[Ccsds-omg-liaison] XTCE Reference Implementation?

[Adrian J. Hooke]

At 09:28 AM 2/7/2005, Brad Kizzort wrote:
>Harris Corporation is implementing a translator to allow XTCE input to its 
>ground system product.  That implementation is currently at an alpha 
>release level, with product release scheduled for later this year.  It is 
>being developed in conjunction with satellite database releases in XTCE 
>format from two different vendors.  However, I don't think those 
>implementations meet the requirement for "independent", since Harris is 
>providing guidance and feedback to those satellite groups.

Brad: the word "independent" is mainly intended to convey that the 
implementations should be done by different groups to avoid systematic 
errors, so your implementation would seem to qualify. However, the CCSDS 
interest is mainly in the interoperability of two independent 
implementations. Has any cross-organizational interoperability testing been 
done?

On that subject, in an off-list communication last week Scott Burleigh made 
an interesting observation, which I will try to paraphrase:
OMG abstractions define a service interface that can be realized by 
specific APIs written in lots of languages. This architecture protects the 
user from vendor lock-in: a mission can select any set of tools from the 
OMG menu, select an underlying communications service to run on, and be 
guaranteed that all of the OMG tools will be able to exchange information 
smoothly.  What the OMG abstraction does not provide is a common "wire 
protocol".  That is, missions or mission elements that have selected a 
given underlying communications system cannot interoperate with those that 
have selected a different underlying communications  system.
This reflects a fundamental (and probably complementary) difference in 
emphasis between what OMG aims at and what CCSDS aims at. CCSDS is all 
about "wire protocols" with little if any attempt to standardize APIs.  OMG 
nowadays is all about APIs -- POSIX-like -- with very little ongoing work 
on wire protocols.
Would you agree with this assessment?

///adrian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ccsds.org/pipermail/ccsds-omg-liaison/attachments/20050207/5eb54679/attachment.htm