xAPI-Spec Remove validation of target statement when voiding

The following line should be removed from the voiding requirements to allow for distributed systems:

"Upon receiving a Statement that voids another, the LRS SHOULD return a descriptive error if the target Statement cannot be found."

May 16 '13 13:05 dnjohnson

+1

May 17 '13 21:05 garemoko

Looks like this has to be a 1.1, though, as it actually changes meaning. Unless we have another spot in the document that is completely opposite of this point (and not just hinting at systems being distributed). I think we will see our fair share of these types of things that we all want, but bump the version to 1.1

May 20 '13 12:05 andyjohnson

This is just a best practice, thus can be a patch issue.

Oct 01 '14 20:10 andyjohnson

We are okay removing this language as this is not actually a best practice and no enforcement is done by the LRS for SHOULD. Make this a PR.

Oct 15 '14 19:10 andyjohnson

Since existing LRS might have already implemented this error, do we need to add something for 1.1 to say that the LRS MUST NOT error if the target Statement cannot be found, and MUST retain the voided statement in case it receives the target statement in future?

If it's not erroring, the client will assume the voiding was successful.

Oct 23 '14 13:10 garemoko

The interesting thing now is that the spec makes no determination on what an LRS SHOULD/MUST/MAY do when the target Statement cannot be found. The next version of the spec should address in details what we expect as voiding behavior. I think this deserves more consideration as you suggest @garemoko , so the issue will remain open as a marker for that LRS section as it is written.

Oct 23 '14 14:10 andyjohnson

@andyjohnson do you have a link to the forum discussion you mentioned in your email?

The latest discussion is on your PR here: https://github.com/adlnet/xAPI-Spec/pull/515 is that what you meant?

Nov 04 '14 23:11 garemoko

Ha! Yes I did! I thought it was a lost conversation, but it was just another issue :)

On Tue, Nov 4, 2014 at 6:02 PM, Andrew Downes [email protected] wrote:

@andyjohnson https://github.com/andyjohnson do you have a link to the forum discussion you mentioned in your email?

The latest discussion is on your PR here: #515 https://github.com/adlnet/xAPI-Spec/pull/515 is that what you meant?

— Reply to this email directly or view it on GitHub https://github.com/adlnet/xAPI-Spec/issues/327#issuecomment-61731342.

Andy Johnson ADL Technical Team 608-318-0049

Nov 05 '14 13:11 andyjohnson