Re: [dxwg] Is one of dcat:accessURL, dcat:accessService and dcat:downloadURL required in a dcat:Distribution? (#1197)

Hi @nfreire and @aisaac,

> Hello DXWG,
> At Europeana, we are designing a profile where we need to use the class dcat:Distribution without using any of the three properties dcat:accessURL, dcat:accessService and dcat:downloadURL.

I  admit I have some problems imagining the use case.  What do these distributions represent?

> The dcat:Distribution states in the last paragraph of section 6.7:
> > Links between a dcat:Distribution and services or Web addresses where it can be accessed are expressed using dcat:accessURL, dcat:accessService, dcat:downloadURL, as shown in Figure 1 and described in the definitions below.
> In the usage notes for these three properties, it is writen:
> > dcat:accessService SHOULD be used to link to a description of a dcat:DataService that can provide access to this distribution.
> > dcat:accessURL SHOULD be used for the URL of a service or location that can provide access to this distribution, typically through a Web form, query or API call.
> > dcat:downloadURL SHOULD be used for the URL at which this distribution is available directly,
> Many thanks in advance.

Your use of dcat:Distribution is not the classic intended case, but in general, I like to take a liberal interpretation where "anything is possible unless not explicitly forbidden".
In this case, I do not remember an explicit prohibition, but other editors/contributors might recall otherwise here. 

 I have reported their [bcp 14]( definitions below,  as I think the difference between "MUST" and "SHOULD"   helps, 

1. **MUST**   This word, or the terms "REQUIRED" or "SHALL", mean that the definition is an absolute requirement of the specification.
3. **SHOULD**   This word, or the adjective "RECOMMENDED", mean that there may exist valid reasons in particular circumstances to ignore a particular item, **but the full implications must be understood and carefully weighed before choosing a different course.**

In this spirit,  I think you can use the dcat:distribution without the properties above provided that
 (i) you are not  "reinventing" a solution for a use case we have already solved differently;
 (ii) you are not using other properties in place of those suggested by DCAT to say what you would have featured with the original DCAT properties.

I hope this general consideration is helpful. If it is not, I am afraid that we need further info about your specific case to frame the reply more.

GitHub Notification of comment by riccardoAlbertoni
Please view or discuss this issue at using your GitHub account

Received on Wednesday, 18 December 2019 18:38:51 UTC