Having well-defined, portable, and easy to use tools for publishing
vocabularies is an essential aspect of promoting and publishing linked
data. I'd love to pool together tools and get a sense of whether the
existing tools work, if there are gaps that need to be filled, and figure
out what the barriers to publishing vocabularies currently are.
I'll list the frameworks I'm aware of below and would love to hear if any
of you have used these frameworks, whether you think they are helpful, and
if there are any I'm missing. Also, it would be great to hear if you've
stitched something together yourself using more general tools like an RDF
library and your favorite programming language.
The tools below range from complete publishing solutions to simple RDF/OWL
-> HTML renderers.
It probably makes sense to start with this:
- W3C best practices for publishing vocabularies (see httpRange-14, etc...)
- SpecGen (Python)
Currently used to publish FOAF and SIOC (among others). There are many
different forks/versions of SpecGen, here are the ones I've found:
- OntoSpec (SWI-Prolog)
Developed by Yves Raimond to publish the Music and Event ontologies.
- vocab.org Toolchain (XSLT)
Created by Ian Davis for the vocab.org vocabularies (relationship, frbr,
- Neologism (Drupal, PHP)
Developed by DERI and claims to be a more "out-of-the-box" solution. Based
on the Drupal content management framework.
- OWL2XHTML (XSLT)
Developed by Martin Hepp et al. for the GoodRelations ontology.
- OWLDoc (HTML)
A plugin for Protégé, OWLDoc renders documentation for ontologies in HTML.
We have a task group starting up in DCMI to find a possible successor to the
current Vocabulary Management Tool (above) -- something which DCMI could use
both to manage its own vocabulary and to recommend for use by other vocabulary
maintainers. We have posted some basic criteria for such a platform, e.g.,
that it be open source and maintained by an open-source community , and we
have started a wiki page for collecting references such as the ones you list
Aaron, may we add your links to the list? If anyone here would like to
participate in DCMI effort or help edit this or other pages on this topic on the
DCMI wiki, please drop me a line and I'll give you access...
> A plugin for Prot�g�, OWLDoc renders documentation for ontologies in HTML.
> Aaron Rubinstein
> Archivist for Digital Collections
> Digital Collections and Archives
> Tufts University
> v: 617-627-2498
> http://dca.tufts.edu <http://dca.tufts.edu/>
Tom Baker <t...@tombaker.org>
Wonderful - I have added your references to  - many thanks!
I have also added you to http://wiki.dublincore.org as a user - password
sent under separate cover... If anyone else would like wiki access, please
drop me a line!
> I'm particularly interested in this:
> "7. Serves as a DCMI "best-practice exemplar" by being deployable by
> anyone wishing to maintain a vocabulary"
DCMI would like to be in the position of lending momentum to an open-source
community in this area through "leading by example".
> I think the w3c guide is fantastic, though it hasn't been updated
> since 2008. Are these still the w3c recommended approaches?
That's a very good question. In addition to  (which I co-edited in its
first iteration), there are guidelines for Cool URIs , likewise from 2008.
Also from 2008, however, is the RDFa specification. But it would be good to
identify the advantages and disadvantages of publishing vocabularies with
RDF embedded in the HTML. For example, the FOAF specification  gets a 200
response code but includes RDF in the attributes [4, 5]. I have added this
to the wiki page  as well...
 curl --header "Accept: application/rdf+xml" http://xmlns.com/foaf/spec/
Tom Baker <t...@tombaker.org>
First off, thanks for starting this conversation here on the lod-lam
list. Part of my reason for prompting you to post here is that I'll be
co-chairing, along with Gordon Dunsire, the DCMI Task Group Tom
I very much like your list of tools, particularly because they're very
"bare-metal" type solutions, focused on creating simple web-pages for
documentation and on publishing the raw RDF itself. I'd like to have a
closer look at these to see if any offer an intuitve editing
interface, what sort of version control they provide, and what other
sort of "terminology services" are included.
In the DCMI Registries community, for years now, there has been
discussion of Metadata Registries as places for managing, versioning,
editing, translating & disseminating vocabularies.  Some such
systems are also designed to support documenting best-practices for
mixing properties from different vocabularies, DCMI's notion of
Application Profiles. Early examples of these typically used more of a
service model, such as SCHEMAS & CORES [2,3]. Among the earliest of
these that was released as software was the Dublin Core Metadata
Registry, which exists as a service as well as an Open Source
Similarly, the Open Metadata Registry , being used for many RDA &
IFLA Vocabularies, is both an application or a service provided by
Metadata Management Associates, and vocab.org is a hosting service for
vocabularies as well as the tool chain you mention.
Other more service-model oriented appraoches include the JISC
Information Environmnet Metadata Registry project , and the
Meta-Bridge project described in a DC-2011 Paper . Currently, the
interface to the latter is Japanese only, though the report
acknowledges that it needs an English interface for wider adoption.
The paper also does an excellent job of listing out functional
requirements for this *kind* of vocabulary management system.
One significant question for any given use case is whether the focus
is on publishing simple documentation and raw RDF or on providing a
more robust infrastructure for managing, versioning, translating &
collaborative editing vocabularies. Another is whether the prefered
solution be a software application or set of tools to host and manage
locally or a hosted service.
Does anyone mind if I forward this entire thread to the DCMI Task
Group when that work kicks off in a few weeks?