#365 Possible typo in the value of the "persistent" attribute

[Roman Cheplyaka]

SBML Level 3 Version 2 Core, Release 2, 29 March 2019

Page 84, lines 17–18.

My understanding is that the purpose of the condition «that either (i) have Trigger objects with attributes persistent=“false” or (ii) have Trigger expressions that did not transition from “true” to “false”» is to ensure that these events are still scheduled for execution. In that case, shouldn't it say persistent=“true” rather than "false"?

[Michael Hucka]

Hi – that particular passage is about events that have not been triggered. What it's trying to say is, for all the events that didn't get triggered, there's a chance that the events that did get triggered change the priority of the untriggered events. So, after the triggered events fire, the priorities of the remaining events should be reevaluated before picking which event(s) fire in the next round.

[Roman Cheplyaka]

Dear Michael,

Thank you for the response.

First, I assume that by "triggered" you mean "executed" (i.e. their assignments got applied), and not merely that their trigger evaluates to true.

So what you're saying is that the executed events may change the priorities of the events that haven't been executed—that makes sense.

My specific issue with that passage is the specific condition on lines 17–18 that says «that either (i) have Trigger objects with attributes persistent=“false” or (ii) have Trigger expressions that did not transition from “true” to “false”». In order for an event to remain "triggered" (scheduled for execution) after the execution of the previous events, the event either has to have persistent="true" (not false, as the spec currently says) or a Trigger expression that still evaluats to true.

[Lucian Smith]

Moved to https://github.com/sbmlteam/sbml-specifications/issues/363