Informal notes on AR/VR discussion on Thursday August 3rd 2017 Present: Meurig, Steve, Mike, Alexandra, Ebtehal, Kim, Jake, Devon, Nick, Jonny, Richard C Meta-discussion What we know re problematic issues in design and engineering education - islands in the manufacturing process - aspiration to exploit VR - barriers to use of VR: cost, interface, pedagogical challenges / challenges to pedagogy [cf. Warwick report by Rob O'Toole] Perception that "technology is mature" and problems are no longer principally technological in nature Counter view: Importing the conceptual framework associated with digital technology into engineering has been very damaging The way in which computer scientists conceptualise construction is ill-suited/matched to engineering practice and promotes design paradigms not well-oriented to future needs Need a profound fundamental re-orientation in way in which we think about digital technology in order to do justice to engineering practices, and to realise its full potential Complementary to rather than consonant with the revolution that digital technology is initiating through automation, AI, big data etc. [That is a very important significant development, but not converging to the same goals where design and pedagogy are concerned?] cf. Humanities Computing Relevant observation conceptual ----- technical ?how to conceive this distinction [old] WHAT? HOW? declarative/functional procedural ... now influential in practice: [simple] functions as basis for dependency [Alluded to as FP, but conceptually remote from FP as originally conceived] Another instance of similar problem: FRep STL: "polygonal soup" [obstructive to effective practice] "can specify shapes that can't be realised in geometry / as physical objects" Never seemed plausible that the design paradigm promoted in CS classically: specify requirement then create/construct stage by stage is suited to engineering cf. Erkki empirical, observation-led "EM" ? To what extent does transition to "EM" orientation demand - new principles, conceptual perspective, enviroment/tools/instruments Argue that all three are necesssary ... A: conceptual frame - radical empiricisim B: principles - ODA perspective C: environment/instrument - Construit "MCE" A + B - prior work of the EM group C - work of the CONSTRUIT! project History: EM = BC fby AB fby C SW culture = C fby B Richard Cartwright "The modern sw development process is making construals" [media industry] SW development + prog driven by experience "How has this come about?" / "Will need for principles become an issue?" SW culture ODA ---> mechanisms to implement this are proliferating Bret Victor, 'EVE' Chris Grainger, Exploratory Explanations dependency features prominently "little functional programs" "What if?" in context of preconceived objects (bridge to classical CS) ? Is there any alternative: question begged by computational stance on universal meaning ? making construals is programming programming may in practice involve 'crafting experience enlisting meanings through association' but this is not licensed by any characterisation of programming as a principled activity -- even though 'it works' and in some contexts works better than anything EM has offered. This kind of activity is not what is consciously practised in 'agile' approaches for instance. association can be conjured without any presumption of purpose (cf. "free association") purpose might be seen as making sense / understanding Suspicious of whole complex of ideas around - specification for a purpose - interfaces for different users and levels of expertise - progressive disclosure of functionality Clarity about what something means is alien to making construals I don't know / I'm not sure / Let me see ... I'd need to think about that ... / I'm confused / I'm puzzled Something's not quite right / I don't understand cf. pedagogical frame in which the nature of what is to be learnt / understood / conveyed is clear and unambiguous cf. construal as object to think/converse with cf. hiding complexity from the learner until they are ready to understand ? preconceived semantic framework know what is to be explored / discovered [? "guided disccovery"] presupposing a level of confidence in the correct interpretation / the "right answer" that is at odds with the learner's personal construals and potential misonceptions cf. the primary school teachers 'dog' story as 'abstracted from experience' learning the nature of cricket from watching it on TV experience first - interpretation later {possibly} crafting experience and interpretation / applications 'open mind' Why invest in this? A+B not compelling enough C is essential Want to show that A+B with appropriate C delivers more than can be achieved otherwise ... AR/VR is an expensive process that readily exhausts financial and human resources -- as practised currently needs more investment than is feasible to realise the aspirations for the technology ... to resolve this beneficiaries / adopters / "users (!)" have to be engaged in co-design have to be leveraging domain expertise want not just to argue speculatively and abstractly that this is desirable but to demonstrate that this is feasible / plausible Qualities of the VR solar system construal economy of definition -- very few required comprehensibility can converse about observables with familiar associations collaboration with negotiation of meanings supporting ambiguity, concurrency, conflict, co-design authentication of undertanding 'reality' in line with the exalted notion of radical empiricism Possible studies - adding labels -- make conditional on "looking at the object" - Mars probe journey (Track to Mars -- #175) - anthropomorphic robot 'experience' / cricket match cf. BBC plans - 'super observation' how fast would Sun be travelling in pre-Copernican model of universe? 'visually' identifying nearest point of approach of Mars is it feasible to 'observe' (e.g.) distance of planet from Earth by looking / measuring?