The entity and entitybyuri queries currently return an EntityReference that contains zero or more knownEntityDescriptions. They do this even if there is just one knownEntityDescription. This behavior is not what most clients desire

Would recommend the following:
(a) (re-)add the tag that was present in the LexGrid model that could be used to identify one of the coding systems as the "primary"
(b) State that the default behavior of entity and entitybyuri is to redirect to the only or (when present) tagged entry in the specific codesystem version. The id is that //entity/12345 will redirect to //codesystem/

/version/

/entity/12345 if it is the only describing code system version or its code system (version?) is listed as the primary
(c) add an additional parameter (fwd?) that, when false, says that redirection should not occur.

Logged: https://github.com/cts2/cts2-specification/issues/129

No Data Available

Both the format and language can be supplied as a url parameter or as part of the http header. The current specification doesn't address what to do if both are present.

Recommendation is that the command line takes precedence over Accept and Accept-Language in the header.

Logged: https://github.com/cts2/cts2-specification/issues/130

No Data Available

The change set includes a CLONE type, but we haven't specified a method for asking that a CLONE be performed.

We would anticipate that the method signature would include:
nameOrURI of the the versionable resource to be Cloned
the new version identifier (optional - could be assigned by service or error if it doesn't)
the new version URI (optional)
the change set URI

(we need to check the documentation to see whether there is anything else)

Logged: https://github.com/cts2/cts2-specification/issues/135

No Data Available

The CTS2 specification describes the minimum (and, at the moment, only) content that can be in directories of various types. Implementers are discovering that they need to add information to this to meet special use cases (e.g. ChangeSetURI and ChangeSetNotes for a special value set authoring environment). We need a way to allow directories to be extended for service specific implementations.

Logged: https://github.com/cts2/cts2-specification/issues/137

No Data Available

The resolveAsCompleteSet function allows the client to ask for a ResolvedValueSet (vs. IterableResolvedValueSet). There is currently no way to do this in the REST functions.

Would propose that we add a parameter to queryControl or update the possible values for maxToReturn so a client can indicate that they don't want an iterator, they want the whole thing. We will have to supply a new error message as well - something to the effect of "too much to return all at once"

Logged: https://github.com/cts2/cts2-specification/issues/125

No Data Available

The current tag, "referencelanguage" is a bit unwieldy from the REST perspective. Would recommend adding a synonym, "lang" to address this issue.

Logged: https://github.com/cts2/cts2-specification/issues/127

No Data Available

CodeSystemCatalogEntry.xsd is missing "hasOntologyLanguage" and "includes" elements. These were added to cts2/spec/submission/OMGServerMap/codesystem/CodeSystem.xsd in revision 7641

Logged: https://github.com/cts2/cts2-specification/issues/121

No Data Available

The mapFrom and mapTo elements in Map Entry are currently of type URIAndEntityName, which doesn't allow an explanatory designation to be provided, meaning that it may be necessary to resolve the returned types against a code system. This type should be changed to EntitySynopsis. Changes have been submitted to CTS2 svn repository as change 7853

Logged: https://github.com/cts2/cts2-specification/issues/122

Disposition: See issue 18532 for disposition

While it is possible to construct a URL that requests the resolution of the CURRENT version of a value set, there is no href in the ValueSet schema that allows this query to be performed without knowing URL construction rules. There should be another optional link called "resolution" or something similar where a service can provide an href that allows direct access to the value set resolution.

Logged: https://github.com/cts2/cts2-specification/issues/134

No Data Available

Currently there is no way to resolve a value set definition directly without knowing a specific version - it is first necessary to find the definition via query statements. The REST signatures should include:
//valueset/resolution[?tag=

], which will resolve CURRENT or tag directly. This has already been implemented in the SVS mat implementation.

Logged: https://github.com/cts2/cts2-specification/issues/123

No Data Available

The documentation for //valueset/

/definition/

/resolution found at http://informatics.mayo.edu/svn/trunk/cts2/spec/psm/rest/html/cts2.html#id1171374283352 makes no sense - we need to check the WADL in this area to see what is going on.

Logged: https://github.com/cts2/cts2-specification/issues/124

This is just an WADL -> HTML generation issue – no change needed, documentation is correct in the WADL.
Disposition: Closed, no change

Line 60 in MapVersion.xsd has the wrong type for versiontag. Change:

<xs:element name="versionTag" type="core:MapVersionReference" minOccurs="0" maxOccurs="unbounded">

to

<xs:element name="versionTag" type="core:VersionTagName" minOccurs="0" maxOccurs="unbounded">

Change submitted to svn - cts2/spec/submission/OMGServerMap/mapversion/MapVersion.xsd - revision 7877

Logged: https://github.com/cts2/cts2-specification/issues/133

No Data Available

The REST specification doesn't specify what the service behavior should be on "byuri" type queries. Some implementations use a 303 to get to the base resource while others leave the URI as entered.

Would propose that the specification say that it should always redirect.

Logged: https://github.com/cts2/cts2-specification/issues/128

No Data Available

EntityDescription provides URI's for children, descendants and ancestors, but it only provides parents directly. While this is useful, it means that applications that are traversing graphs have to have two different mechanisms - one for the first three categories and a second for parents. We should add one more attribute, "parents" which allows parents to be accessed by URI as well.

Logged: https://github.com/cts2/cts2-specification/issues/136

No Data Available

Section 2.2.1.1 (Operation resolve), input parameter 'directory' is specified as type CodeSystemCatalogURI. Type should be 'CodeSystemCatalogEntryDirectoryURI'

Section 2.2.1.2 (Operation resolveAsList), input parameter 'directory' is specified as type CodeSystemCatalogURI. Type should be 'CodeSystemCatalogEntryDirectoryURI'

Logged: https://github.com/cts2/cts2-specification/issues/96

This issue does not exist in the "CTS2 - Code System and Code System Version Catalog Services, v1.0" document.
Disposition: Closed, no change

There is no method with a signature: f(conceptDomain, context[0..*], codeSystemVersion[0..*], tag[0..1]) --> ResolvedValueSet

(In that there is no way to say "Give me the resolved value set for this concept domain in this context")

https://github.com/cts2/cts2-specification/issues/149

No Data Available

Issues #141 (http://www.omg.org/issues/cts2-rtf#Issue18470) and #122 (http://www.omg.org/issues/cts2-rtf#Issue18486) add an optional description to valueSetEntry, mapFrom and mapTo. It appears that, with the exception of EntityType and ResourceType, the remainder of the URIAndEntityName references should all be changed to EntitySynopsis, allowing an optional description. Would propose removing EntitySynopsis completely from the model and moving its description property to URIAndEntityName.

Logged: https://github.com/cts2/cts2-specification/issues/143

No Data Available

ResolvedValueSet has 'member' elements while IteratableResolvedValueSet has 'entry' elements.

Recommend moving all elements to 'entry' as that is the pattern for everything else.

Logged: https://github.com/cts2/cts2-specification/issues/147

No Data Available

CTS2Exception in the PIM consists of message and severity. The XML schema has an extra element that makes no sense - ExceptionType. This appears to be an artifact that was accidentally left in the UML model, but is never used in the spec itself.

This should be removed from both places - the UML and the exception schema

Logged: https://github.com/cts2/cts2-specification/issues/148

No Data Available

In the description of the Code System Catalog Entry the identification is done via the Code System name: “codeSystemName - the local identifier that uniquely identifies the code system within the context of the implementing service. Note that the about URI is the globally unique identifier”.

The HL7 culture is used to manage messages and documents via OIDs which. An optional attribute: CodeSystemOID is needed. The issue here is not turning an OID into a URI but having the possibility to have SIMULTANEOUSLY the URI and OID present (an additional attribute aside from about). In all IHE domains we use OIDs to identify objects. There is no strong consensus and a mechanism of sharing URI so that complete interoperability can be achieved. In this transition context we need a means to assure this transition while assuring interoperability. https://github.com/cts2/cts2-specification/issues/144

No Data Available

In cases where the associationID is supplied, it is the ID that gets you from the directory contents to the actual association.

This change has been added to revision 7733 in cts2/spec/submission/OMGServerMap/association/Association.xsd on 10/4/2012

Logged: https://github.com/cts2/cts2-specification/issues/118

No Data Available

Generating associationID's when none exist just produces noise. The associationID parameter should made optional in the PIM where the service doesn't support AssociationRead. This needs to be changed in the schema AND the spec.

Schema has been updated in revision 7752 in cts2/spec/submission/OMGServerMap/association/Association.xsd

Logged: https://github.com/cts2/cts2-specification/issues/119

No Data Available

The PropertyReference class in Filter, and corresponding elements in the query signatures doesn't specify exactly how one should create a referenceTarget for references of type Attribute. As an example, how would one go about specifying the uri attribute in the source component of a sourceAndRole element? The behavior is underspecified if the reference not to a leaf (e.g. "sourceAndRole") Recommend it should specify match anything except href.

If it is going to require URIAndEntityName (vs. 'or'), we need to clearly state what URI is used in the CTS2 spec. Otherwise we need to change the signature to allow just a name (preferred).

Logged: https://github.com/cts2/cts2-specification/issues/151

No Data Available

The association graph structure is considerably easier to use if has a dept indicator. While this could be computed, it is much easier to traverse if explicitly present.

This change has been added to Association.xsd in cts2/spec/submission/OMGServerMap/association

Still needs to be documented and added to the spec itself

Logged: https://github.com/cts2/cts2-specification/issues/120

No Data Available

Section 2.2.1.1 (Operation resolve), input parameter 'directory' is specified as type ConceptDomainDirectoryURI. Type should be 'ConceptDomainCatalogEntryDirectoryURI'

Logged: https://github.com/cts2/cts2-specification/issues/97

This issue does not exist in the "CTS2 - Concept Domain and Concept Domain Binding Services, v1.0" document.
Disposition: Closed, no change

URIAndEntityName does not represent what is in the UML. The UML says that the only thing that is required is the uri – the name and href are optional. The Schema says that the namespace and name is required.

Logged: https://github.com/cts2/cts2-specification/issues/98

No Data Available

Appendix 10 of the OMG "Hitchhiker's Guide" (http://www.omg.org/cgi-bin/doc?hh) specifies a directory structure for organizing submission documents. While older specifications were organized as "spec.omg.org/...", the current direction is that the documents should be organized in "www.omg.org/spec//, where version would be "Beta2", "1.0" for the base spec and "yyyymmmm" for accompanying files.

The OMG AB believes that this set of requirements would imply that the CTS2 URIs should change from "http://spec.omg.org/cts2/1.0/" to "http://www.omg.org/spec/cts2/1.0/".

Logged: https://github.com/cts2/cts2-specification/issues/90

Resolution (PIM)
N/A

Resolution (PSM)

In each of the following files, all references to: "http://schema.omg.org/spec/CTS2/1.0"
are being changed to: "http://www.omg.org/spec/CTS2/1.1".

Change Location

AdvancedAssociationQueryService.wsdl
AssociationHistoryService.wsdl
AssociationMaintenanceService.wsdl
AssociationQueryService.wsdl
AssociationReadService.wsdl
AssociationTransformService.wsdl
BaseExportService.wsdl
BaseImportService.wsdl
CodeSystemCatalogHistoryService.wsdl
CodeSystemCatalogMaintenanceService.wsdl
CodeSystemCatalogQueryService.wsdl
CodeSystemCatalogReadService.wsdl
CodeSystemVersionCatalogHistoryService.wsdl
CodeSystemVersionCatalogMaintenanceService.wsdl
CodeSystemVersionCatalogQueryService.wsdl
CodeSystemVersionCatalogReadService.wsdl
ConceptDomainBindingMaintenanceService.wsdl
ConceptDomainBindingQueryService.wsdl
ConceptDomainBindingReadService.wsdl
ConceptDomainCatalogHistoryService.wsdl
ConceptDomainCatalogMaintenanceService.wsdl
ConceptDomainCatalogQueryService.wsdl
ConceptDomainCatalogReadService.wsdl
EntityDescriptionHistoryService.wsdl
EntityDescriptionMaintenanceService.wsdl
EntityDescriptionQueryService.wsdl
EntityDescriptionReadService.wsdl
EntityDescriptionTransformService.wsdl
MapCatalogHistoryService.wsdl
MapCatalogMaintenanceService.wsdl
MapCatalogQueryService.wsdl
MapCatalogReadService.wsdl
MapEntryHistoryService.wsdl
MapEntryMaintenanceService.wsdl
MapEntryQueryService.wsdl
MapEntryReadService.wsdl
MapResolutionService.wsdl
MapVersionHistoryService.wsdl
MapVersionMaintenanceService.wsdl
MapVersionQueryService.wsdl
MapVersionReadService.wsdl
ReasoningService.wsdl
ResolvedValueSetLoader.wsdl
ResolvedValueSetQueryService.wsdl
ResolvedValueSetResolution.wsdl
StatementHistoryService.wsdl
StatementQueryService.wsdl
StatementReadService.wsdl
UpdateService.wsdl
ValueSetCatalogHistoryService.wsdl
ValueSetCatalogMaintenanceService.wsdl
ValueSetCatalogQueryService.wsdl
ValueSetCatalogReadService.wsdl
ValueSetDefinitionHistoryService.wsdl
ValueSetDefinitionMaintenanceService.wsdl
ValueSetDefinitionQueryService.wsdl
ValueSetDefinitionReadService.wsdl
ValueSetDefinitionResolution.wsdl

Change Description

Changes to machine readable files are easily identified by Git using the color red to indicate “deletion” and green to indicate “addition.” Furthermore, the ‘-‘ minus sign is also displayed to indicate a deletion, and a “+” plus sign is displayed to indicate an addition.

View Changes: https://github.com/cts2/cts2-specification/commit/890558063ee42dd3a21419f942abe16f0a634677

Due to number of changes, please review the PDF.
http://informatics.mayo.edu/svn/trunk/cts2/spec/submission/RTF_1.1/RTF_Issues/CTS2_RTF_1.1_17318.pdf

On page 39 of Core the description of the documentURI attribute in ResourceVersionDescription stops in mid sentence with a "100".

Logged: https://github.com/cts2/cts2-specification/issues/139

Resolution (PIM)
Location 1
Document "CTS2 - Core Model Elements, v1.0"
Section 2.7.1.5
Page 39
Change Description 1
Change made to Attributes section.
Old Text:
documentURI - a URI that identifies the specific version, language, and notation of the about resource. This URI needs to be constructed in such a way that, if necessary, it will be possible to differentiate resource versions that were loaded from different document syntaxes. As an example, if an image of the wine ontology was loaded from a resource that was in Manchester Syntax, it should be given a different URI than the image loaded from the RDF/XML syntax. The reasoning behind this is, even in cases where different syntaxes are 100.
New Text:
documentURI - a URI that identifies the specific version, language and notation of the about resource. This URI needs to be constructed in such a way that, if necessary, it will be possible to differentiate resource versions that were loaded from different document syntaxes. As an example, if an image of a the wine ontology was loaded from a resource that was in Manchester Syntax, it should be given a different URI than the image loaded from the RDF/XML syntax. The reasoning behind this is, even in cases where different syntaxes are 100% compatible the transformation into the CTS2 model may not be identical.
Resolution (PSM)
Change Location
Core.xsd
Change Description
Changes to machine readable files are easily identified by Git using the color red to indicate “deletion” and green to indicate “addition.” Furthermore, the ‘-‘ minus sign is also displayed to indicate a deletion, and a “+” plus sign is displayed to indicate an addition.
View changes: https://github.com/cts2/cts2-specification/commit/138ee6180cb52fa1176ac0f3bbfd93b54de34205

The current description of entityType in section 2.1.1.1 of the Entity Description Services (page 6) reads:
entityType - the set of type(s) which the entityReference is an instance of. Because this is a terminology service, entityType must include one of owl:class, owl:individual, rdf:predicate, or skos:concept, although it may carry many other types as well.

This is incorrect - it should read:
"... must include one of owl:Class, owl:Individual, rdf:Property or skos:Concept"

Note that this may potentially effect some implementations that have either transcribed the information literally or have used "rdf:predicate" instead of "ref:Property"

Logged: https://github.com/cts2/cts2-specification/issues/142

At the moment, SourceAndNotation is a required field. In many situations there is nothing of value that can be supplied here - it should be made optional.

Logged: https://github.com/cts2/cts2-specification/issues/140

see pages 35 - 37 of prc/2013-06-02 for details

AssociatedEntitiesReference.referencedEntity and SpecificEntityList.referencedEntity are currently of type URIAndEntityName. This does not provide any place to include a designation or description of the entity. The type should be changed to EntitySynopsis

Logged: https://github.com/cts2/cts2-specification/issues/141

See issue 18532 for disposition

The DocumentURI was intended to differentiate situations where there were potentially multiple sources of the same version of a resource (e.g. SNOMED CT 20120731 from RF1, RF2, RRF or Owl). This is the exception, rather than the rule, and the DocumentURI field should be made optional for clarity.

Logged: https://github.com/cts2/cts2-specification/issues/138

Resolution (PIM)
Change Location 1
Document "CTS2 - Core Model Elements, v1.0"
Section 2.7.2
Page 40
Figure 2.17
Change Description 1
Updated diagram in Figure 2.17
Old Diagram:

In each Query Service, it would be helpful for the user to be able to select certain attributes of the resource to be included (or excluded). For instance, a user may only be interested in the Definitions, or a given named property. Right now, for the Query Service, the user is limited to either 1) the directory entry, or 2) the list representation, which is the entire resource. It would be helpful to have the option of something in between.

Logged: https://github.com/cts2/cts2-specification/issues/155

See OMG Issue 18474 (#137) for disposition.

Disposition: See issue 18474 for disposition

Currently, to get an association graph in REST, the graph is retrieved by:

/codesystem/cs/version/1.0/graph

It would be more correct to be like:

/codesystem/cs/version/1.0/associations/graph

to take into account the 'graph' taking in the association directory as an input. Right now, the 'graph' resource is used in place of the 'associations' resource, but this makes it very hard to do queries such as:

/codesystem/cs/version/1.0/entity/C12345/children/graph

Logged: https://github.com/cts2/cts2-specification/issues/152

No Data Available

In the schema: http://www.omg.org/spec/cts2/201206/mapversion/MapEntryServices.xsd, the types 'MapTargetList' and 'MapTargetListList' need corresponding Message elements, such as 'MapTargetListMsg' and 'MapTargetListListMsg.'

The WADL would need to be updated accordingly to represent the new return types.

Logged: https://github.com/cts2/cts2-specification/issues/154

No Data Available

The UML model in figure 4.4 of the ValueSet document lists the cardinality of includesResolvedValueSet as [0..1]. It should be [0..*].

https://github.com/cts2/cts2-specification/issues/150

No Data Available

ValueSetDefinitionResolution 'resolve' method has incorrect return type. It should be IterableResolvedValueSet.

6.1.1.1 Operation: resolve

Resolve the value set definition against the supplied code system versions and/or version tags.

....

Return Type: ResolvedValueSetDirectory

should be:

6.1.1.1 Operation: resolve

Resolve the value set definition against the supplied code system versions and/or version tags.

....

Return Type: IterableResolvedValueSet

Logged: https://github.com/cts2/cts2-specification/issues/159

No Data Available

The specification states that derivation can be ASSERTED, INFERRED or omitted, meaning that the derivation is unknown. The UML diagram shows it as required, which is wrong. The XML Schema shows it with a default of ASSERTED, which also needs to be removed.

Logged: https://github.com/cts2/cts2-specification/issues/157

see pages 258 - 260 of ptc/2013-06-02 for details

In the WADL, two MapEntry URLS were omitted:
map/

/version/

/entrybyuri/resolution?uri=.

Logged: https://github.com/cts2/cts2-specification/issues/153

No Data Available

The derivation attribute (INFERRED/ASSERTED) needs to be visible in an association directory, as it helps consumers understand the abstract structure.

Logged: https://github.com/cts2/cts2-specification/issues/156

No Data Available

Type of maxtoreturn is string in the WADL definitions. It should be NaturalNumber (or equiv)

Logged: https://github.com/cts2/cts2-specification/issues/158

No Data Available