Rhizo 14 : A Rhizomatic Learning cMOOC in Sunlight and in Shade

The authors present findings from the first stage of research into a “home-grown” connectivist MOOC, Rhizomatic Learning: The Community is the Curriculum (Rhizo14). We compare the surface view of the MOOC that has been presented in a range of open blog posts and articles with the view from beneath the surface that we have found in data we have collected (some anonymously). Our analysis reveals a positive, even transformative, experience for many participants on the one hand, but some more negative experiences and outcomes for other participants. These findings highlight the need for further research on the ethical implications of pedagogical experimentation, interrelated processes of community and curriculum formation, the role of the MOOC convener, and learner experiences within MOOC communities. In this paper we report on the alternative experiences of Rhizo14 participants and identify issues that we will explore in deeper analysis in forthcoming publications.


Introduction
Higher Education is in a state of flux: perhaps it always has been. Education is framed as a means of change through changing people and society, but is also the subject of change when it is characterized as "broken" and in need of transformation (Fullick, 2014). Will education be reformed, transformed or deformed, or a little of all three? Into this state of flux have emerged Massive Open Online Courses (MOOCs): first the connectivist MOOCs (cMOOCs) like CCK08, later the xMOOCs. With the proliferation of MOOCs since 2012, commonly referred to as the year of the MOOC (Watters, 2012), there has been an increasing output of published research into the MOOC experience (Haggard, 2013;Gašević, Kovanović, Joksimović & Siemens, 2014;Liyanagunawardena, Adams & Williams, 2013;Yousef, Chatti, Schroeder, Wosnitza & Jakobs, 2014). Much of this research has been into xMOOCs, i.e. MOOCs which have taken a traditional pedagogical approach to teaching and learning (using video lectures, readings, quizzes, tests, discussion forums) and moved it into open online learning environments. There has been less research on cMOOCs which were originally designed to challenge traditional approaches to teaching and learning by experimenting with a new pedagogical approach (Haggard, 2013;Rodriguez, 2012). This new approach was aligned with a proposed new learning theory, connectivism, which encourages learner autonomy, diversity, openness and interaction (Siemens, 2004;Mackness, Mak & Williams, 2010).
A recent cMOOC, Rhizomatic Learning: The Community is the Curriculum (now known as Rhizo14) took an experimental approach to teaching and learning. Designed with Deleuze and Guattari's (1987) principles of rhizomatic thinking in mind, it sought to challenge traditional conventions normally

Rhizo14-the context
Rhizomatic Learning: The Community is the Curriculum (Rhizo14) was an open course, convened by Dave Cormier in January 2014 to explore the possibility of open learning and provide a space for considering rhizomatic learning (Cormier, 2013). Officially the course ran for 6 weeks from January 14 to February 25. Unofficially it continued another six weeks until April 17; during this time course participants independently continued to discuss topics of interest both in the course site (P2PU) and on Facebook.
Rhizo14 attracted more than 500 participants. The exact number is not known; many participants did not formally register for the course. No demographic data was collected, but our participant experience indicates a diverse mix of people from across the world, from schools, further education and higher education and with different levels of experience of MOOCs and open learning (Mackness, 2014a).
The course was (fairly) massive, open access and free (no entry fee or barriers) and therefore a MOOC. It shared many of the characteristics of the original connectivist MOOCs (cMOOCs Twitter, a Google+ community, participant blogs, and cMOOC activities were promoted-aggregation, remixing, repurposing and feeding forward. cMOOC principles of learner autonomy, diversity, openness and interaction were essential to the course. Despite these similarities with the original cMOOCs Rhizo14 differed from them in some significant respects. The intention was that the community would be the curriculum. There were no course objectives and virtually no course content was provided (Cormier, 2014c). The course was designed around weekly provocative statements and questions as follows: Week 1-Cheating as Learning (Jan 14-21) Week 2-Enforcing Independence (Jan 21-28) Week 3-Embracing Uncertainty (Jan 28-Feb 4) Week 4-Is Books Making Us Stupid? (Feb 4-Feb 11) Week 5-Community As Curriculum (Feb 11-Feb 18) Week 6-Planned Obsolescence (Feb 18-?) These topics were not pre-planned but were chosen "on the hoof" in response to weekly discussions and posted with very short (av.3min) introductory videos (Cormier, 2014b). Participants were then left to create their own curriculum.
Rhizo14 also differed from prior cMOOCs in that it was "home-grown." Dave Cormier ran the MOOC in his own time, often convening the weekly Hangouts in the evening from his own home. Despite this, his intention was that there would be no centre to the course; he would be one of the participants.
In Rhizo14, Deleuze and Guattari's (1987) notion of the rhizome as a metaphor for thinking was used as a metaphor for learning. The metaphor affected the design and learning in the course, which could be regarded as being at the extreme end of the c/x MOOC spectrum, where "c" in this case could mean complex, chaotic, and/or connected.

Methodology
This research was to some extent precipitated by concerns that we, the authors, have about our own and other participants' experiences of learning in Rhizo14. As such, we are conscious of the dangers of finding what we are looking for in our research, as outlined by Stephen Downes in his presentation on MOOC Research (Downes, 2014). In outsider research where researchers see themselves as outside the researched situation, objectivity is a key element of the traditional sciencebased approach normally adopted. We were engaged in insider research, we became participant observers in Rhizo14. We acknowledge the danger of reduced objectivity and have taken measures to counteract it. We have also benefited from our subjectivity as researchers and the subjectivity of other participants who shared their views and experiences with us, being simultaneously insiders and outsiders (Dwyer & Buckle, 2009). Subjectivity can bring a significant contribution to research in complex situations involving people and their relations with material things and with each other, as we are "entangled in a web of relationships and practices" understanding agency as "a flow of forces in which the subject is continuously performed and performative" (Fenwick & Edwards, 2010, p. 21).
Our first steps were to declare that we were conducting research, and to engage with other participants (via a Google doc) on what would be ethical ways of using data in our research. Having consulted and planned our data collection, we shared this as widely as possible in all the spaces in which Rhizo14 was evident and on our blogs (Bell, 2014a;Mackness, 2014d).
Our research process developed organically. A representation of this is shown in figure 1.

Data collection
We participated in the MOOC as fully engaged participants and collated resources from the MOOC on a private wiki. These resources include Facebook threads, Twitter streams, annotated readings, discussion, survey results, links to videos and our own participant observation/reflection. Following the MOOC we created a Survey Monkey survey related to a botanical drawing of a rhizome from which we wanted to elicit qualitative rather than quantitative data. The survey included 4 questions: • How does the image of a rhizome relate to your prior experience of teaching, learning? • How does the image of a rhizome relate to your experience of learning during Rhizo14? • How might the image of a rhizome represent your future practice? • If the above questions did not allow you to fully explain your learning experience in Rhizo14, then please comment in the box below on those aspects of the course which were significant for you, and what kept you in the course or caused you to leave early.
We posted the link to the survey on Facebook, in the Google + group, on our blogs and on Twitter.
The link was also sent to all P2PU participants by Dave Cormier. In an attempt to ensure that we reached as many participants as possible, not only those who were still active at the end of the course, we identified non-registered participants and bloggers and sent them individual invitations to respond to the survey. Most importantly, the survey allowed for anonymous responses. It is difficult to know exactly how many people the survey reached, but we received 47 responses and more than 30,000 words of data. Following this initial survey we sent out further questions by email to 35 survey respondents who agreed to receive these follow up questions. These respondents were also asked an individual question which sought clarification of their original survey response (Mackness, 2014b).

Figure 1: Organic research process
The final stage of data collection was a Skype interview with Dave Cormier, convener of the Rhizo14 MOOC. A summary of the outcomes of this interview was posted at Bell (2014b).
As well as being mindful of and explicit about our roles as researchers, we are very conscious of the partial nature of the data we have collected and have analysed. The distributed nature of the spaces, the mix of public / private, and the number of survey respondents (47) combine to remind us that we must be missing some important perspectives. What does encourage us is that despite this partial view, our decision to allow for confidential and electively anonymous responses to our surveys, has enabled a light to be cast on what people are thinking, and not saying, in public and semi-public forums. This research will make a contribution to the hidden MOOC experience.

Literature that informs this research
Although the quantity of published research papers into learning in MOOCs is increasing, this research area is still in its infancy. Gašević et al. (2014) report in their review of MOOC research that much of it to date has been lacking in sufficient methodological and theoretical rigour. Liyanagunawardena et al. (2013), report a lack of reference to ethics in much of the research literature.
Given that many MOOCs, particularly cMOOCs, could be regarded as pedagogical experiments, i.e. trying out new approaches on participants, then ethical behaviours in relation to teaching and learning within MOOCs and consideration of ethics in MOOC related research must be part of a rigorous methodological and theoretical approach. With academic freedom comes responsibility (Marshall, 2014). This responsibility relates to both the collection of data from learners (Prinsloo & Slade, 2014) and the tension between innovation/ experimentation and a basic teaching and research principle of "do no harm." The vulnerability of learners must be recognized by both teachers and researchers (Markham & Buchanan, 2012;Barnett, 2007).
In addition to the need for a more ethical approach to teaching and research in MOOCs Liyanagunawardena et al. (2013) and Yousef et al. (2014) have pointed to further gaps in current research into MOOCs. Liyanagunawardena et al. (2013) suggest that the creator/facilitator perspective is not being widely researched. This was also noted by Kop, Fournier and Mak (2011). Yousef et al. (2014) note the research emphasis on "top-down, controlled, teacher-centred and centralized" MOOCs, i.e. xMOOCs. They say that "Attempts to implement bottom up, studentcentred, really open and distributed forms of MOOCs are the exception rather than the rule" (Yousef et al.,p. 16). More research is needed into learner experiences, especially in these "exceptional" MOOCs.
With respect to learner experiences, Gašević et al. (2014) have highlighted the importance of socialisation in MOOCs. This echoes the much earlier work of Wenger (1998) on social learning theory, the even earlier work of learning theorists such as Vygotsky, as well as the work of Garrison, Anderson and Archer (2000) on social presence in online environments. In the Homenet study, links between social isolation and Internet use dissipated over time (Kraut, Kiesler, Boneva, Cummings, Helgeson & Crawford, 2002). In online learning, social isolation is contextual to the course/learner experience (Haythornthwaite & Kazmer, 2001). In MOOCs, opportunities for establishing trust can be limited because of their shorter duration (Gašević et al., 2014). The question of what constitutes significant socialization and interaction therefore becomes an important area to research. Brinton, Chiang, Jain, Lam, Liu and Wong (2013) found that in the early stages of a MOOC much of the discussion is small talk or chatter that does not relate to course content and that this small talk is a major source of overload in the forums. This led Eynon, Hjorth, Yasseri and Gillani (2014) to question how meaningful and successful learning in a MOOC can be recognized and how "the Open Praxis, vol. 7 issue 1, January-March 2015, pp. 25-38 invisible rules that forum discussions encompass" are understood. Gillani, Yasseri, Eynon and Hjorth (2014) recognise the importance of group dynamics in MOOCs and that interactions at group level are important for learning. Their findings suggest that the structure of MOOCs and large-scale crowd-based learning can limit communication between learners. A small proportion of very vocal participants can significantly influence the discussion and "modularity in MOOC forum networks appears to 'trap' information in small learner groups" (Gillani et al., 2014, p. 6). Milligan, Littlejohn and Margaryan (2013) found that active participants are key to the success of a MOOC, but suggest that course facilitators should design courses that support a diversity of learners.
Current research therefore suggests that the learner experience and how learners interact in MOOCs is not well understood and the high drop out rate from MOOCs has been widely reported. This has implications for diversity (Mackness, Waite, Roberts & Lovegrove, 2013). As the numbers drop, diversity decreases and it becomes increasingly difficult for some learners to sustain their weak connections. Yang, Sinha, Adamson and Penstein (2013) have noted that MOOCs do not develop in the same way as better understood online communities, but tend to "grow in an unruly manner" and lack shared practices. The original cMOOCs, based as they were on connectivist principles, were never intended to foster the development of communities. Rather they emphasized learning in networks. At the time, Downes (2007) was clear about the difference between groups and networks and why the principles of connectivism were the principles of learning in networks. More recently, in a comment on Tony Bates' blog, Stephen Downes has written: "the two play different roles: the communities embed knowledge and standardize practice, while the MOOCs disrupt existing patterns of thinking and introduce people to new connections and new ideas" (Bates, 2014a). Wenger, Trayner and de Laat (2011) have also distinguished between the different purposes of networks and communities. However, Kop et al. (2011) highlighted the importance of social and teaching presence, support structures and the creation of community in MOOCs (see also Kop, 2011) and more recently, MOOCs from both ends of the c/xMOOC spectrum have been attempting to combine both learning at scale (the massiveness of networks) with community building. Examples of this are Coursera's MOOCs, Modern and Contemporary American Poetry (ModPo) and E-Learning and Digital Cultures (EDCMOOC), and Rhizo14. Jeremy Knox (2014), a convener of the EDCMOOC found in his research that if a MOOC is designed to obviate the need for tutor presence, then participants try to "replace the need for a teacher with the necessity for community" (Knox, 2014, p. 172).
If MOOCs are thought of as communities, how do educators position themselves in these learning environments? Ferguson & Whitelock (2014) have researched this and have found that educators in MOOCs outline "the trajectory of the course, acting as both host and instructor, sometimes as fellow learner, and often as an emotionally engaged enthusiast" (Ferguson & Whitelock, 2014, p. 563). Others have observed that the role of the educator in relation to MOOCs is changing (Masters, 2009;Johnson, Adams, Cummins, Estrada, Freeman & Ludgate, 2013). Educators in both xMOOCs and cMOOCs are taking a minimal intervention approach (Rodriguez, 2012). xMOOCs achieve this through distant charismatic, celebrity professors and automated responses; cMOOCs achieve this through teachers who adopt a facilitator and co-learner role (Ross, Sinclair, Knox & Macleod, 2014). Biesta (2013) claims that "teaching is more than facilitation and that teaching matters, that teachers should teach and should be allowed to teach" (Biesta, 2013, p. 36). But Ross et al. (2014) point out that what it means to teach and learn is as much a mystery as it ever was, never mind in a MOOC and measuring success in a MOOC remains an elusive endeavour. "There is more complexity and variation in the notion of the teacher than MOOC debates and literature have yet engaged with" (Ross et al., 2014, p. 61).
If the role of the teacher in MOOCs needs further research, so too does the meaning of the word "course" in particular in relation to content in a course. xMOOCs have taken traditional courses, with video lectures, readings, tests, a weekly structured syllabus and so on and put these courses online (Bates, 2014b). In cMOOCs the focus is less on the content and more on "how" to learn through networking and connectivity. Participants in these MOOCs play a significant role in content creation. "Teaching is subordinate to learning in a connectivist MOOC" (Ross et al., 2014, p. 60) and the community is the curriculum (Cormier, 2008). Despite this learners need some sense of coherence in content and conversations ) even if it is agreed that MOOCs are "a large public experiment exploring the impact of the Internet on education" (Siemens, 2012, n.p.).
Rhizo14 was such an experiment and provides an opportunity to explore and address some of the gaps that have been identified in current MOOC research. Of particular interest in relation to our research into the Rhizo14 MOOC are the ethical implications of experimenting on learners, the role of the facilitator, socialization, group dynamics and the impact of community on learning in MOOCs and how course, content and curriculum are understood in MOOCs.

The light and dark sides of Rhizo14
Rhizo14 was a connectivist MOOC. It fulfilled many of the characteristics of cMOOCs. It could be considered to be on the extreme "c" end of the c/x MOOC spectrum (Haggard, 2013;Rodriguez, 2012;Bates, 2014b) because unlike prior cMOOCs, the course was designed to have no centre. It was also designed to have no content or assessment; the community would be the curriculum. Dave Cormier was the course leader: his intention was "to invite a bunch of people to a conversation about my work to see if they could help me make it better" (Cormier, 2014b, n.p.). The course was about rhizomatic learning (Cormier, 2008). Ideas related to rhizomatic learning stem from the metaphor of the rhizome and some of the principles of rhizomatic thinking outlined by Deleuze and Guattari in their book A Thousand Plateaus (1987); principles such as Connections -a rhizome ceaselessly establishes connections and affords multiple points of entry; Heterogeneity -any point of a rhizome can be connected to any other and must be; Multiplicity -a multiplicity is, in the most basic sense, a complex structure that does not reference a prior unity and requires no central pivot point, being a-centred and de-subjectified; Asignifying rupture -if you break a rhizome it can start growing again on its old line or on a new line. Connections are constantly breaking (deterritorialisation) and reforming (reterritorialisation); Cartography and decalcomania -the rhizome is like a map and not a tracing. You can enter a rhizome at any point. Maps are always unfinished and subject to revision. These principles, whilst not discussed in relation to Deleuze and Guattari during the course, were nevertheless influential in the way in which the course was designed and experienced.
In these terms Rhizo14 was unlike all prior xMOOCs and cMOOCs. It was therefore an "exceptional" experimental course (i.e. an exception to prior MOOCs) and as such warrants focused research (Yousef et al., 2014).
Our survey and interview responses have revealed that for many participants Rhizo14 was a very positive experience. Many participants valued the metaphor of the rhizome for teaching and learning. Quoting from survey responses, participants of the Rhizo14 course thought that teaching and learning based on this metaphor is "subconscious," "subterranean," "subversive," "a non-linear, multi-directional underground web of connections." Learning is "haphazard," "messy," "serendipitous, " "esoteric," "dynamic," "unbounded," "unpredictable," "adaptive," "self-organising" and "non-hierarchical." In the words "The rhizome metaphor gives me a new way of framing education, exploring education, and thinking about education" (survey respondent).
Open Praxis, vol. 7 issue 1, January-March 2015, pp. 25-38 These participants valued the "lack of a centre," i.e. the lack of traditional tutoring and the lack of prescribed content. They valued the high emphasis on learner autonomy, self-organization and handing over control to learners. The course challenged traditional hierarchical modes of thinking about education and deconstructed formal rules of learning. They valued their perceived liberation from the aborescent structures of education institutions, classrooms and curricula. There was a sense of freedom that led, as one participant noted, to "movement, spontaneity and creativity," such that the course was experienced as a spirit of exploration, openness and experimentation. These were thought to be important aspects of Rhizo14.
For these participants and for the authors, as participant observers, this sense of excitement was palpable and was visible in activity on the web site and across a range of social media spaces, particularly the Facebook group which continued to remain very active for weeks after the end of the course and remains open. A survey respondent remarked "I became fascinated by really excellent discussions and I found the enthusiasm of some of the younger teachers and researchers infectious." For some the course promoted deep and wide learning, was transformational and had a positive impact on classroom practice. This was the "light" side of Rhizo14.
There was also a "dark" side for those participants who did not feel connected and could not find a voice in the community. Whilst the majority of our survey results were positive, the possibility for anonymous returns meant that the voice of participants who experienced the "dark side" could be heard. In MOOC research it is difficult to gain access to these voices since "open access" also means "open exit" and many participants "vote with their feet" and are difficult to contact following their departure. This has been noted by many researchers who have published MOOC drop out data (see for example, Yang et al., 2013;Jordan, 2014).
The "dark" side of Rhizo14 related to many of the gaps in MOOC research that have been noted by other researchers and referenced in the review of literature. Rhizo14 participants for whom the experience was less than positive felt isolated. They felt unable to make meaningful connections despite in some cases being experienced "MOOCers." One viewed the emphasis on community as an unnecessary pressure, which led to artificial effects, exclusion and limited learning. Another viewed the community as "disjointed networks of pre-established subgroups." Another described the community as having a "dark edge." These participants felt that there was a lack of appropriate facilitation, and that there were inappropriate exhibitions of power and politics in the course. Some felt that the course was based on weak philosophical foundations and that the rhizome is an empty signifier. Some questioned the lack of content in the course and felt that it lacked depth and theoretical discussion. For these participants the rhizome is "A pernicious, pervasive weed, rooted in a lot of dirt and "SH***""; " . . .a 'thug' and can be very badly behaved"; "Part of one big family/ plant-joined at the hip"; "Clones of the "same damn plant." One respondent wrote "I knew before that the arborescent paradigm was a problem. The rhizome is a contrasting alternative, but I learned in the course that this alternative has a lot of connotations with ugly and weed-like characteristics which are not necessary for every complex or even chaotic network" (survey respondent).

Rhizo14 as an experiment-ethical implications
Rhizo14 was designed as an experiment (Cormier, 2014c) and challenge to traditional ways of thinking about teaching and learning. There were no course objectives. In effect learners in this course were "lab rats," as was recognized by one survey respondent.
"I have felt very 'lab-ratty" at many different moments during rhizo14, and I mean that in a good way, and also in a critical way, meaning that I was aware that I was taking part in an experiment, a learning experiment. (. . .) For me, it was a journey within myself and how I function as a learner. I let myself be the 'lab rat', so to speak, to then take a step back and observe the big picture of the paths/directions I/rhizomes chose to follow/sprout towards, and that includes the people with which I most densely connected with and read and engaged most often" (survey respondent).
For this learner, being a "lab rat" was ultimately a positive experience, but as we have seen this was not the case for all participants. Another survey respondent noted that the psychological safety of participants should be of concern to the MOOC convener and wrote: "Clearly, my personal, admittedly tacit understanding of cMOOCs has been that they are intended to be a relatively safe (psychological and intellectual) space for differing opinions and world views. My tacit understanding is also that those who have more visible positions in cMOOCs, either through overt or indirect facilitation and "leadership," have both a high level of responsibility and a high (perhaps too high?) level of pressure to model and demonstrate, and perhaps even protect, this psychological safety and conceptual openness" (survey respondent).
The responsibility of the "teacher" was also discussed by Marshall (2014) in his paper on the ethical implications of MOOCs: " . . . we have a professional and social obligation to ensure that we are not abusing a position of trust and responsibility and acting, irrespectively of our wider goals and intentions, in an unethical manner" (p. 250).
The question of ethics in relation to MOOCs has been identified as a gap in the MOOC literature (Liyanagunawardena et al., 2014). Recent concerns have more often been related to the ethical use of participant data for research, but less has been written about the ethics of teaching in MOOCs. Marshall's paper is an exception (2014). In this he raised a number of further concerns about the ethics and responsibilities of teaching in MOOCs. These related to the ethics of convening a MOOC according to personal interest and thereby introducing personal bias; the responsibility for being alert to the potential for some group cultures to become disempowering for other participants; the obligation to provide a learning experience which is likely to be successful for all, and the duty of care that educators have for their learners.
These ethical considerations, which were also raised as concerns by some Rhizo14 survey respondents have been given very little attention in MOOCs, although the ethics of teaching has long held an established place in educational literature, as exemplified by the work of Noddings, who as far back as 1984 wrote: "The primary aim of all education must be the nurturance of the ethical ideal" (Noddings, 1984, p. 6). For Noddings the ethical ideal is an ethic of caring.
The "massive" of MOOCs has necessitated a change in the role of the teacher, but nevertheless "The teacher persists in the MOOC: though reworked and disaggregated, the teaching function and teaching professionalism remain central" (Bayne & Ross, 2014, p. 57). We would argue that well-established ethical principles also persist. Some have argued that teaching in a MOOC is a shared responsibility between teachers and learners (McAuley, Stewart, Siemens & Cormier, 2010). If this is so, then it follows that well established ethical principles should also be adhered to by all participants of a MOOC, as was noted by one survey respondent who listed six rules of engagement for teaching and participating in MOOCs.
1. Do no harm 2. The expectation is that interactions will be mutually respectful 3. Provide and allow space for reflection 4. Ad hominem attacks should not be permitted as a method of discussion 5. There should be a duty of care or necessarily emotional labour on the part of those calling together/convening/organizing/providing these amorphous spaces 6. All cMOOC participants have a duty of care and nurture and responsibility toward others or for themselves, mitigating the need or desire to externalize (blame) their learning and experience on others.
Open Praxis, vol. 7 issue 1, January-March 2015, pp. 25-38 These concerns were echoed by Lau (2014) who wrote of Rhizo14: One of the key lessons that I have taken from my MOOC experiences is that regardless of how participatory the learning experience is designed to be, it is worthwhile for MOOC instructors or facilitators to be mindful that participants are likely to look towards them for guidance on behavioural norms within a MOOC-and that they have both the power and responsibility to model attitudes and actions that support the full range of voices in a MOOC to be heard (Lau, 2014, p. 240).

Conclusions
The Rhizo14 MOOC was an experiment, an experiment which for some participants was very successful. It was innovative and challenged hierarchical and traditional ways of teaching and learning. It encouraged learner autonomy and openness, a participant co-created curriculum, co-creation of knowledge and community building. However, alongside exciting, inspiring and transformational experiences, there were others that were demotivating, demoralizing, disenfranchising and even disturbing. Many participants acknowledged Rhizo14 as a seminal experience that has changed their teaching and learning practice: some had mixed feelings and experiences; whilst others dropped by, lurked in the background or melted away, as they felt increasingly disconnected.
To an extent teaching is always experimental, a research process in action. As long ago as 1975 Robert Stenhouse described classrooms as laboratories. Almost forty years later a Rhizo14 survey respondent reflects on being a "lab rat." Education needs innovative, experimental approaches to teaching and learning, never more so than now in this fast changing digital age, but with innovation and experimentation comes increasing responsibility for learners. The will to learn is fragile (Barnett, 2007); a sentiment also reflected by a survey respondent: "You become a stranger to yourself so that you can later think yourself in strange terms. . ." There is a pedagogy of risk associated with treating teaching as an experiment (Mackness, 2014c). "There is an ethics of educational space, which has surely not been excavated" (Barnett, 2007, p. 146).
At this stage in our research, we can identify that adult learners, already immersed in formal education systems, can benefit from "doing" Rhizomatic Learning, forming community, and creating curriculum in a community setting. Educators can learn from their own experiences to change their professional practice but this is not always straightforward.
Our preliminary analysis has raised some issues that demand more detailed analysis and explanation if these benefits are to become available to a wider range of learners than those who experienced the "light" of Rhizo14. We are fortunate enough to have a wealth of data to inform further analysis.
In future writing, we will explore: • Interrelated processes of community and curriculum formation in Rhizo14 • The positive and negative effects of emotion and alienation • Moderation and leadership roles in the design and conduct of de-centred courses • Distributed spaces, technologies and services in a multi-platform MOOC • The rhizome as a metaphor for teaching and learning cMOOCs, to date, have tended to focus on the processes and practices of digital and networked teaching and learning and have attracted professional participants who work in the education sector Milligan et al., 2013). Critical analysis and reflection on cMOOC participant experience, which seeks to explore both the "light" and "dark" sides of learner experience in MOOCs, therefore becomes relevant to a wider constituency of learners. This paper reports on a first step to exploring these alternative perspectives and more complex issues related to learning in a MOOC.