In 2.1 we introduced to read the schema name from configuration, to be abel to produce same name as in 2.0. However we discovered, in last minute, that this was wrong. The schema name shoudl be DATEXIISchema_[model base version]_[tool major version]_[tool minor version].xsd

The last minute change did not revert back to the old rules, instead we just changed to configuration file. That will create a wrong schema name if you use a 2.0 model or if you use a future version. So the best is to revert back to old rules.

In the DATEX-Tool, it is not possible to restrict the multiplicity of an attribute, which is mandatory (at least I suppose this is the point).

Example:
'accidentType' has multiplicity 1:*. It is not possible to change it to 1:1, which should be possible according to the rules. The Tool does not accept the value 1 - it does block any further action in this case, until the 1 is deleted again.

Note: This problem differs from the one described just before.

In the current version of the DATEX Tool, multiplicities and enumeration literals can be modified according to the rules of DATEX II 2.1, but these changes don't have any effect on the resulting schema (i.e. all literals are present, multiplicities are unchanged).

Deseleting just one enumation literal with the Tool took me 5 minutes (!) and 15 seconds from click to reaction. The Tool occupied 50% of my dual core CPU within this time.

As you can see in the pciture, the enumaration atribute (here: 'boundary' from 'referentType') is quite deep inside the tree, which is obviously the problem.

The DATEX Tool of version 2.1 crashes after reading a specific XMI-file into the DATEX Tool and klicking the tab "selection".

The error ist attached as picture, Jonas Jäderberg is informed about the more detailed error text and the XMI causing this error.

The component LinearAlongLinearElement resp. PointAlongLinearElement can be attached at maximum once to the component Linear resp. Point.

But there is the need of using this container multiple times to describe the same location. In Germany, for example, there is an inter-urban georeferencing method, which can be best represented with this ISO 19148 container - in addition to some edge representation, for example.

Proposal: Change upper multiplicity for this aggreagtion from 1 to *, so that the usage of this container is not longer restricted (in each case Point and Linear).

The ability to extend enumerations whould have been a nice feature to hava and have been discussed earlier. For future versions it would be interesting to elaborate this.

I found an interesting articel on this subject.
http://www.ibm.com/developerworks/library/x-extenum/

I alsi attach the articel as a PDF

Solution four looks very interesting and promising.

The DATEX-definition for the attribute "defaultLanguage" is

"The default language used throughout the payload publications, specified by an ISO 639-2 3-alpha code."

and similar for "language" of MultilingualStrings.

In our perception, this definition is ignored, and usually the two-alpha code instead of the three-alpha code is used (even in the XML-examples on the website).
The internal UML schema "language" does allow both variants.

Proposal: Change definition(s) to ISO 639-1 two-alpha code (or maybe allow usage of both variants).

there is a need to indicate generically the number of heavy vehicles (as well as light vehicles) within an accident in real time.. 

for this purpose  the reusable class GroupsOfVehicleInvolved can be used with a specific VehicleCharacteristcs with VehicleType set as "heavyVehicle" that was in the past 1.0 version of the Datex2 model.

heavy vehicle is a category that includes all vehicles that are not light vehicles ( whose number is possible to manage using carOrLightVehicle as vehicleType ) 

there is a lack of information for planned events events

the information managemnt uses 3 attributes to manage this  kind of information:

probabilityOfOccurrence  in SituationRecord which may contain the following values:

• certain The source is completely certain of the occurrence of the situation record version content.