Proposal 8 - Open issue - Sign for optional descriptors

that which may, or may not, be in version 2.0!
Locked
derpmann
Posts: 460
Joined: Tue Apr 24, 2007 5:32 am
Location: Volkswagen AG, Wolfsburg - Germany

Proposal 8 - Open issue - Sign for optional descriptors

Post by derpmann »

Which sign should be used for optional/additional descriptors?

In proposal 8 the following specification is used in the main document 5.1:
The mandatory and optional descriptors are described in the related electronic document A. Additional partner specific descriptors may be agreed between the transferring parties. They have to start with a “+” sign (ASCII 43).
Proposal 8:
In the main document we distinguish between mandatory and optional descriptors.
Both types are described in RED A, but the optional descriptors may be omitted.
The optional descriptors have to start with a ".".
All other descriptors agreed between the exchanging partners have to start with a "+".

Problems:
  • 1) Some descriptors are mostly optional but sometimes optional esp. for a specific channel or testobject type
    (see RED A 1.2.3.1 and 1.3.4.1).
  • 2) Is the descriptor 'Comments' mandatory or optional?
  • 3) If the attribute mandatory/optional is changed in the future, the descriptor name has to be changed.
Solutions:
  • A) We ignore the problems and accept the proposal 8
  • B) Optional descriptors (which are officially specified in RED A) don't have to start with a "."
    Additional descriptors have to start with a "."
  • C) Optional descriptors (which are officially specified in RED A) don't have to start with a "."
    Additional descriptors have to start with a "+"
  • D) I describe another solution
My preference:
I prefer solution C
Last edited by derpmann on Fri Apr 23, 2010 6:13 am, edited 2 times in total.
DiVe
Posts: 411
Joined: Tue Apr 24, 2007 9:14 am
Company: IAT mbh
Location: IAT Berlin - Germany
Contact:

Re: Proposal 8 - Open issue - Sign for optional descriptors

Post by DiVe »

Hello Peter,

in my opinion there is also one more fact to be taken into account:
For version 1.x we suggested to use a "." in front of any additional descriptor introduced by the exchanging partners.
So there are already existing implementations of descriptors starting with a dot. If we change this in the future to "+" and the "." will then be used in a different way (optional descriptor in the standard) this might confuse the users.

May I am overvalue this effect. In principle the change to V2.x will result in several significant advancements in the data format and so the change in the marking character of a descriptor will be of reduced importance.

Beside this concern I personally agree that the "+" character better expresses that some additional information is exchanged here.

If we decide to have a "." for optional descriptors in the standard, then the usage should not be optional. If we intend provide an indicator for what is optional (and depending on the application) with lower importance for the exchange, then the indicator must always be visible.
So we need to change the descriptor to ".Comment",.....
In principle we have also made a change of the same extend before, when we removed the leading "/" from the descriptors in the ISO EGV channels.

Bye
Dirk
User avatar
Paul Wellicome
Posts: 176
Joined: Mon Apr 23, 2007 2:20 pm
Company: HORIBA MIRA Ltd
Location: HORIBA MIRA Ltd, England

Re: Proposal 8 - Open issue - Sign for optional descriptors

Post by Paul Wellicome »

Open issue - Sign for optional descriptors

We discussed whether to retain the ‘.’ to indicate an optional descriptor. It was felt that this was confusing and that it should be dropped. Additional descriptors – those not in the RED-A – will start with a ‘+’ instead. We decide to adopt solution c) [Optional descriptors (which are officially specified in RED A) don’t have to start with a “.” Additional descriptors have to start with a "+"].

Mandatory / optional descriptors are indicated in the tables in RED-A.
Locked

Return to “MME 2.0 discussion”