Linked Data Subject pages: distributions or data services or non-of-both?

[bertvannuffelen]

I have the following discussion on which I have no clue about how to categorize them.

Online consultation of data is that a distribution, a data service or non-of-both?

Consider Linked Data subject pages:

if one dereferences a URI then a document URI is returned. This html document can be nicely styled, having the data organized in such a way that I as a reader find it pleasant to read it.

The collection of all subjectpages for URIs in a dataset is that a distribution of a dataset? Or is it a data service because it renders the data in a pleasant reading way?

When the subject page shows also the location on a nice map, and offers additional further exploration links based on the location and the associated properties, can it then still be considered a distribution?

As we consider Linked Data Does if the subject page shows the labels of the codes and organisations by dereferencing the URIs , can it then still be considered a distribution?

If Linked Data Subject pages is categorized as a distribution, is then a map viewer also a distribution?

Are Linked Data Subject page data services because they use HTTP as protocol? Thus is using http as protocol sufficient to categorize it as a dataservice?

This discussion is probably related to many issues such as #1242, but as it is a concrete case I have registered it separately.

[andrea-perego]

@bertvannuffelen , you're touching a number of different aspects which, although interrelated, may need to be addressed separately.

It would be beneficial to the discussion if you could break down the issues you're raising, and complement your comments with (ideally, real-world) examples.

[bertvannuffelen]

A more concrete example: the collection of all html pages of DBPedia is that a distribution of the DBpedia dataset? E.g. https://dbpedia.org/page/Belgium

To my knowledge, this collection of all html pages is not downloadable, only online consultable. (It is not listed on https://www.dbpedia.org/resources/)

Observe that the html rendering has been done some "data interpretation": namely the flag is shown next to the abstract. The same information, not interpreted, can be found in the data below.

If that is not a distribution, is the html rendering then a dataservice? Namely a nice html rendering service of the data available at http://dbpedia.org/data/Belgium.ttl

[smrgeoinfo]

If all of DBPedia were accessible as a single download consisting of a set of individually identifiable html documents, that would constitute a kind of distribution. But it doesn't exist. A dataset can exist independent of any distribution. http://dbpedia.org/sparql is a data service that allows one to extract data from DBPedia.

An html (or ttl) page that renders a single DBPedia record could be considered a dataset whose subject is the subject of that DBPedia record, and its distribution would be a data download.

[davebrowning]

Project/Milestone modified.

Explanation: As DCAT v3 moves through review and hopefully ratification, we want to make sure that open issues and feedback that have yet to be completely addressed are properly recorded and tagged/assigned in github to both clarify their status and to help review and prioritise as a source of improvements and new requirements in future DCAT versions