Reply to thread

Message

<blockquote data-quote="clearstream" data-source="post: 8289624" data-attributes="member: 71699">I am presently accepting SP as a label, notwithstanding that I share your concerns for its possible hazards.That aside, having had cause recently to dig into &quot;Skilled Play&quot;, I suspect the label is being applied to two different modes of play. One mode is better labelled &quot;Gygaxian Successful Play&quot; (GSP, going forward). GSP is happy to use all of the rules to achieve the ends of the players. Spells like passwall and knock, and Find Traps checks, are by no means verboten to GSP. &quot;Modern Skilled Play&quot; (MSP, hereafter) is more interested in improvised, diegetic actions. A Find Traps check elides the fiction so has less appeal at an MSP table, where the DM is expecting a description of how the trap is searched for by the character in the imagined world. This seems to me foremost a matter of tradition, and the bats and silence discussion is very relevant here.That's not what we find when measuring player skill. Skill does indeed differ between challenges that call upon fundamentally different abilities, but skill is consistent across the same or related abilities. Bayesian algorithms produce good predictions of player skill at simple games even without using an AI-based approach. One up front piece of work is modelling your challenges - correctly identifying which abilities they will be stressing - so one would envision a cluster of AIs with various functions (a challenge-maker, variously skilled challenge-players, a player-ranker).</blockquote>

[QUOTE="clearstream, post: 8289624, member: 71699"] I am presently accepting SP as a label, notwithstanding that I share your concerns for its possible hazards. That aside, having had cause recently to dig into "Skilled Play", I suspect the label is being applied to two different modes of play. One mode is better labelled "Gygaxian Successful Play" (GSP, going forward). GSP is happy to use all of the rules to achieve the ends of the players. Spells like [I]passwall[/I] and [I]knock[/I], and Find Traps checks, are by no means verboten to GSP. "Modern Skilled Play" (MSP, hereafter) is more interested in improvised, diegetic actions. A Find Traps check elides the fiction so has less appeal at an MSP table, where the DM is expecting a description of how the trap is searched for by the character in the imagined world. This seems to me foremost a matter of tradition, and the bats and [I]silence[/I] discussion is very relevant here. That's not what we find when measuring player skill. Skill does indeed differ between challenges that call upon fundamentally different abilities, but skill is consistent across the same or related abilities. Bayesian algorithms produce good predictions of player skill at simple games even without using an AI-based approach. One up front piece of work is modelling your challenges - correctly identifying which abilities they will be stressing - so one would envision a cluster of AIs with various functions (a challenge-maker, variously skilled challenge-players, a player-ranker). [/QUOTE]

Verification