Object Management Group
Manufacturing Technology &
Industrial Systems Task Force
MantisLogo

Manufacturing Technology &
Industrial Systems Task Force

Technology Submission Evaluations

CAD Services V1.0
Submission Evaluations

CAD Services V1.0 RFP: mfg/2000-06-07


Index


New!

20 August 2001


Organization

Chairs:

 Uwe Kaufmann
 Fraunhofer, IPK
 Russel Claus
 NASA Glenn

Hosting Working Group:
Product & Process Engineering Working Group

Contact information:
The CAD Version 1.0 Submission Evaluation Team may be contacted via e-mail at mfg_ppe@omg.Please copy the following Submission Teams as well.

CAD Evaluation Team Membership:

Member Organization eMail
 Richard Baumann  Fraunhofer IPK  [email protected]
 Russ Claus (Chair)  NASA  [email protected]
 Gilles Debarbouille  Open Cascade  [email protected]
 Peter Denno  NIST  [email protected]
 David Flater  NIST  [email protected]
 Ron Hare  GE Aircraft Engines  [email protected]
 Rohinton Irani  GE Corp., R&D   [email protected]
 Mattias Johansson  Eurostep  [email protected]
  Larry Johnson  MSC.Software   [email protected]
 Uwe Kaufmann (Chair)  Fraunhofer IPK  [email protected]
 Thommy Kuehn  CADSYS  [email protected]
 Steffen Nowacki  Computer Graphics
 Center
 [email protected]
  Mitch Silverman  UGS   [email protected]
 Matthias Strebel  Fraunhofer IPK  [email protected]
 Ilan Weitzer  Ford  [email protected]

Submission Evaluation

Submission Evaluation Procedures

Schedule:

Date Item
Danvers Meeting
 8-July-2001
 Sunday
 13:00-17:00
Evaluation Team Meeting (Minutes)
 9-July-2001
 Monday
  13:00-17:00
Presentation of Submissions to ManTIS (mfg/2001-07-05)
 10-July-2001
 Tuesday
 15:00-17:00
Evaluation Team Meeting
(Evaluation Report Development)
No Minutes Provided.
 12-July-2001
 Thursday
 09:30
Recommendation Report Presented to ManTIS as follows:

CAD Services Evaluation Team Report
Uwe Kaufmann

The Evaluation Team reviewed the compiled issue list. All issues have been resolved as (a) being dismissed, (b) accepted, and (c) with a few remaining issues identified as addressable in an FTF.

The changes to the submission in the Errata are assessed to be non-substantive and acceptable.

The Evaluation Team's recommendation is to go forward.

 12-July-2001
 Thursday
 09:30
Vote to Vote
(The motion carried 9-0-0)
12-July-2001
Thursday
09:30
Vote to Recommend
(The motion carried 8-0-1)

Evaluation Issue Log

The following is a table of the issues identified by the ManTIS Submission Evaluation Team through the Danvers Meeting, July 2001 and the disposition of those issues. These issues, among others were forwarded to the OMG Issue Log ([email protected]) and addressed by the CAD Finalization Task Force.

Source

No.

Clause/Para

Type

Comment

Proposed change

Resolution

Keith Duddy

1

 

Editorial

Bounding Box description incorrect

Improved description in document mfg/06-03

Agreed. Changes made in mfg/2001-07-09

Keith Duddy

2

 

Technical

#define CadConnectionFault incorrect

Correct definition

Agreed. Changes made in mfg/2001-07-09

Keith Duddy

3

 

Technical

CORBA::Object instead of Cad object type in Tessellation data structures

Known type in place of Object

Disagreed. Known type creates infinite recursion.

Keith Duddy

4

 

Technical

event_open_model(and the other event operation) does not specify how the event service is used

 

Specify event mechanism or remove

Agreed. Event operations removed in mfg/2001-07-09

Keith Duddy

5

 

Editorial

DnyAny text needs clarification

Fuller description in document.

Agreed. Text changes in mfg/2001-07-09

Keith Duddy

6

 

Technical

Structuring the spec in terms ofcompliance points is unconventional

 

Restructure IDL

Agreed. IDL changes in mfg/2001-07-09

Keith Duddy

7

 

Editorial

Copyright waiver, contact points, not in standard form

Include waiver.

Agreed. Text changes in mfg/2001-07-09.

Keith Duddy

8

 

Technical

UML looks like Corba profile, but is not quite

Clarify diagrams.

To be resolved in FTF.

Keith Duddy

9

 

Editorial

Italics used for IDL in running text rather than IDL font

 

To be resolved in FTF.

Keith Duddy

10

 

Technical

modification date - only a "recommended" format. This should be mandatory

ISO standard required.

Agreed. Changes noted in mfg/2001-07-09.

Keith Duddy

11

 

Technical

unique_entities_count should return unsigned long

IDL modified to use unsigned long.

Agreed. Changes noted in mfg/2001-07-09.

Keith Duddy

12

 

Editorial

close_model -  does this save or not

Clarify text.

Agreed. Changes noted in mfg/2001-07-09.

Keith Duddy

13

 

Technical

EntityAttrib is strange... if it can hold an any, why not justuse an any, and then check for the specific types other than any specified by the enum?

Remove any.

Agreed. Changes noted in mfg/2001-07-09.

Keith Duddy

14

 

Editorial

IDL extracts in the document are inconsistent about showingmodule declarations

Insert module declaration

To be resolved in FTF.

Keith Duddy

15

 

Technical

Some calculated properties, such as surface areaor volume, use an inout “accuracy” parameter to set the desired accuracy of the calculated properties. Many CAD systems do not support this functionality and shall return a negative value for “accuracy” to indicate lack of support for this parameter." There is no behaviour defined for the out param when accuracy is supported. Is it acceptable to return a different accuracy from the one specified?

Clarification.

Text changes require implementations to define accuracy parameters. Mfg/2001-07-09

Uwe Kauffman

16

 

Technical

Feature set is too generic

 

To be resolved in FTF.

David Flater

17

 

Technical

Model level accuracy not addressed.

 

Accuracy task added to roadmap.

David Flater

18

 

Editorial

Need sequence diagram for PDM Enablers.

Add diagram as appendix.

To be resolved in FTF.

 

 

 

 

 

 

 

Steffen Nowacki, ZGDV

19

 

Usability

Different Abbreviations for Attribute used

CadUtility::EntityAttrib

CadConnection::NativeCadAttr

Use full word Attribute instead of abbreviations

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

20

 

Usability

User defined type starts with lower case

CadGeometry ::indexStruct

Rename CadGeometry ::indexStruct to CadGeometry ::IndexStruct

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

21

 

Usability

Lower case letters in enumerator names

CadUtility ::AttribTypes

{ longtype,  doubletype,  stringtype,

    booleantype,  anytype };

Use only upper case letters and underscores in enumerator names

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

22

 

Usability

Different Abbreviations for point used CadCurve::LineStruct member pnt;

CadGeometry:: PointOnSurfaceStruct member pt;

Use same member name “the_point”

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

23

 

Technical

The return type for the application specific information (cost, maximum load, ...) is too generic in CadFoundation::Attributable::get_info

CadFoundation::Attributable:: get_info should use CosPropertyService::PropertyDefs as return type

DynAnys provide enhanced flexibility – can support this request

Steffen Nowacki, ZGDV

24

 

Technical

The in parameter type for the application specific information (cost, maximum load, ...) is too generic in Attributable:: set_info

CadFoundation::Attributable:: set_info should use CosPropertyService::Properties as in parameter type

DynAnys provide enhanced flexibility – can support this request

Steffen Nowacki, ZGDV

25

 

Funtionality

Unnecessary get operationes for EntityProps members in CadFoundation::Entity

Remove get operations from CadFoundation::Entity for values which are EntityProps members

Disagreed. Vendors want both access approaches.

Steffen Nowacki, ZGDV

26

 

Modeling

Missing Entity with EntityGroup semantics

(for entity type BlockReference in AutoCAD)

Define interface that inherits from Entity and EntityGroup

Semantics provided through ModelInstance Interface with embedded operation added to Model Interface.

Steffen Nowacki, ZGDV

27

 

Technical

union CadUtility::EntityAttrib has no advantages compared to an any

Remove CadUtility::EntityAttrib and use any in CadFeature::Parameter::set_value and CadFeature::Parameter::get_value

Disagree. union provides some convenience. However, any removed from union.

Steffen Nowacki, ZGDV

28

 

Technical

ModelInstance is an interface, but is only used to transfer data

Make ModelInstance a struct whith members location and component

Disagree. Entity inheritance is important.

Steffen Nowacki, ZGDV

29

 

Usability

CadGeometry ::BodyTessellation member

Object obj_ref

 

A more specific interface type and a

more meaningful member name should be used

 

Disagree. Narrowed reference will create an infinite recursion.

Steffen Nowacki, ZGDV

30

 

Usability

CadGeometry ::BodyTessellation has no suffix “Struct”

Declare and use consistent rules throughout the document, wether a struct name has the suffix “Struct” or not

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

31

 

Usability

CadBrep ::PropertyStruct member names (Surface_Area, Volume, ...)

Use only lower case letters and underscores in member names

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki, ZGDV

32

 

Funtionality

Missing a way to obtain all attribute values (standard + specific) from an Entity in one operation call

Extend the Attributable interface

Disagree. Team suggests that current structure is logically grouped and further grouping would not provide significant performance or Use Case benefits.

Steffen Nowacki, ZGDV

33

 

Funtionality

Missing a way to get the attribute values of many Entities with one operation call

Define an iterator, which can return the attribute values of many entities in one operation call

Recommend use of Visitor interface

Steffen Nowacki,

ZGDV

34

2.1.2.3

Editorial

#define CadConnectionFault is incomplete

#define CadConnectionFault  CadConnection::ValidationError, CadConnection::InvalidProperties, CadConnection::PermissionDenied

Agreed. Changes in mfg/2001-07-01.

Steffen Nowacki,

ZGDV

 

35

IDL,

2.1.3.2

Usability

The name “ModelList” is misleading

Because it is used for a list of model names

Rename ModelList to ModelNameList

Disagree. Name is also misleading – can be identifier. Prefer existing.

Resources:


Return to Technology Submission Evaluations

Return to ManTIS home


This page was updated on 7 February 2002. Please send comments and suggestions to [email protected] by email.

Last updated on: 11/09/2007