Exporting data from ShowVoc

[Christopher Mutel]

Dear all-

Thanks so much for creating this whole tool chain, it is a great step towards making semantic data more easily usable by the wider audience! It's also great that it has gotten the seal of approval from big international organizations.

I would like to download data from the repos hosted by the EU and FAO. I tried looking through the user manual, searching on this list, and playing with the graphical interface, but I didn't see a way to easily download the complete set of terms and alignment tables in a programmatic way (and the download links in "Metadata" must be manually copied/pasted and also seem to be obfuscated in Javascript for some reason). I am not sure that the SPARQL query interface is suitable for prgrammatic use either, though this could be wrong.

Is there an easy way to get the data behind a ShowVoc instance? Are building building software on top of ShowVoc, or is this considered out of scope for this project?

Thanks for your patience, and sorry in advance if this is a very simple question!

Sincerely yours,

Chris Mutel

[Armando Stellato]

Dear Chris,

 

By first, thanks for the appreciation for the VocBench and ShowVoc platforms!

 

The reason for not allowing directly to download a resource by dumping it is to avoid, by default, that any user can put too much stress on the interface with the triple store and, in the end, on the triple store itself. For instance, FAO’s whole LOD stack goes in the range of millions of accesses (of various types, depending on the application, might be access to pages/LD RDF descriptions, might be SPARQL queries, etc..) per month, and having such an operation completely open could wreak havoc on the exposed services. That said, not all instances have this volume of accesses while may still have good gear and bandwidth, so it might make sense in the future to put this as an optional feature that the administrator can toogle.

 

That said, there is not obviously a gap: the idea is (but you found it already) that users should download pre-prepared versions of the repository. The links are currently available in the metadata section but will be moved in the future in a dedicated area called “downloads”.

 

I’ve tried by taking a random dataset from the installation at the EU:

 

https://showvoc.op.europa.eu/#/datasets/ESTAT_Combined_Nomenclature,_2022_%28CN_2022%29/metadata

 

and indeed it works (try the first link, which is the zipped RDF dataset, even though all links work).

 

Access via Web API is possible, but it (currently) requires authentication. There is though, for ShowVoc, a public user that can be used to get the same level of authorization that you would get with the UI.

 

Instructions for accessing via Web API follow from the backend platform (the same for VocBench) and are here: https://semanticturkey.uniroma2.it/doc/user/web_api.jsf

 

The public ShowVoc user has the following credentials:

user: pub...@showvoc.eu

pwd: showvoc

 

The above instructions can be used also for VocBench, by using registered users, while the above SHowVoc Public user is the only one that will be universally configured this way in all SV installations.

 

Finally, about the “software on top”, well, ShowVoc is kind of end-user application. Conversely, the backend behind ShowVoc (Semantic Turkey, https://semanticturkey.uniroma2.it/) has indeed been reused in various applications from time to time. Some of them are client-specific, some of them have been discontinued or were proof of concepts for some research project. Currently, VocBench and ShowVoc are both using this framework and will continue for the years to come. Semantic Turkey is currently undergoing a major update with several architectural changes, which will result in updated VocBench and ShowVoc as well.

We apologize for the delayed release (we expected to deliver within March) but we are fighting with quite some issues in moving to the updated RDF4J 4.x and GDB 10.x and we want to be sure that all users will have a smooth migration of their data from previous versions. In the meanwhile, we are adding some novel features in parallel, so the additional time won’t be wasted :-)

 

Kind Regards,

 

Armando

--
You received this message because you are subscribed to the Google Groups "vocbench-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vocbench-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/vocbench-user/e7850e18-0669-4e74-9e67-85da720c324cn%40googlegroups.com.