the time, and what most people don’t know is that the OMR is still where RDA/RDF data originates — it wasn’t supplanted by Git/Github, but is chugging along in the background. The OMR’s RDF CMS is still visible and usable by all, but folks managing larger vocabularies now have more options.

    One important aspect of the use of Git and GitHub was the ability to rethink versioning.

    Just about a year ago our paper on this topic (Versioning Vocabularies in a Linked Data World, by Diane Hillmann, Gordon Dunsire and Jon Phipps) was presented to the IFLA Satellite meeting in Paris. We used as our model the way software on our various devices and systems is updated–more and more these changes happen without much (if any) interaction with us.

    In the world of vocabularies defining the properties and values in linked data, most updating is still very manual (if done at all), and the important information about what has changed and when is often hidden behind web pages or downloadable files that provide no machine-understandable connections identifying changes. And just solving the change management issue does little to solve the inevitable ‘vocabulary rot’ that can make published ‘linked data’ less and less meaningful, accurate, and useful over time.

    Building stable change management practices is a very critical missing piece of the linked data publishing puzzle. The problem will grow exponentially as language versions and inter-vocabulary mappings start to show up as well — and it won’t be too long before that happens.

    Please take a look at the paper and join in the conversation!