Discursive Conclusion

Discursive Conclusion

Research Questions - March 2024

What pedagogical tools and processes are available to support novices to overcome barriers to participation in game coding processes?
How can game design patterns support the development of coding practices with novices?
How can learners build agency in an evolving community of game makers?

Outline notes

The working draft of this chapter now merges a proposed discussion chapter with the conclusion.

Initial Discussion Outline

Discussion around agency: Compare agency in design to different conception in research Tads sannino, relational etc
Analyse 3M design with this lens of agency
- Game design to understanding games as a design artefact for empowerment purposes
- Tension between walled garden and wider choice <!– - Compare key elements of 3M other studies in comping coding domain , primm, umc,
- shifts in levels of abstraction –>
- Start with motives? address transfer?
Synthesis of conceptions of designing for learner agency in an emerging community of coders
- compare Resnick / Paperts features of ‘designing for tinkerability’, with TADS and creating hooking points for double stimulation (warping anchors)
- note different emphasis on the agency of the learner to one focusing on curriculum knowledge and resulting reccommendations
- also compare concept of computational fluency of Resnick.
- differing focus on the role of the designer
- similarities in process - iterative, empowerment

Initial Conclusion Outline

Recap discussion around comparing aims of coding for empowerment/ learner agency with STEM pipeline narrative
- simple duality - critique of CT, pipeline narrative to fill vacancies in job market
- complexify simple duality interpretation, as authenticity and linking to real practices ARE empowering
Synthesis of 3M model
Recap analysis of framework to facilitate designing for an emerging community of coders
- Explore possibility to extend this to other domains
Final caveats / reflections on research
- Advantages and limitations of Home Education context
Recommendations arising from this research
- Recommendations for facilitators
- Recommendations for researchers
Related next steps for research
- Directly relevant next steps for research to address limitations
- Tangential next steps for research
  - Use of physical computing elements and embodied approaches

Diagrams / Illustrations

3 foci and 3M element compared - save for the conclusion/ discussion? https://docs.google.com/document/d/1FIPn2dEB2aCKZi5z05j69cnHgpGNXFh1_SPV6h5l12g/edit#heading=h.kci0lnvn7f60

Research outputs & discussion on outputs

Research output: 3M Game Making Model

In summer 2020 I began early dissemination of results and wanted to be able to present the essence of the learning resources and design to fellow practitioners and participants as well as researchers. To do this I created the frame of a 3M model, the aim being to simplify a complex set of resources and teaching processes. The resources have been created under an open licence (CC-BY-SA) which allows them to be freely used and adapted.

A recap of the key features of the model and the motivations behind them.

Figure 1.3. Synthesising diagram of 3M Game Making Model { width=80% }

Missions	Maps	Methods
Simple code changes yield quick feedback	A map of learning dimensions flexibly linked to main missions/patterns can be used by both learners and facilitators	Play Testing in each session aids short term motivation. Showcase events help longer-term motivation and aid project prioritisation
Free choice of Patterns increases learner engagement and ownership	Tracing the learner pathway on an attractive physical map in the learning space can help integrate navigation and reflection into the creative process	Drama and fictional scenarios can help explore issues and reduce learner anxiety though coding in a role
Restrict Game Type and number of Patterns to reduce Facilitator stress		Adding electronics to control the game via arcade buttons and cabinets increases engagement and perceptions of project authenticity
Limit complexity of patterns. Some are simple but cause a large change in the game
Side missions which explore and celebrate different ‘maker types’ (from Bartle’s player types)

Table 1.2. Key Features of 3M Game Making Model

Missions

The main challenge of the 3M model is to create a playable game around a theme for a real or imagined audience with learners given the choice to add many optional features to the game.

In addition, side missions encourage social and playful coding approaches which help develop a community of coders. The purpose of mirroring the language of adventure game experience is to create a playful environment, a theme explored in the following chapter.

Maps -

Learning Dimensions - in Phase 5 Physical maps in Phase 3.

Methods

varied inclusive methods.

Research output - Learning dimensions map

WAS Conflicts stemming from tensions between privileging participant choices or underlying computing and systems concepts

While the focus on this research is informal environment, I wanted to create resources and processes which are of potential value to practitioners in the UK classroom. Chapter two explored ways in which digital game making is commonly aligned with curricular areas [@kafai_connected_2016]. It also surfaced tensions that emerge between the exploration of curricular concepts participants desire to follow activities following learner expression and choice [@hoyles_pedagogy_1992].

This section addresses my interventions to resolve this tension. From my own perspective as a research designer/facilitator, there are two dimensions of contradiction which are explored in the following section: one my perception of the contradiction between curriculum and learner choice; the second a conflict which manifested via my hesitancy to introduce tools and activities designed to address this tension.

Learning Dimensions Map

Research on concept maps [@canas2003summary] as methods for teachers and researchers to identify key learning suited to particular computing projects has gained traction in the field of computing education [@keppens_concept_2008; @noauthor_quick_2020; @weinerth_concept_2014]. The value of facilitators presenting concepts and asking students to create maps of concepts for formative assessment. In addition, there is value in mapping which are likely to occur in project work especially when such concepts are embedded in activity in a way which resists superficial analysis. Research on hands-on tinkering in science museums by Bevan and Petrich [-@petrich_it_2013] together with educators examined video footage of families interacting with exhibits to make a structured list of the learning they observed. Their resulting map of learning dimensions included underlying science concepts but also contained more general skills and helping behaviours involved in exploratory learning processes.

Similarly, one of the outcomes of my research was to extract from hours of recorded material some of the concepts and practices that learners engaged with when making games. Table 4.x. shows my resulting map of learning dimensions. While some concepts are common to existing computational thinking frameworks others, including systems thinking concepts, are more unique to game making. The choice to include some elements over others is based on a synthesis of the experience of participants, existing frameworks in this domain (previously explored in the LR), and my own interpretation which privileged under explored elements like systems patterns. Due to the dominance personal input, this framework is offered as starting point which could benefit from the type of collaborative analysis undertaken in Bevan and Petrich’s process.

Coding Concepts	Systems Patterns	Design Practices
Sequences	Systems Elements	Goal Setting
Variables	Systems Dynamics	Being Incremental and Iterative
Logic	Reinforcing Feedback Loops	Developing Vocabulary
Loops	Balancing Feedback Loops	Web Navigation
Arrays		Problem Solving
Creating Functions		Version Control
Change Listener		Debugging
Input Event		Reusing and Remixing

To illustrate these concepts in more depth, I created concise descriptions of the concepts [^2], and cross referenced with examples of the concepts in practice in the application common game features[^3]. This process aimed to be an aid to my existing attempts to highlight relevant concepts in-situ.

In journal notes I note my attempts to find learning relating to abstract concepts happening in-situ. A common example of use of loops or arrays or functions. I would highlight their use by participants and link to their activity. This mirrors the concept of the value of abstract in context for wider application is underlined in the work Davydov of rising to the concrete [@valsiner_cultural-historical_2007].

The potential of learning and navigational maps

This process of mapping this framework is potentially of use for teachers and other researchers to help their observation of students in game making activities to align with design, coding and computational concepts. Offering student choice in project work can create stressful demands on a teacher trying to support a diverse design paths [@blumenfeld_motivating_1991; @cole_fifth_2006]. To address this issue practically, rather than students working on varied game types and features, I restricted the genre of game and offered a significant but limited menu of features, a process explored in the next section. The mapping of specific design features to the related concepts, as outlined above, has potential to aid teachers and students to track and correlate the underlying curricular concepts being explored in these divergent learning pathways.

The process of adapting this learning design to the needs of more formal settings can be aligned to existing research on this area. For example the concept of semantic waves is relevant here as a way of visioning the alternation between concrete design experience and communicating the underlying abstract concepts [@curzon_using_2020; @maton_making_2013].

However, despite this alignment and potential utility in a school setting, I was reluctant to introduce these resources in this informal context, reflecting the lack of external motivation for participants to align with curricular demands driven by exam requirements. Introducing additional mediational tools with the aim of the communicating and reinforcing abstract computing or systems concepts risked effectively establishing a new competing activity system. This perspective validates my reluctance to introduce additional distracting stimulus, instead prioritising time spent in on-going hands on experimentation, peer learning and playful activities. Additionally, while this approach aligns with situated understandings of knowledge which problematise the separation of the content to be learned from the embodied social activity itself [@rogoff_observing_1995; @leontev_problem_1974], this alignment did not halt my concern over remaining locked into concrete experience and not developing awareness of repeated abstract concepts which could be of future generalised utility [@maton_making_2013]. My ongoing discomfort surrounding the validity of the learning experience without direct alignment to abstract content, stem from wider concerns around avoiding danger of tacit learning by reinforcing learning and spur metacognition [@gherardi_toward_1998]. Even if generalisable concepts can be encountered in the context of an learner-driven activity, they may not be explored without guidance from others. Hoyles and Noss term this tension a ‘play paradox’ [@hoyles_pedagogy_1992, p.46]. This theme is explored in more detail in chapter six concentrating on social processes rather than external documentation.

Commentary / Discussion on Learning Dimensions

More discussion on the practical value of CT concepts

CT as a practical skill or a more abstract concept.

A general discussion of varied views and my own based on data of this study.

It’s fun but are they learning

Exploring the tactic of reflective practices to back up CTs being found in the wild. Both for practitioner and participants as per [@petrich_it_2013].

Discussion on UMC and PRIMM in relation to use of 3M model (Practitioner focus)

UMC - should also be covered in the Remix part of the design ?

Note on PRIMM and personal knowledge.

Recap on PRIMM ? - THIS WILL BE IN LIT REVIEW MOST PROBABLY

Da and Te - In these interaction the parent seems keen for the child to be able to express and demonstrate their personal knowledge of coding, games and maths concepts.

This is an interaction which in some ways is comparable to the PRIMM model which is suggested for classroom use. However to do it in the way demonstrated by this pair would involve a level of coding familiarity, and ongoing familiarity with the structure and thinking of the participant which would be difficult to achieve in many classrooms.

UMC on the other hand while being less described and open to interpretation seems to match the more informal, iterative game making of many participants.

Synthesising the characteristics and potential of the 3M pedagogy

BE CONVINCING HERE OF POTENTIAL OF THIS APPROACH.

3m This is a domain specific framework for teaching , can it have wider applications?

If so how can characteristics be abstracted?

And if so are they distinct from existing frameworks explored in literature review?

Walled garden of limited design patterns
Playful methods of community creation
Drama narrative to create audience a la heathcote
Regular informal feedback and chances to discuss evolving designs

3M through the lens of agency

In this section I summarises the potential for learners the concepts and techniques of achieving agency in the 3M design.

In PBL, UDL and other models - learner choice and building agency is shown to be of value in increasing engagement and ownership over learning. However, building the relevant knowledge to identify, agree on and implement game patterns from scratch takes time and without relevant scaffolding too so long that it negatively impacted motivation.

In a similar way to the play paradox [hoyles_pedagogy_1992], there is a tension between structure and freedom. However, the issue here relates to tensions between different forms of agency, instrumental, authorial and transformative.

TADS - interventions and affordances as double stimulation

Revisiting terms; The term intervention and affordances do not nearly align with the broad conceptual framework of this study infering a duality between subject and object. FIND REFERENCE

Backwards and warping anchors.

How does the process of offering out ledging points where anchors can catch, as a form of learning design match with conceptions of agency?

PROPOSAL Coding impact on design/ processes on learner agency - boiled down -

Answering RQ - on pedagogical resources / agency / use of GDPs

THIS IMAGINED FOR CONCLUSION BUT ALSO CODE TO IT? OR ABOVE?

learners developing agency through personal expression of home and other Identities
agency through choice over pathway via GDPS
agency through flexible design cycle pathways, especially via community playtesting
developing agency through celebration of different game maker styles
authenticity of tool use, especially docs and access to Community
authenticity of audience - and role of drama to facilitate reflection

This summary can be explored using conceptions of agency explored in previous chapters including instrumental, effortful and authorial agency. In particular the use of TADS and other conceptions of agency to shed light on the role of the designer /facilitator can be highlighted at this point.

Implications and recommendations for practitioners

This section is specifically oriented towards teaching coding in informal settings but also potential in classroom contexts.

The value and processes of creating space for emergent practices.

In the last chapter there was discussion on

metaphors for creating space, harbours, walled gardens

These metaphorical ideas have implications for practitioners that can be framed perhaps best as tensions.

know the territory but don’t pre-plan the route for your learners

Supporting different game making approaches via scaffolding within choice with limits.

A different strokes commentary. Depending on aims of the practitioner and context of the game making different tactics may be applicable.

At this point it is of value to return to one of the key challenge of a more bricolage approach to making and undertaking projects. How can you balance an approach based on choice of what to implement next, and the difficultly of participants navigating and overcoming blocks.

In chapter four and five we explored the value of a menu of tutorials and code snippets to support participant choice and a more active approach to learning which still contained the needed technical information.

This can be framed as Just-in-time learning [@riel1998education] but with certain modifications which offer additional scaffolding to reduce learner confusion.

Recommendations / Commentary / Discussion on toolsets

Some of the recommendations that this research makes have already been implemented over the lifespan of the research which were not in place at the start of the process.

Code playgrounds are now common and suggested by CS

However phaser while authentic has mismatches. Exposed by Belen, as I am using version 2 instead of 3. Thus the resources available are harder to navigate.

Alternative exist using the same 3M approach.

Further recommendations regarding tools sets

However, here are some more possible directions.

Make-code arcade has been created and adopted by educators. I developed this over lock down and gathered data on use which is available for further study.

P5.js perhaps provides a better base which is more aligned with learners needs. Recreating resources in P5.js is a strong possibility to mnake this approach more familiar and less porous / daunting.

Design patterns as a teaching aid

TO DEVELOP OR MOVE back TO GDP CHAPTER

See journal notes and contrast with existing research with GDP and informal / formal coding programming debates. Asking what does this research add to these debates?

Some of the following may be relevant;

Learner choice and levels of abstraction techniques
From the literature what are complexities involving the practitioner practicalities surrounding creating communities of coders
What is special about the use of GDP in this 3M model?
How is this model different from Game Jams in particular.
Whereas Game Jams focus on x, 3m has a focus on y.
Exploring GDP and different kinds of types of coding errors?

Game Design Patterns vs CT and more established “knowledge”

My end design highlights game patterns as a starting point for user interest and to draw participants in to other elements of learning dimensions. More abstract dimensions like systems thinking, and more concrete like coding concepts.

Compare formal teaching of CT and CS with an approach based on a freer choice of patterns.
Discuss the use of gdps as a gateway to other concepts as explored in CS in a more abstract, and more concrete direction [@eriksson_using_2019] .
Leading with GDP as a way to explore other Learning Dimensions
Discussion - The use of GDPs hits a sweet spot as a “middle” level construct, (see Eriksson and Bjork)
Discussion - Rising to the concrete - as a key AT concept - how does this fit?

GDP, CT & Transfer

The heart of the matter here is how transferable knowledge is without a contextual meaning attached. This is a contested area. Firstly how wide should reuse be imagined? To other coding projects? Or wider claims as with Wing’s interpretation of CT.

Technical of 3M concepts suiting transfer

Learning maps - additional discussion on learning dimensions - or conclusion / next steps.

While there are technical elements of transfer that were identified and included in the learning map I sidelined the study of them.

Specifically in examining the concrete implementation of GDPs there are transferrable concepts and practice involving practice which include;

Use of levels as a array which is parsed by a loop to create layout
Logic statements of varied degrees of complexity
Functions - code patching into the right area

Strengths and Weaknesses of the GDP construct / terminology

PERHAPS MOVE TO DESIGN? The interpretation of GDP that this thesis adopts contains some weaknesses. The specificity of the term is weak in that in seeking to categorise a wide diversity of features that participants would want to add to the game to fulfil a pedagogical motivation of providing guidance and inspiration to the novice designers. The pragmatic nature of this interpretation can be contrasted to more formal definitions of say Game Mechanics related to user actions in the game [@sicart_defining_2008].

The term is particularly week in the area of Game Polish which involved were more aesthetic and narrative choices.

[]

These choices do not fall neatly in to the argument being constructed around the potential of GDPs to unlock repeatable, transferable practices. For example, some participants, specifically one who came as a friend of of others and didn’t have direct parental help, spent a majority of time working on these patterns especially if assets are being replaced.

It is perhaps the term pattern that is problematic here. It may imply a repeatable structural element rather than an aesthetic choices. Other academics have given a wider name of game elements [@tekinbas_gaming_2014] or lenses [@schell_art_2008].

On CTs and CTs emerging in the wild.

As explored in a previous chapter, the 3M design leads with concrete game design patterns chosen by participants. This approach can be compared to others which lead with more abstract computational thinking concepts which are introduced and explicitly taught from first principles.

However, while not explicitly taught, the CT concepts are contained within the learning dimensions of the 3M model allowing participants and especially facilitators to be aware of examples of practice allowing for development of these skills. This section explores examples of CT in action that emerge from video data of game making.

NOTE - A quick summary of examples?

Compare formal teaching of CT and CS with an approach based on a freer choice of patterns.

The design choice of leading with GDP was based on increasing the agency of participants over the pathway of their design activities. LINK TO LITERATURE ON CHOICE AND ENGAGEMENT.

It is problematic to compare participants engagement with this learning design with another fictional game where they learn CT concepts first and then put them into practice.

However, it is helpful to identify key characteristics of the design which run counter to a principles first approach and describe how participants experienced these elements.

Elements of contrast with 3M include;

The remix element of code (explored in previous methods chapter perhaps) - which allows participants to add new features quickly. They can also see and replicate existing patterns in the code. This could compare to many hours writing from scratch and many abstract concepts explained before getting to anything looking like a game.
Lack of value placed on CT as overarching knowledge. No posters, no curriculum content, no tests on concepts, no lesson objectives and end tests,
Code patching from existing code examples, as previously explored, allows for participants to take code and add it into their game without full knowledge of the syntactical complexities of each line they use.

Synthesis of reflections on designing for learner agency

Pulling all together with recommendations from research . add in from notes

learners developing agency through personal expression of home and other Identities
agency through choice over pathway via GDPS
agency through flexible design cycle pathways, especially via community Playtesting
developing agency through celebration of different game maker styles
authenticity of tool use, especialy docs and access to Community
authenticity of audience - and role of drama to facilitate reflection

Thoughts Log - things that may end up here.

GDPs that are already in the template

Some GDPs were not made a part of the menu of options as they were already in the template. For example Reward.

Is this a missed opportunity to surface and discuss those patterns? cf [@eriksson_using_2019]

Breaking out of the walled garden:

Te and Da’s decision to explore a different starting point and add uncharted GDPs like following enemies. A potential tension which was resolved quickly in some ways. But at the cost of maintaining a working game.

Synergies of use of GDP by practioners and learners in research designs

The use of GDP by practitioners are compared to those of participants and synergies are explored.

On transfer of gdps - debates

In this context of novice coders in an informal environment it seems appropriate to introduce some levels of reusable practice whilst initially prioritising individual learner engagement and community participation. Game design patterns contain elements of more abstract and thus repeatable structures thus fulfilling some of the promise transfer offered by advocates of Computational Thinking. This is explored in more depth in the next chapter.

Synthesis of this research in relation to other research

How does this fit into other domain specific research - Constructionism, Computer supported collaborative learning CSCL

This section follows up on gaps in the reserach exposed in the introduction

Attempt to show differences and possible lines of inquiry that have been neglected in other areas.

Limitations of this research and potential areas for future research

Gender based issues of inclusion

This research did not address gender or racial issues directly. But further research would suit that.

The possibilities align with similar research, home interests etc. How will this impact pedagogies?

Missing data on starting game / quick start cards

Study on the use of templated starting cards was unfortunately hindered by the logistics of the recording process which began only after this process had happened and may participants had already undertaken one phase.

A separate study of a stand-alone group engaging with this starting template and cards would be welcome, and may be a legitimate follow up to this study.

Advantages and limitations of Home Education context

Comments include:

Reflecting back on interview there were very highly insightful comments from home educators that did influence the design.
Thus a very fruitful area for jointly created design research
However it is important then to not extend all conclusions beyond this setting but to extract the essense and try to seed in other settings using accelerated ways to construct and scaffold the kinds of support that home ed parents can offer.

Physical Computing and Game Making:

There is scant evidence for this in the video data.
I can make observations from journal notes and try to find mention of it in interviews.

(Imported Writing) Physical Computing and Game Making: The use of physical computing to create concrete and tangible activities can increase the engagement and motivation of learners [@kaloti-hallak_students_2015]. Making the digital concepts physical, and thus allowing exploration via diverse means, also aligns with inclusive learning principles. To support my game making projects, I created simple arcade cabinets out of wood with retro arcade buttons. Connecting arcade buttons to the computer via simple electronics is a project which can be completed quickly. The process of students building their own arcade cabinets for a games showcase increased their perception of the authenticity of their end goal. Some families created low-tech, customised arcade cabinets using cardboard. Although my studies have been small-scale the self-reported effects on learner engagement and motivation of this part of the program were significant.

Exploring CT development via the use of GDPs

Brief commentary on the enduring value of CT concepts and how they could be studied in relation to GDP pattern implementation.

Developing understanding of CT concepts in isolation is a potentially weak point of the 3M approach in comparison to from first principles approach. However the process of patching into different areas of the code and creating functions does address this. The following transcript of Sh’s interaction with myself shows that the process of copying patterns in the template code and patching in new lines from the supporting documentation gives ample opportunity for sequencing mistakes to be made. The process of working with students on their existing code is valuable to explain why and how the order of lines of code and when functions are called.

The comparison raises a potential discussion point on what the practical value of this CT concept is.

Using game design to explore games as a design artefact for empowerment purposes

Bruno advocates process that expose that are surrounded by designed objects. This process of making lifts the lid on that process.

See example between helper and Pa in 2019-03-18 at the end. “It makes you think” . They marvel at the complexity of code needed to create even a simple game.

The motivation to allow participants to get under the hood of new technology is a foundation of many technical programs built on empowerment. We can take Mozilla, mouse, etc.

It is also present in design programs which take care to convey ethical perspectives as part of the design process as a way to counteract the possible effects of unindended side effects. This is especially relevant in the age of big data and highly powerful corporations.

Yet this approach is not without critique.

“Battle not with monsters, lest ye become a monster, and if you gaze into the abyss, the abyss gazes also into you.” Friedrich Nietzsche

This expression, conveys a perspective that the medium is the message. By engaging with the technology its influences may permeate the subject.

This raises an additional question. How can we help learners build agency in technological environmental without becoming monsters.

How can the use of design patterns map to other computing educational domains?

HCI and webdesign already share a similar process at times. What about physical computing / electronics.

Can the same process help unearth accessible design patterns.

Comparing aims of coding for empowerment/ learner agency with STEM pipeline narrative

Recap discussion around comparing aims of coding for empowerment/ learner agency with STEM pipeline narrative
- simple duality - critique of CT, pipeline narrative to fill vacancies in job market
- complexify simple duality interpretation, as authenticity and linking to real practices ARE empowering
Recap limitations and contextual factors of computing in schools and diverse environments