HCLS/Banff2007Demo/HCLS/Banff2007Demo/KnownBugs

From W3C Wiki
Jump to: navigation, search

This is a list of known bugs in the data or documentation associated with the Demo. As they are fixed we will leave a note here to that effect.


Reversal of whole/part in queries

Reported by RayHookway.

Reversal of ?whole and ?part in the GO queries that have to do with the part_of relation. For example, in the query


prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>
prefix owl: <http://www.w3.org/2002/07/owl#>
 
select  ?whole ?part ?wholeName ?partName
from <http://purl.org/commons/hcls/20070416>
where
 {
   ?whole rdfs:subClassOf ?sub.
   ?sub rdf:type owl:Restriction.
   ?sub owl:onProperty <http://purl.org/obo/owl/part_of>.
   ?sub owl:someValuesFrom ?part.
   ?whole rdfs:label ?wholeName.
   ?part rdfs:label ?partName.
   filter (?whole = <http://purl.org/obo/owl/GO#GO_0000119>) .
 }


?whole and ?part are reversed. The query returns the following which implies that DNA-directed RNA polymerase II, holoenzyme is part_of a mediator complex when in fact just the opposite is true. This problem also exists in the wiki.

Choice of PURLs

Some PURLs in the /science/ (i.e. Science Commons) domain probably ought to be in the /commons/ (i.e. community)domain instead. For example, http://purl.org/science/article/pmid/16228203 designates an article, and if community process accepts this as a "good" URI for that article, then the URI ought to live in the /commons/ domain, not the /science/ domain. See http://sw.neurocommons.org/2007/uri-explanation.html .

It may also be desirable for administrative reasons to reverse the databank and type components of the paths. E.g. we may want to write http://purl.org/commons/ncbi_gene/record/1003064 instead of http://purl.org/commons/record/ncbi_gene/1003064 . (It has also been suggested that the type be placed following the record identifier with a . separator instead of / , e.g. http://purl.org/commons/ncbi_gene/1003064.record .)

As indicated elsewhere, these PURLs are to be considered provisional, and subject to review may be changed. If this happens it will most likely happen in 2007.

Inferred part_of for GO use different URI for part_of

The originals have: `http://purl.org/obo/owl/part_of` The inferred relations described here use `http://purl.org/obo/owl/obo#part_of`