One tension running under several of the themes discussed is that between narrative and system, or in broader terms borrowed from structural linguistics, diachrony and synchrony, where diachrony refers to the temporal progression of history (the narrative of change) and synchrony to the snapshot of a language at a particular time (system). This tension underlies, for instance, the difference in opinion about the story missions in Battletech: one of the panelists said something very interesting in that the story missions seemed, to him, to be interruptions in the management of a space mercenary group. This would imply that it's not the quality of the story missions that is in question, but rather their existence at all. This panelist (sorry, I'm bad with names) would thus valorise the system (a synchronic set of rules that do not change and thus can be manipulated) over narration (the diachronic dimension which is change as such, by definition). There was also a brief mention of how the map in JA2 is not procedurally generated. This can again be seen in terms of the system-narration distinction: a non-procedural map in games can be understood as a narration projected onto a plane, because the player's experience changes as the map is traversed in a way that is ultimately to some degree arbitrary and not rule-bound, whereas procedural generation, even when it creates a map across which the player's experience changes, creates changes according to rules, and is thus still ultimately a product of system.
The interesting tension between system and narration perhaps speaks to a certain kind of historiography which has aspirations of becoming more like the rigorous sciences, a historiography that seeks to create systems out of the narrative of history.
(PS: Signing up for the Idle Thumbs network with "Three Moves Ahead" as the answer to the security question doesn't seem to work.)