Dear all,
It’s been a while since the last release announcement :-)
Version 15.1 of VocBench and 6.1 of ShowVoc have been released today! The reason for this very long wait is actually a mix of different aspects:
- a large number of new features contributed by different funders
- a major reengineering work due to:
- the introduction of a new version of the dataset metadata vocabulary: https://w3id.org/mdr
- the eponymous metadata component (the MDR) now centralizing management of all dataset, versions and distributions
We hope the wait will be compensated by the large number of improvements and new features that have been brough to the platforms: you can check the changelogs for both version 15.0 (which has not been released) and version 15.1 of VocBech, and 6.0 and 6.1 of ShowVoc below in this email.
So, here we are, with the usual info:
Versions 15.1 of VocBench and 6.1 of ShowVoc are available at the following pages:
VocBench: http://vocbench.uniroma2.it/downloads/
ShowVoc: http://showvoc.uniroma2.it/downloads/
GraphDB and RDF4J compliancy:
There are some important news here:
The RDF4J client (and its embedded server, when using local repositories) installed in Semantic Turkey, the backend of VocBench, is 5.1.4. If you are using an external RDF4J triple store, you should update it to this version or to a later one.
About GraphDB, the officially supported version would be 11.x, however, there’s been a recent change in the licensing, that coincides with versions >= 11.0, restricting free use of the platform.
To this end, we manage to maintain compatibility with both versions 10.x and 11.x but there are different instructions to follow.
The instructions are
https://vocbench.uniroma2.it/doc/sys/#separate_triple_store
contains and summarizes all of the above information.
For the future, we want to assure all users using having different needs, that none of them will be let down
- Compliancy with GraphDB 10.x will be kept for a time as long as possible, thus avoiding users previously adopting GraphDB-free to be abruptly forced to move to a licensed version or to another technology
- In the meanwhile, we will strengthen support for RDF4J free repositories and improve performances when adopting them (e.g. by adopting Lucene indexes for text searches, checking compliancy with other free storage solutions etc..) thus guaranteeing compliancy with state-of-the-art free and open-source solutions.
- Users relying on licensed versions of GraphDB can also be assured: GraphDB will remain fully supported (and new features will be added as they come) in the future!
- Also, for those planning to move to a licensed version, remember that GraphWise applies specific discounts for GDB licenses used in combo with VocBench or ShowVoc. Please drop us an email if you need more information.
Release notes
Release notes are reported below for the two applications, for their common service backend and for its main related applications (only for those that changed)
VocBench
15.0.0 (22-10-2025)
- Introduced a new management system for configuring remote triple store instances
- reorganization of the UI for the various improvements following MDR 3.0 (see ST's changelog)
- UI changes following introduced support for oAuth
- Revamped project creation form
- Introduced templates for project creation
- Enabled sorting of alignments (by left/right entity or measure) in Alignment Validation tool
- Accepted IRI without angle brackets <> in the add value manually dialog (Thanks to Saku Sepp?l? for the contribution)
- Support for configuring connection to Diffing tools and for hosting ChangeSet files
- Support for querying the support repository (requires appropriate capability)
- Disabled several tools in Alignment type projects
- Enabled search syntax when backend repository technology is GraphDB
- Option for changing the suggested baseuri when switching the working graph
- Option for spporting oAuth in the ShowVoc deployer
- Introduced new DocTagging project model for interacting with SELEN
- Added users' last session start time to the project overview table
- Fixed wrong rendering resolution of entities in Alignment Validation tool
- Minor bugfixes
15.1.0 (12-03-2026)
- Improved MDR and adapted to the final version of the vocabulary
- Introduced registration of a new/binding to an existing mdr:DatasetArchetype (simply referred as Dataset in the UI) when creating a project
- Introduced Machine as an alternative authenticated actor (available only with OAuth2)
- Added support for multiple versions of a Dataset within a project and for binding more projects under a same Dataset
- Introduced support for new configuration options (see Semantic Turkey changelog) for loaders/deployers in the versioning page:
- admins and PMs prepare a base configuration for deployers that can be stored at system level or at project level
- each archived distribution stored through a given deployer can be automatically restored by exploiting the respective inverse loader
- fields declared as parameters can be omitted by admins and PMs; they will be later requested to the users deploying a new distribution
- the version placeholder can be used in fields to be automatically expanded with the version of the dataset being deployed
- Automatic access to projects upon their creation
- Made profiling optional when using data preload in project creation
- Update suffix of core repository name in project creation form
- Implemented lazy loading for structure tabs: each tab is now initialized only upon first activation
- Added reset to History and Validation filters
- Added support for SPARQL querying MDR repository (only available for administrators when "Experimental features" option is enabled)
- Moved persistency of "show deprecated" setting server side
- Dynamically initialized the textarea height based on its content when inline-editing a literal value in ResourceView
- Supported sticky header in several tables
- Fixed an issue preventing listing instances of classes when browsing the class tree in the context of property values
- Fixed an issue preventing editing of project facets
- Fixed a bug causing loading from Spreadsheet files to fail
- Fixed an issue preventing details of a selected element in UML-like diagrams from being displayed
- Fixed a bug that caused the language value filter to not work with SKOS lexicalizations
- Fixed a bug in ResourceView template editor which affected "Include in/Remove from all resources" quick actions
- Fixed infinite loop when any of custom broader or narrower properties is no more available
- Minor bugfixes
- Updated several dependencies addressing vulnerabilities
ShowVoc
6.0.0 (22-10-2025)
- Introduced support for Dataset Catalogs, with functionalities for:
- automatic discovery of new versions of an acquired dataset
- automatic discovery of new datasets in a "watched" catalog
- acquisition of metadata from a dataset acquired through a catalog
- logging of provenance information for datasets/versions of datasets acquired through a catalog
- support for Annotation of Digital Objects (see ST changelog)
- showing existing annotations
- showing statistics about existing annotations
- showing existing annotated objects
- Introduced a new management system for configuring remote triple store instances
- Overall reorganization of the UI for the several improvements following MDR 3.0
- see ST's changelog for the underlying changes in understandign how these affected SV's UI
- complete revision of the download section, now featuring a per-version list of downloads
- downloads are now available through permanent URIs and not service invocations
- structured support for dataset, linkset, changeset, generic downloads
- pervasive support in the UI for versions and distributions explciitly acknowledged in the MDR
- partially editable metadata entries in the dataset catalog
- Revamped project creation form
- Introduced templates for project creation
- Enabled search syntax when backend repository technology is GraphDB
- Converted most of the components as standalone
- Fixed wrong order of SKOS lexicalizations
6.1.0 (12-03-2026)
- Improved MDR and adapted to the final version of the vocabulary
- Introduced registration of a new/binding to an existing mdr:DatasetArchetype (simply referred as Dataset in the UI) when creating a project
- Added support for multiple versions of a Dataset within a project and for binding more projects under a same Dataset
- Added the possibility to view and select the version involved in several project operations, such as indexing, loading, and deleting data
- Introduced Machine as an alternative authenticated actor (available only with OAuth2)
- Implemented lazy loading for structure tabs: each tab is now initialized only upon first activation
- Added "open/close all" options in datasets dashboard
- Made profiling optional when using data preload in project creation
- Restored the save query option in SPARQL editor for authenticated users
- Added support for SPARQL querying MDR repository (only available for administrators when "Experimental features" option is enabled)
- Allowed SPARQL update to admin
- Added setting for showing/hiding deprecated resources in trees/lists
- Added timestamp info to Annotations of Digital Objects
- Supported sticky header in several tables
- Added support for several presentation customizations:
- header/footer entries with placeholders for logos, title, footer menu etc..
- custom additional informative pages
- new (optional) header custom menu: each entry can point to an external link or to a newly created custom page
- customizable color themes
- Restricted dataset profilation to authenticated users only
- Restricted dataset discovery to authenticated users only
- Dynamically updated the title of browser tab based on the current page
- Fixed an issue preventing the Parameterized Query editor from appearing
- Fixed an issue preventing details of a selected element in UML-like diagrams from being displayed
- Fixed a bug that caused the language value filter to not work with SKOS lexicalizations
- Fixed a bug in ResourceView template editor which affected "Include in/Remove from all resources" quick actions
- Fixed minor issue in lexicalization sets charts when the same language occurrs in multiple lexicalization models
- Minor bugfixes
- Updated several dependencies addressing vulnerabilities
SemanticTurkey
https://bitbucket.org/art-uniroma2/semantic-turkey/src/master/ChangeLog.txt
VocBench Web App
https://bitbucket.org/art-uniroma2/vocbench3/src/master/ChangeLog.md
ShowVoc Web App
https://bitbucket.org/art-uniroma2/showvoc/src/master/ChangeLog.md
Sheet2RDF
https://bitbucket.org/art-uniroma2/sheet2rdf/src/master/ChangeLog.txt
CODA
https://bitbucket.org/art-uniroma2/coda/src/master/ChangeLog.txt
LIME API
https://bitbucket.org/art-uniroma2/lime-api/src/master/ChangeLog.txt
MAPLE
https://bitbucket.org/art-uniroma2/maple/src/master/ChangeLog.txt
Migration from a previous version of the system
As usual, for people migrating from a previous version of VocBench 3 / ShowVoc, it suffices to:
- keep the “SemanticTurkeyData” directory (you might want to make a backup just in case anything happens, even though the system has been extensively tested for migration)
- consider that some migration routines will make the data directory not entirely compatible with prior versions of VocBench
- remove the previous distribution (directory semanticturkey-x.y) and replace it with the new one inside the vocbench3-15.1-full.zip (or equivalent of ShowVoc) archive
- if using an external triple store:
- before moving/running anything, pls backup your data!!!
- for GDB only (both 10.x and 11.x)
- for GDB 10.x
- if moving to GDB 11.x
- deploy the st-changetracking-sail-15.1.jar, that can be extracted from:
/semanticturkey-15.1.0/lib/st-changetracking-sail/15.1
into the lib directory of the triple store - deploy the st-trivial-inference-sail-15.1.jar, that can be extracted from:
/semanticturkey-15.1.0/lib/st-trivial-inference-sail/15.1
into the lib directory of the triple store
For developers willing to recompile the project:
all of the dependencies of VocBench and ShowVoc based on other software written by us:
have already been committed to Maven Central, so they may choose what to re-build and what to take as-is from the dependencies
Cheers,
The VocBench/ShowVoc team