# realtime mutiplayer modding garme future perfect future perfect by unknown worlds seems an important (if still far too small) step in blurring already blurred lines between play / players and development / developers // video here however it still seems based on several standard garmey gamisms such as real(r) gravity physics and (endless) and tedious tweaking - instead of being an 'enabler of play' instead of asking "how thick should this wall be" future perfect could be asking how the computer could be instantly generating multiple permutations of varied play scenarios; it's still far too based on static design principles rather than dynamic ones - ie. computationally emergent / fully procedural despite (eg.) 'being a fun garme' in itself one admittedly able to change on the fly tasking players with building the garmes they want to play in still generates essentially static spaces - mere '2d' levels and maps - bland stages or backdrops against which play takes place future perfect could instead be thinking in terms of processes rather than objects - of whole abstract systems instead of discrete isolated elements unable to communicate meaningfully with one another - that is artificially intelligent in short future perfect seems stuck an old development paradigm when it could be thinking in terms of an enoesque ecology of play metaphorically player-devs need the directness and simplicity of beach balls not merely the beaches upon which to play with them. project spark seems to have the same kind of limitation // video here // republic of bob