If I understand it correct, "inneresFormat" means the original (for the data usage much more important) format of the data. And "format" means the compression (e.g. .zip). DCAT-AP format means the data format (the compression is missing in DCAT). So the mapping to DCAT needs to be adjusted. Note: both are important and should be considered in OGD!
Anonymous (not verified)Fri, 17/06/2016 - 10:32
Anonymous (not verified)Fri, 17/06/2016 - 15:22
Thank you for the issue that we will check.
Indeed one of the bigger CKAN /DCAT strutural differences:
To my understanding this says nothing about container or not and in the case of a ZIP-File as a distribution, the MimeType "application/zip" would be used.
Also other people cared about this mapping https://github.com/ckan/ckan/issues/1336 arguing in your direction that DCAT-AP Mimetype must be the content in the container and not the container itself. But what about the checksum in those cases? Checksum of ZIP and MediaType of content does not make much sense to my understanding.
But as we are on joinup and we can easily ask the authors:
I raised the issue in the DCAT-AP issues, just to be sure about the exact mapping.
DCAT-AP unterscheidet nur Format oder MediaType, meint aber die Distribution.
Das dies ohne Container gemeint ist, wird meines Erachtens aus der DCAT-AP Spezifkation nicht klar. Eine Zip-Ressource hat nach meinem Verständnis den Zip MimeType.
Hier https://github.com/ckan/ckan/issues/1336 stützt man aber die These, dass DCAT-AP nicht die Zip-Ressource, sondern das innere des Containers meint. Dann ist aus meiner Sicht aber die Prüfsumme fragwürdig, denn dann wäre es ja die Prüfsumme der Datei(e)n im Container.
Aber wir sind ja auf Joinup und können die Autoren von DCAT-AP diesbezüglich einbeziehen.
Vielen Dank für Ihren Beitrag. Wir werden nun eine Weile brauchen, um alle Hinweise zu prüfen und Ihnen dann hier im Portal eine Rückmeldung geben.
Thanks a lot for your input. We will now take some time to review all posted issues. Afterwards you will receive our feedback on this website.
Comments
Thank you for the issue that we will check.
Indeed one of the bigger CKAN /DCAT strutural differences:
CKAN features both mimetype and mimetyp_inner, http://https://lists.okfn.org/pipermail/ckan-discuss/2011-April/001141.html whereas DCAT-AP only knows for the same level the distinction between MimeType and if not the newer MediaType.
To my understanding this says nothing about container or not and in the case of a ZIP-File as a distribution, the MimeType "application/zip" would be used.
Also other people cared about this mapping https://github.com/ckan/ckan/issues/1336 arguing in your direction that DCAT-AP Mimetype must be the content in the container and not the container itself. But what about the checksum in those cases? Checksum of ZIP and MediaType of content does not make much sense to my understanding.
But as we are on joinup and we can easily ask the authors:
I raised the issue in the DCAT-AP issues, just to be sure about the exact mapping.
https://joinup.ec.europa.eu/discussion/question-usage-mimetype-and-mediatype-zippedcontainer-distributions
-----
Vielen Dank für den Hinweis, den wir gern prüfen.
CKAN kennt beides, https://lists.okfn.org/pipermail/ckan-discuss/2011-April/001141.html
DCAT-AP unterscheidet nur Format oder MediaType, meint aber die Distribution.
Das dies ohne Container gemeint ist, wird meines Erachtens aus der DCAT-AP Spezifkation nicht klar. Eine Zip-Ressource hat nach meinem Verständnis den Zip MimeType.
Hier https://github.com/ckan/ckan/issues/1336 stützt man aber die These, dass DCAT-AP nicht die Zip-Ressource, sondern das innere des Containers meint. Dann ist aus meiner Sicht aber die Prüfsumme fragwürdig, denn dann wäre es ja die Prüfsumme der Datei(e)n im Container.
Aber wir sind ja auf Joinup und können die Autoren von DCAT-AP diesbezüglich einbeziehen.
Ich habe dies soeben getan:
https://joinup.ec.europa.eu/discussion/question-usage-mimetype-and-mediatype-zippedcontainer-distributions
What if in a ZIP-File are multiple files, hence, multiple entries for inneresFormat?
Shouldn't be the cardinality of "inneresFormat" be [0..*] instead of [0..1]?
Can examples of data be provided which needs to be compressed as ZIP-files?
This issue is definitely something to be discussed at the W3C DCAT workshop in Amsterdam later this year.
I'm not sure whether inneresFormat is a good idea in any case, but if it is kept then I second the suggestion of allowing multiple formats.