ISSUE-187: Do we want to keep the same occurrence constraints as defined in DAQ (for example, that every metric should belong to exactly one dimension)?
Do we want to keep the same occurrence constraints as defined in DAQ (for example, that every metric should belong to exactly one dimension)?
- State:
- CLOSED
- Product:
- Quality & Granularity Vocabulary
- Raised by:
- Antoine Isaac
- Opened on:
- 2015-06-12
- Description:
- Do we want to keep the same occurrence constraints as defined in DAQ (for example, that every metric should belong to exactly one dimension)? In this specific case this may be demanding too much on quality data publishers: it could be that a metric does not clearly belong to a dimension, or that a metric is in scope for several dimensions.
- Related Actions Items:
- No related actions
- Related emails:
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from amgreiner@lbl.gov on 2015-09-18)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from nmihindu@fi.upm.es on 2015-09-18)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from amgreiner@lbl.gov on 2015-09-16)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from nmihindu@fi.upm.es on 2015-09-15)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from amgreiner@lbl.gov on 2015-09-14)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from aisaac@few.vu.nl on 2015-09-13)
- Re: DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from Jeremy.Debattista@iais.fraunhofer.de on 2015-09-11)
- DQV ISSUE-187 - cardinality of the link between Metric and Dimension (from aisaac@few.vu.nl on 2015-09-11)
- dwbp-ISSUE-187: Do we want to keep the same occurrence constraints as defined in DAQ (for example, that every metric should belong to exactly one dimension)? [Quality & Granularity Vocabulary] (from sysbot+tracker@w3.org on 2015-06-12)
Related notes:
This issue has been discussed in the following call:
http://www.w3.org/2013/meeting/dwbp/2015-09-11
The group is generally inclined to not constrain the number of dimensions a Metric is connected to.
Nandana knows of quality metrics that could be classified in several dimensions (if just because such classifications are not crisp). He also observed that the restriction may actually not be formally stated in the formal ttl representation of daQ (http://butterbur04.iai.uni-bonn.de/ontologies/daq/daq.ttl).
However the doc says "A metric belongs to exactly one dimension" (http://purl.org/eis/vocab/daq#)
The topic needs more investigation.
http://www.w3.org/2013/meeting/dwbp/2015-09-11
Regarding the metrics with multiple dimensions in the 2015-09-11 call, I was concretely referring to the LODQM model [1]. It says "The quality metrics for assessing these quality dimensions are proposed at the second level of the model, some of which are used for measuring two quality dimensions." Figure 1 illustrates this (page 5). For my use cases, I am ok to use DAQ with metrics having only one dimension.
[1] http://2014.eswc-conferences.org/sites/default/files/phdpaper_17.pdf
we decided to not keep the constraints from DAQ but provide guidelines http://www.w3.org/2013/meeting/dwbp/2015-09-24#resolution_9
it is worth noting that DAQ seems move on the same direction.
Display change log
