Jump to: navigation, search

<< Back to MoodleNet index


Meeting notes

Note: unless otherwise specified, all of these meetings took place between Doug Belshaw, Project MoodleNet Lead, and the person (or people) mentioned.

Martin Dougiamas (23rd April 2018)

Martin is CEO and Founder of Moodle. His Wikipedia page can be founder here.

  • We're going to run a Design Sprint with Outlandish, week beginning 21st May 2018.
  • Adverts for the Technical Architect position can now be posted. Martin wants to be involved at the interviewing stage.
  • We're definitely going with the name 'MoodleNet'. Martin is sure this is the best option.

Mike Larsson (16th April 2018)

Mike is a software developer with Iridescent. His Twitter profile can be found here.

  • Going to talk to former colleague about search stuff.
  • Google Custom Search could be used for the MVP, but could get expensive quickly beyond that. There's an API but they charge per use. Various monetisation options which get complex. Claims to confirm to OpenSearch specification. Can limit to Creative Commons licensed content, or certain aspects described by
  • Bootstrapping problem in that search looks like its necessary to get people using MoodleNet, but it might not be widely used after that initial flurry of use. Search potentially the hardest part of the project!
  • There's a difference between 'searching the internet' and 'searching the metadata of stuff that is referenced inside the system'. Need to do some more work on this.
  • Decentralised approach would mean that a user not only searches their instance of MoodleNet and the internet, but other federated instances.
  • Mastodon is based on ActivityPub which, in turn, is based upon Activity Streams specification. ActivityPub is something that Mastodon and other platforms can conform to (notifications-based) and Activity Streams defines the (extensible) vocabulary to be used. As a result, we could use the vocabulary of Activity Streams to define the functionality of MoodleNet, and ensure that the commenting functionality conforms to the ActivityPub specification.
  • We should probably build in, or have in mind from the start, federation and decentralisation. However, it's probably not a good idea to turn it on / flag it up until v1.0 (to prevent fragmentation).
  • Mike would need to do some research to see if there are any chat apps based on ActivityPub.
  • Google bought Firebase which is a platform to quickly build mobile apps and multiplayer games. There's chat functionality built in, as part of a 'serverless' approach / database schema.
  • With a design sprint, you can move more quickly with fewer people, but more likely to miss the mark r.e. what's useful.
  • The Thoughtbot playbook is a useful resource when planning early stages of tech projects.

Emma Richardson (16th April 2018)

Emma is President of the Moodle Users' Association (MUA).

  • Calling the project 'MoodleNet' seems a little narrow-minded when it could (should?) be used by anyone/everyone who's an educator.
  • The following should be included in MoodleNet:
    • Individual support options (paid services from particularly helpful Moodle users)
    • Better search interface
    • Forum-type functionality (Q&A, knowledge base, respond to threads via email)
    • Social network (casual conversational relationships: user/user, public, organisational)
    • Chat functionality (double-tick, typing)
  • Need to think about functionality and dependencies.
  • Would a modular approach to MoodleNet (and Moodle core) work best. Could take users on a journey to help them configure the Moodle that works best for them (technical / social /pedagogical)
  • Decentralisation: organisations would want to be able to have their own version of MoodleNet, which even if it was built off Mastodon/ActivityPub could still look however we want. That would then allow orgs to surface stuff of their own choosing, including proprietary, paid-for subscription content.
    • Would probably want to build in this functionality but turn it off until v1.0 to ensure development work isn't sharded.
  • Agrees that we need a separate Moodle-specific resource repository that can be queried/search via MoodleNet.

Rohan Hardie (16th April 2018)

Rohan is CFO/COO at Moodle.

  • Agreed r.e. design sprint --> prototyping --> MVP
  • Need to come up with lightweight business case that includes a well-defined problem statement.

Mark Pegrum (4th April 2018)

Mark is an associate professor in the Graduate School of Education at The University of Western Australia in Perth, where he is the Deputy Head of School (International), with responsibility for overseeing offshore programmes and international connections. His personal website can be found here.

  • People are already doing what Project MoodleNet proposes, so the ideas has been validated (Pinterest, Scootle, etc.)
  • The main question is how to draw people away from existing solution (especially Primary teachers)
  • Good idea to target teacher educators/trainers (like Mark) - student teachers have to demonstrate PLN (blogs, Pinterest, etc.)
  • Is MoodleNet the right name? The link with Moodle is positive as it's by educators, for educators. However probably best not to use given institutional politics (perhaps "powered by Moodle"?)
  • A fundamental question is whether Project MoodleNet is a separate project to Moodle core? Or is it for Moodle users?
  • Need to import content from as many platforms as possible (e.g. existing Pinterest collections)
  • Two-pronged strategy: students teachers / existing teachers
  • Social networks used by teachers, in Mark's experience:
    • Twitter
    • Facebook
    • LinkedIn
    • Diigo
    • ScoopIt
    • (+blogs, wikis, Wix, Weebly)
  • Australia has a new system of standards for teachers (effectively on probation for three years after qualifying). Teachers have to demonstrate ability to move up to next level through a portfolio of evidence (including attendance at courses and networking)
  • Perhaps focus on international schools and the IB? Connections with schools in other countries very important in that regard (esp. first-hand resources)
  • Reputation? LinkedIn-style endorsements plus comments on work you've done. Ensure highly-rated stuff floats to the top.
  • Social justice angle to this (flow of knowledge tends to be from global north to global south)
  • Gamification is extremely popular - badges? (pro-social contributions and collecting evidence for portfolios)
  • Language-learning apps (e.g. Busuu) - vote on best translations and get ratings.

Stephen Downes (3rd April 2018)

Stephen works in the Learning and Performance Support Systems program at the National Research Council Canada, which is a multi-year effort to develop personal learning technology and learning analytics. His personal website can be found here

  • Likes the idea presented for the MVP
  • Three elements, all of which have been tried before: curation, social, collections.
  • A Mastodon plugin for a Moodle environment could work, generating communities and sub-communities.
  • In general, people are looking for information within a specific discipline.
  • Self-organisation (e.g. tagging, taxonomies) can be done well, but also really badly.
  • Need to think about whether targeting existing communities beyond Moodle who may be interested (i.e. on social networks like Twitter, LinkedIn, and Google+)
  • A decentralised, open-source approach like Mastodon, or WordPress is great.
  • There's an overlap between Project MoodleNet and what Jim Groom is doing with a domain of one's own.
  • Data mining is a trade-off for features/services (so long as people are aware of it).
  • Separation between teachers and learners is imposed by the LMS - can get around this by focusing on 'exhaust' of OER from learning.
  • Howto videos, Codepen, Glitch are all good places to look at for inspirations.
  • Different disciplines support different things - e.g. History (videos), Business (case studies).
  • There's a big difference between using something and recommending something. It's the difference between what Stephen does with OLDaily and someone's Twitter feed. Next step beyond that would be mapping what Stephen features in OLDaily to what makes it into his papers/presentations.
  • Annotation is important, but so is reducing the workflow and number of steps involved in curation. If what we're doing is essential data collection, and we know the discipline and level, so we can do auto-tagging and classification.
  • A sidebar in the browser could show resources ranked highly by people like you (or your immediate network) - personal graph.
  • What gestures work in social networks? What is useful in this instance (e.g following people on Mastodon vs Twitter). What are the high effort / significant gestures?
  • Could rank according to:
    • Discipline and level
    • Social networks (connections)
    • Use
  • It's a chicken and egg problem with existing reputation systems, which will of course play out in Project MoodleNet. Need to give same capacity to disadvantaged groups. Take away the 'broadcast' feature?

Alan Levine (20th March 2018)

Alan consults with higher educational institutions on digital technologies and the affordances of the open web. His website can be found here.

  • Interest-based communities such as Thingiverse often have a 70/20/10 ratio around engagement.
  • The 'Moodle' name could be a problem for those who don't use the learning platform.
  • He likes the 'cupcakes' approach to development - not just another place to sit around and talk.
  • Badges probably aren't something for the MVP, more important to have a follow button.
  • Open hashtagging could work, as could suggesting a hashtag. Could people have *personal* hashtags for resource collections? Then they could be used a little like Amazon wishlists (i.e. "I might want this at some point in the future")
  • There is power in the default, so although you might want to allow people to create private collections, the default should be public.
  • GitHub profiles have an automatic 'activity' view, but users can also choose their five top / most interesting repositories. Could replicate for Project MoodleNet?
  • When you start with a search, you have to know what you're looking for. It doesn't allow for serendipity.
  • There should be a 'suggest this to other people' button. Also should be email-based, and you shouldn't have to sign-up to see what's in Project MoodleNet collections.
  • The profile as MVP dashboard is a good idea. Give people a choice around saving a search they've performed, watching collections, and configuring stuff.
  • Should talk with Randy Thornton, Moodle consultant.

Martin Hawksey (19th March 2018)

Martin is Chief Innovation, Community & Technology Officer at ALT. His website can be found here.

  • Stack Overflow badges are a good example of promoting the kind of behaviours you want within a community.
  • The ALT annual survey results show a trend of declining OER usage.
  • Martin was involved with the Into The Wild (2012) publication also mentioned by Amber. The SEO angle might be important.
  • How do you break the habit of people 'just googling for stuff' and going to your search engine?
  • Might be worth rewording the overview for Project MoodleNet to talk about 'including' open content rather than focusing on it?
  • Profiles are really important - c.f. CLMOOC. Allows for degrees of digital 'visitor' and 'resident' for individuals and allows them to aggregate activities from across the web. Martin mentioned that Known has been really good for that.
  • Discussion of using profiles as an 'MVP dashboard' led to Martin's TAGSExplorer as a way for communities to see people who might be several steps removed for them. Works as a recommendations engine to build the network.
  • ALT uses BuddyPress a lot, including for the OER18 conference (each session is a BuddyPress group)
  • Martin finds Drupal good for aggregation and viewing data in different ways. Need to think how much personalisation and customisations options to give users.
  • XtLearn allows users to collect and share resources and was developed by Xtensis with some grant funding. Might want to talk to Rod Pavey.
  • ALT offers CMALT (Certified Membership of ALT) which is portfolio-based. They're always looking to map this onto other frameworks and approaches to accredit prior learning.

Sander Bangma (15th March 2018)

Sander is Open Source Development Coordinator / Program Manager at Moodle. His LinkedIn profile can be found here.

  • The Open Badges experience in Moodle is sub-optimal. Need to at least upgrade to v2.0 of the specification.
  • Aiming for Moodle to be 'feature complete' by v3.7. Basing this on standards, including badges.
  • We must consider of what will work well for Core, Project MoodleNet, and Workplace.
  • Need a vision of how badges could be better in Moodle. Separate badging platform?

Alan O'Donohoe (14th March 2018)

Alan works for the Exa Foundation and is a well known Computer Science educator in the UK. His Twitter profile can be found here.

  • Description of Project MoodleNet reminds Alan of "the best features of some of his favourite platforms".
  • Alan spends a lot of his tie on text-based platforms supporting communities. He doesn't want to recommend Facebook, but it's often the lowest-friction way to help people. Compared this with the Computing At School community forum which is very different. People with the most experience aren't the most active, and the CAS forum only really works on laptop with browser (not mobile-friendly) - affects the kind of interactions you get.
  • Being on a social network is "like waiting for treasure to arrive".
  • Facebook has a reward/response loop which encourages positive behaviour. Removes the technical element entirely, so just focused on interactions.
  • Status roles are important within communities, especially in terms of the 'differential' between users. In some networks you can be a guru, whereas elsewhere you'd be a peer. There's a spectrum of engagement.
  • Alan's interested, as he thinks many people are, around analytics to do with his interactions in various communities. He's looking for metrics to encourage/guide him such as:
    • new followers
    • retweets
    • likes
  • Adding value comes through 'high quality' people, a great structure, and continually experimenting with new features/options (e.g. up/down voting in CAS forum)
  • The name 'MoodleNet' could be problematic and off-putting for some people.
  • Slack is pretty low-bar for people as lots are already on it and can log in with the same credentials. Just adding another chat. Compare that with Staffrm where you need separate credentials, etc.

Tom Woodward (14th March 2018)

Tom works in the Academic Learning Transformation Lab (ALT Lab) at Virginia Commonwealth University as the Associate Director of Learning Innovation. He was also involved in the success of DS106. His portfolio can be found here.

  • Will searching repositories be 'guided' or specific to the user in any way? Will there be different default views for various audiences
  • It took Tom ages to discover the term he was looking for - 'faceted search' which allows progressive restrictions on metadata. Example would be FacetWP for WordPress.
  • Need to automate some of the 'standards crosswalks' that are common in K12 education. Some companies specialise in doing this, but could do automatic tagging (e.g. KS3 = Grades XYZ)
  • Curated resources are a good idea, but some organisations / districts will want their teachers to be able to search for centrally-approced content. Project MoodleNet users shouldn't be restricted to individuals, but should be allowed to be organisations, too. Then, in search results, could prioritise 'friends/followers' collections at the top, followed by everyone's collections, and then finally everything in the network. Affinity groupings.
  • If integrating Project MoodleNet into Moodle learning platform through Jetpack-like plugin, then it needs to auto-tag users and resources in some way. Potential for LDAP integration? Although might lead to tumbleweeds/ghost town problem when it actually needs the 'snowball' effect.
  • Progressive onboarding is a good idea - i.e. invite people to apply for an account and give out 100 places with 10 invites each. See who applies as, for example, could create group of History educators, etc.
  • DS106 was successful because people were guiding those involed on what to do. Provided challenges. Could get Mary involved in this and link to badges and stealth assessment?
  • Gardner Campbell might be a good person to speak with. He's an English Professor interested in the broader philosophical stuff around tech, but gave a great presentation on the element of surprise in phpBB around levelling up. He worked with Jim Groom at the University of Mary Washington. Have a look at his talk around No Digital Facelifts
  • Patreon is becoming more of a thing - see Alan Levine's recent post
  • The people-centric vs. resource-centric social networks approach is a good one. Can also think in terms of proactive/reactive and streams vs. one-off searching. Ultimately, people need the right language to be able to search for stuff.
  • A continuous challenge will be how to promote the right 'intimacy' of conversation. Also a challenge around languages: is it going to be one big thing (all the languages together) or forks of the same thing in different language communities? Could do auto-translation between them? (see WPLANG)
  • Back to DS106 and another reason it was successful was the transfer of power as students were creating content. Reputation was based on doing the work instead of your title.
  • Twitter chats are valuable because they're synchronous.
  • The subreddit for Picture Requests is an interesting example of an active community around communal work.
  • Other links shared:

Phil Barker (14th March 2018)

Phil works with technology to enhance learning and creates information systems for education. He's particularly interested in supporting the discovery and selection of appropriate learning resources. Much of the work I do is with Cetis LLP, a cooperative consultancy for innovation in educational technology. His personal website can be found here.

  • Work Phil's doing with for Credential Engine.
  • Look back through Jisc's UKOER programme resources.
  • What would a useful search experience look like for users? There are different approaches to describing the same thing (metadata).
  • Adding things to collections is an interesting/useful way to approach quality - i.e. if in a collection of a trusted/well-known user, then quality stuff. Could add badges?
  • Check out Open Edinburgh
  • The OAI protocol is an older standard, but could use Dublin Core? Where there is metadata, we should use it.
  • Collections could have discussions around them - i.e. a threaded comment section which is public by default. Users could add tags to collections, or suggest them for the collection creator to add. Particularly useful if, for example, created by UK educator and need to add US-related tags?
  • Could standardise age-appropriate levels (e.g. Pre-K, K6, K12, Post-K12) and colour-code resources?
  • Potential for more than one dashboard depending on number of interests, things to follow, etc.
  • Phil isn't sure he'd use for profiles - more for resource descriptions. Talked about Bayesian filters for spam and early machine learning. Involves a trainng process a bit like thumbs up/down for Spotify playlists. Surfaces obscure/long tail stuff.
  • Should talk with Joe Wilson
  • Crowdfunding seems like a good idea as it increases motivation to create resources. Could be particularly useful for 'less sexy' stuff like security/maintenance of courses and resources?
  • Phil could potentially offer consultancy around interfacing with repositories (c.f. Learning Registry and metadata.
  • A project at Heriot-Watt University recommended where users should upload resources, but also kept a copy of the resource centrally.

Bryan Mathers (14th March 2018)

Bryan is a visual thinker who started working life as a Software Engineer. He's founded, grown and then sold companies, and is a founding member of We Are Open co-op. His personal website can be found here.

  • Could visually represent the three approaches to building the MVP as a technical drawing or graphical projection (i.e. front/side/top)?
  • Is a landscape study of peer-reviewed academic articles really necessary to start building?
  • Productive ambiguity is OK, but not too much. So not "Pinterest for educators".
  • Twitter built an MVP in two weeks. What features are absolutely essential for Project MoodleNet?
  • When Wapisasa was building their 'Central' product they used Google Sheets for an agile approach to planning focused on user stories. Tried to make thigs as simple and shareable as possible and worked in two-week sprints. Eventually moved to GitHub. Used 'ROM' (Rough Order of Magnitude) with each issue in terms of granularity.
  • Asked Bryan to create three drawings:
    • 3 'cupcakes' (approaches to building)
    • An overview of the workflow for the resource-centric approach
    • A three-layer cupcake (cake, icing, cherry) to represent: (i) fiding/curating, (ii) discoversing other peoples collections, and (iii) the social networking layer.
  • Could actually order cupcakes for the UK & Ireland MoodleMoot session?
  • Using vus.js to do some interesting stuff at the moment.

Gavin Henrick (13th March 2018)

Gavin is Business Development Manager at Moodle. His personal website can be found here.

  • Authentication (identity) is key - both inside Moodle and outside Moodle.
  • Need to think about what the forums aren't providing at the moment.
  • Just-in-time help for academics and course builders is key.
  • Gavin uses a modified version of PRINCE2 for project planning (summary, risks, etc.)
  • Project MoodleNet needs to ultimately be within Moodle to reach educators and not just techies. Moodle admins will need a reason to turn on the integration (e.g. reduce support costs)
  • Should talk to Mark Brown from DCU.
  • Tweetchats seem to have replaced forums - is that to do with removing status/hierarchy issues?
  • Need credibility from educators, so should get Mary Cooch involved
  • What needs to be decided r.e. tech choices?
  • The tech lead we hire needs to be awesome at authentication/identity systems. Also we should further investigate Gluu.
  • Core of Project MoodleNet is login and profile.
  • IMS have a Learner Information Package Specification that we couldd potentially use.
  • Check the workflows of how people interact with the proof of concept. Call it 'open research'.
  • Should ensure Project MoodleNet integrates with Moodle Cloud
  • How are people going to access Project MoodleNet? Native apps? Progressive web apps? Notifications are key.
  • Revisit some of the functionality of Elgg and potentially talk with Ben Werdmuller.
  • looks useful as an example of the kind of thing we're trying to do with the first part of the resource curation workflow.
  • Could use AI around metadata to suggest tags within a wider folksonomy.
  • The Caboodle projecct was a proof of concept that allowed learners to search for resources within Moodle.
  • The OAIPMH (Open Archives Initiative Protocol for Metadata Harvesting) protocol could be useful.
  • Encourage developers to create interesting ways to display data and create new clients that interface with Project MoodleNet. It adds value to the ecosystem.
  • Wouldn't recommend building out profiles from existing ones on, but could add a new block for Project MoodleNet stuff?
  • Profile as MVP dashboard - add RSS/APIs.
  • Gavin suggests using Drupal for mockups through to MVP. Two Moodle partners are Drupal experts and could help. Can demo Drupal via Acquia Cloud
  • Three layers to sharing (me / friends / public)
  • Gravatar is a useful touchstone, not only because of auto-attaching avatars to email addresses, but also because of the profile card that it creates.
  • Degreed has just had a large investment and focuses on 'show your expertise'.
  • Should ensure great experience for new users. Perhaps hire people to create/curate stuff to begin with?

Amber Thomas (13th March 2018)

Amber is Head of Academic Technology at the University of Warwick. Her personal website can be found here.

  • Amber's team includes Kerry Pinny who is leading Moodle-related stuff at the University of Warwick.
  • Jisc has recently launched a NGDLE report which may prove useful, as might a 2012 report Amber wrote with Cetis called Into The Wild (gives models of resource discovery)
  • Paradata (i.e. activity data related to resources) may prove particularly useful.
  • The JLeRN experiment was about deriving useful information in a passive way - i.e. building services based on data users had consented to share and be processed.
  • Should talk with Martin Hawksey
  • What's the problem we're trying to solve with Project MoodleNet?
  • The Jetpack metaphor made Amber think about LinkedIn and how it integrates with Microsoft (which now owns LinkedIn) has been marketing end-to-end solutions to universities with different models of membership. Offering contextual help.
  • What's the underlying business model of Project MoodleNet? Need to build it differently if users are paying (vs institutions)
  • Need a critical mass, so should choose area to focus on initially very carefully.
  • Clippy / voice assistant metaphor is essentially providing 'just in time' contextual information.
  • Profile element reminds Amber of which started off as very promising, but then pivoted towards something a bit creepy.
  • Piirus took a 'dating' approach to professional networking for early career researchers. It was acquired by and pretty much shut down.
  • How stable is the Moodle community? Are there particular points when people need to create/build a network? Can be victim of own success (i.e. if you bring together the right crowd, they don't need the 'matchmaker' in the middle any more). That's OK, though, if they leave a trace - as with a forum when there's a history of interactions and being helpful for those coming afterwards.
  • Forums vs. social networks? (need both)
  • What's the value proposition / service design? Connecting to resources and connecting to people. Users need a mental map and an incentive to be part of the community (otherwise it could just be newbies talking to each other)
  • Handy to connect in Twitter (potentially via IFTTT)
  • Crowdfunding is something Amber could see herself using, particularly around CPD resources or Moodle design patterns.
  • "One man's expert conversation is another man's n00b question."
  • Tension: those people with the deepest knowledge/expertise have the least incentive to share/help. Could a Patreon model help with this? (esteem + £££).
  • Should talk with Josie Fraser
  • Could create a marketplace of content packages that can be rated.
  • Canvas Commons was useful (but seems to be slowly dying? not much activity)
  • If the business case is focused on creating social good, then perhaps think about interventions in other VLE communities?

Rohan Hardie (13th March 2018)

Rohan is CFO/COO at Moodle.

  • Need for a technical lead to be promoted internally or hired externally:
    • Big-picture thinker
    • 'Full stack' developer
    • Prototyper
  • Project management approach
  • Upcoming conference/events to attend (including OE Global 2018)

Kristina Ishmael (12th March 2018)

Kristina is a Public Interest Technology and Educational Policy Fellow at New America. Her Twitter profile can be found here.

  • Absolutely right to be focusing on curation and discovery, as Kristina has found in her work that's what educators are missing. Only really got OER Commons
  • The Go Open initiative is worth paying attention to. It's starting with small, targeted interventions (e.g. replacing a single textbook with OER). Techies were disappointed not to also have a commitment to open source, but one step at a time.
  • Should talk with Peter Nelson who's based in a school district and is building an OER search engine.
  • The Hewlett Foundation funded a lot of the Go Open work, including Kristina's as a New America fellow. They have a new Education program director in Kent McGuire who is carrying out a strategy refresh. The Hewlett Foundation is trying to find more diverse voices. Kristina reached out to the NAACP but without success.
  • Other angle around equity is focusing on those in rural locations. Smaller systems can be nimble, but larger rural states in the US need some attention too.
  • Kristina's fellowship runs out at the end of November after 18 months. Working on four case studies after convening people in DC earlier this month. Talked about the US MoodleMoot in Denver at the end of October.
  • Angela Debarger, OER program manager at the Hewlett Foundation, wrote a blog post about equity and diversity in future funding work. Worth a read.
  • The Gates Foundation is also restructuring to focus on equity.
  • Lisa Petrides at OER Commons is a good person to connect with.
  • Can potentially connect in-person with Kristina in Delft at the OE Global Conference 2018

Meg Goodine (8th March 2018)

Meg is Manager of Learning Technologies at Kwantlen Polytechnic University. Her team's microsite can be found here.

  • Using Open Badges in Moodle through a CanCred Factory plugin.
  • Issuing badges for information literacy, which are triggered on course completion. Badges are displayed on Moodle profiles, and can be pushed to CanCred backpack / Mozilla backpack (although no-one really is)
  • These badges are for compliance, so not a great pilot.
  • Chose CanCred because it allows people to see what's available without having to login to Moodle. Can apply for badges. Canadian servers, which complies with stringent British Columbia privacy legislation.
  • At the IMS badges summit the Badgr team was demonstrating badge pathways.
  • Kwantlen Polytechnic University is "really into open" and is an open access institution.
  • Looking forward to using badges with v2.0 of specification.

Kin Lane (8th March 2018)

Kin is a technology professional with an obsession for Application Programming Interfaces, also affectionately called APIs. His personal website can be found here.

  • Kin knows people at Thingiverse if we need a contact.
  • The Jetpack metaphor works.
  • OpenID Connect is the right choice for authentication.
  • Open source social networks come in waves, get lots of buzz then disappear. In general, people are looking to create their own network effect.
  • Every API suffers from sandards - need a common language, but vendors like to throw their weight around.
  • is a good choice for profiles, as is Gravatar. You need the same vocabulary, if not the same tooling. Give the schema/specification to enable modules to be built.
  • Kin is going to have a '30 min glance' at the components listed for Project MoodleNet and see what's available in terms of standards and specifications.
  • "I don't believe in code" - comes and goes, unlike the schema behind it. Good idea to focus on W3C specs and bet on the open web.
  • Everything should have an API. The OpenAPI Initiative is a good place to start as it allows you to describe the surface area of an API and focus on users.
  • Good idea to adopt existing schemas and specs and focus on inputs/outputs.

Mike Larsson (8th March 2018)

Mike is a software developer with Iridescent. His Twitter profile can be found here.

  • Federated aspect / JetPack metaphor makes sense.
  • Availability for technical reviews / consultancy.

Scott McLeod (7th March 2018)

Scott is an Associate Professor and CASTLE Founder. His personal website can be found here

  • Curriki - sponsored by Cisco and like Teachers Pay Teachers but free.
  • Is the name 'MoodleNet' the right one? It sounds like it's just for Moodle users.
  • Amazon Inspire is an OER repository. Why would someone use Project MoodleNet instead of just going there?
  • Is Project MoodleNet going to use existing ratings (users / resources) and collections from repositories (e.g. via the API). Annotation layer?
  • From a professional development point of view, this mainly happens in the US through courses that are part of state-wide learning systems. They include a credit-tracking system linked to diagnostic tests which discover gaps for teachers to learn new things. This then matches them up with vendors supplying courses. Could include in Project MoodleNet?
  • Portfolio systems have curriculum standards embedded - e.g. LiveText by Watermark
  • Scott uses IFTTT but it h as limitations Could use bots? AI?
  • In terms of searching, what's a good metaphor for how Project MoodleNet will work? Google Custom Search approach where search happens in certain domains first, and then the whole web?
  • Need to talk with US Dept of Education and their 'Go Open' initiative. Good contact relating to that is Kristina Ishmael.

Martin Weller (7th March 2018)

Martin is Professor of Educational Technology at the Open University. His blog can be found here.

  • What's the problem to solve here? Is that clear?
  • Professional development is a process, not an object. You 'get it' by doing it. Twitter is a bit like swimming in that you just have to jump in and start doing it.
  • Other than Twitter, the early edublogosphere and forums were where comunities were at.
  • JetPack metaphor works, but pitching Project MoodleNet as 'Thingiverse for education' looks like a promising approach. Be aware of what's gone before such as Ariadne (meta-search), OER Commons, OER World Map
  • Main differentiator between projects like this that succeed or fail seems to be branding/UX. In a good position as 'critical mass' through existing Moodle community.
  • In terms of making the case, Martin has used the OU's financial crisis to talk about 'efficiency savings' of using OER. Particularly good case to make if authoring tools and search engines are combined. People need to be able to surface stuff they didn't even know they needed (c.f. Google Docs research mode)
  • The Blended Learning Consortium have an interesting model: colleges pay to join the 'club' (around £5k/ year) and then people are employed to create and curate resources.
  • ICDE (International Council for Open and Distance Education) is lead on a new EU bid called PANOR. About discoverability of OER and bringing communities together.
  • OER Commons is working with the US government on a Go Open initiative.
  • The Open University have relaunched OpenLearn Create which allows users to make their own online courses / MOOCs. Providing the tech infrastructure for others to create is a powerful thing to do.
  • Should talk to:
    • Creative Commons (Cable Green)
    • Open Education Consortium (Paul Stacey)
    • David Wiley (good balance of Open Education and commercial stuff)
    • Andrew Law (Open University - was head of OpenLearn until recently)

Mary Cooch (7th March 2018)

Mary is Moodle's Community Educator. Her personal website can be found here.

  • Ultimately, Project MoodleNet has to be the place for idea and resource-sharing. This boils down to courses and resources.
  • The Edmodo community seems to be sharing resources back and forth all the time. The community is great, but different in that regard.
  • Of around 50 courses submitted to the existing every week, only one is likely to be accepted. This is for a couple of reasons:
    • 90% are shared mistakenly (publish/share button used incorrectly)
    • Poor quality of those intentionally shared
  • Unity / Open Academic Environment looks promising
  • is old and only works for a small group of Moodle users. Even at Moots, people suggest continuing conversations and sharing stuff via WhatsApp instead of the forums.
  • The 'default view' for Project MoodleNet should be something like a dashboard where users can see people with suggested ideas/resources. Community activity.
  • Suggested talking to Alan O'Donohoe

R. John Robertson (6th March 2018)

John is a Digital Education Librarian at Seattle Pacific University. His LinkedIn profile can be found here.

  • JetPack metaphor works
  • How does Project MoodleNet with hosted Moodle instances? What about partners?
  • Leading with the social element of Project MoodleNet could be challenging
  • Makes sense as an aggregator: "OER repositories have a horrible half-life"
  • Metadata on resources is never going to be 100% correct, so just have to deal with that.
  • Seems like a good candidate for Open Badges
  • Giving people money (i.e. crowdsourcing element) can be tricky based on their contracts (MoodleCoin?!)
  • Why wouldn't this just go away when the next shiny aggregator comes along?
  • Canvas' community has no OER aspect to it, but instead includes the following (third particularly useful):
    • Asking questions
    • Social profile
    • Feature request voting
  • Perhaps include an annotation layer and/or paradata? (see W3C annotation spec
  • OER course shells (i.e. sharing entire courses) can work when people need a significant number of resources and sometimes this is a quicker/easier way of sharing.
  • Recommends talking with Phil Barker

Bryan Mathers (6th March 2018)

Bryan is a visual thinker who started working life as a Software Engineer. He's founded, grown and then sold companies, and is a founding member of We Are Open co-op. His personal website can be found here.

  • Current number of registered users:
    • 125 million registered Moodle users overall
    • 2 million registered users at
    • 40,000 active users at (2%)
  • Difference between Open Academic Environment and Mastodon is that the former is by default private, and the latter is by default public. There has to be a need/reason to make something public with OAE, unlike Mastodon.
  • Why is Slack the default way for new communities to communicate? Other options (open source)
  • Are forums the same as social networks? Do we use them for the same purposes? Twitter is like a cafe where anyone can pull up a chair. Forums are like a dive bar where the music stops when you walk in!
  • There's huge value in going to where people are. How could Project MoodleNet use something like IFTTT to meet people halfway?
  • Perhaps what's needed is an email-based system? (e.g. email DMs)
  • Daily digests are useful (c.f. Google Groups / OLDaily / - how many people are getting email digests from the latter without logging in and creating an active session?)
  • How do people want to be seen/notified? (e.g. Slack turn off email notifications when you install the mobile app)
  • Voice and tone is importance (e.g. MailChimp / Slack / Virgin group) MailChimp has a dedicated website on this.
  • What are the actual needs of Product MoodleNet vs the hunches people have about what's needed?
  • The JetPack metaphor needs refining - perhaps just refer to as an uber-plugin?
  • Should commit to milestones, not dates.
  • What are the experiments that could be done to avoid pain in the future? How could they test assumptions?
  • There's a difference between searching Slack (stuff I know exists) versus something like Stack Exchange (stuff I hope exists)
  • If Project MoodleNet is not itself going to be a repository of resources, how could the system direct people to upload content to the right place? (e.g. YouTube / OER Commons)
  • Twitter chats, etc. are useful because they create spaces without hierarchy which could be conceptualised as Temporary Autonomous Zones.
  • What kind of plugins are available for Mastodon? What additional functionality already exists?
  • With events, the system could scaffold the user if they're working towards a MoodleMoot versus, for example, a small meetup. Could feature crowdsourced guides
  • One way to connect a system with separate parts is to use hashtags.
  • Could experiment with Mastodon at the upcoming UK & Ireland MoodleMoot? What problems could it solve? Issue badges?
  • Worth talking to Mastodon developers r.e. colour-coding different types of interaction (e.g. public versus direct messaging)

Tim Klapdor (6th March 2018)

Tim is the Online Learning Technology Leader at Charles Sturt University, where he's also held a variety of roles relating to the implementation, development and production of eLearning resources and systems. His personal website can be found here.

  • JetPack metaphor works
  • Community around Federated wiki kind of died due to problems with Mozilla Person as authentication (single email address)
  • Shibboleth approach to authentication? Unity network based on Open Academic Environment looks useful (if a little research-focused)

Project MoodleNet team (6th March 2018)

Present: Martin Dougiamas, Marina Glancy, Tom Murdock, Ryan Wyllie,

  • Need to re-think the JetPack metaphor as it's potentially confusing for those who haven't self-hosted WordPress, and also is technically-focused, which might throw some people off what we're actually trying to achieve. What's NOT like JetPack? How is MoodleNet different? Can people still use it if they're not using Moodle?
  • MD mentioned that we're "first and foremost targeting existing Moodle users" as well as those who aren't currently using anything.
  • Connecting people between MoodleMoots will be an important part of MoodleNet. Need to make sure that's in the vision somewhere.
  • Potentially remove 'forum' as a distinct component. Shouldn't be distinct to 'social networking' component.
  • Could add 'dashboard' feature to MVP?
  • Needs to be a sense that new users are joining something. Mastodon perhaps a bit too disparate and Twitter-like for that?
  • Important to start prototyping (paper-based if necessary) before we start coding. Iterate on that, potentially on weekly basis. We can then break down into work packages.

Ian O'Byrne & Greg McVerry (1st Mar 18)

Greg is a digitally networked scholar. His personal website can be found here. Ian is an academic who thinks about literacy, technology, and education. His personal website can be found here.

  • Scratch has a great user community focused on making.
  • The JetPack metaphor also allows you to push things through the network (e.g. upcoming Gutenberg update)
  • LTI compliance is important.
  • Project MoodleNet will potentially allow the community to identify people by what they make.
  • Portability of data is important for those leaving insitutions, or who teach in more than one.
  • ORCID data is now being linked to reviews, as well as academic papers.
  • What's the current status of Mahara? Could it be used as a basis for the profiles part of Project MoodleNet?
  • Interoperability is key - could use something like Bridgy?
  • A Project MoodleNet account could work like IFTTT and connect to everyone's network (kind of a 'Bring Your Own Network' approach) - bring in your contacts, but also those suggested to you by the system based on what it knows about you.
  • Headless system definitely something to aim for (c.f. Siri / Alexa / Google Now)
  • Relationships are based on trust but also need to have an element of ambiguity. Privacy settings, on the other hand, should be unambiguous and granular. Think of Google Docs model or local (=faculty), federated (=field), and public.
  • There are lots of open source dashboard solutions we could build from.
  • Mozilla had an Open Scholarship team that had something almost ready at the end of last year?
  • Interesting Hybrid Pedagogy article on Education in the (Dis)Information Age
  • Onboarding is important - the user shouldn't already have to know anything (particularly about Moodle) to get started - could use pop-up modals to pre-populate profiles with templates to ensure there aren't irrelevant fields for user to populate.
  • What's the best way for the community to give feedback on the project?

Nitin Parmar (1st March 2018)

Nitin is a digital education consultant. His Twitter account can be found here.

  • Working with Premier League, has been Moodle user for ~15 years.
  • Challenge will be for Project MoodleNet to get buy-in from established user community.
  • How do users give back? (money / development time)
  • Should talk to Lewis Carr (consultant, previously at Leeds City College)

Grainne Hamilton (1st March 2018)

Grainne is an education consultant. Her Twitter account can be found here

  • User profiles - would want to share:
    • Badges
    • Social media (LinkedIn, Twitter, Instagram)
    • Blogs (Medium, WordPress)
  • Social cards which 'unfurl' after sharing from Project MoodleNet could allow for 'story bits' to be added, much like the Mozilla Discover project was aiming to do.
  • How do we go about trusting people within social networks? How does that trust emerge? How can we verify people and information?
  • Users should have choice, particularly around restricting who they share with and tagging / adding metadata to their resources.
  • A missing part of resource sharing is the path which the individual took to create it. Example of the visual thesaurus to find things in the same ballpark as what you're searching for. Could be generated by relationship between popular folksonomic tagging.
  • Potential to create a badge pathway of resources?
  • Focus should be on Project MoodleNet as an educator curator network.
  • Create affinity groups with Project MoodleNet (which could be renamed 'M-Net' or similar to widen its appeal?)

Clint Lalonde (28th February 2018)

Clint is the Manager of Education Technology and Development at BCcampus and Associate Faculty in the School of Education and Technology at Royal Roads University. His website can be found here.

  • How users will need a compelling reason to take part in Project MoodleNet, and that will differ depending on sector, location, etc.
  • As much as we talk about inter-disciplinarity, faculty hold onto their disciplines, and that's their community.
  • Most of the success Clint's team has had with OER and open textbooks is by having advocates within disciplines ('faculty fellows' programme).
  • Clint liked the following in the white paper in particular: Communities of Practice, focus on professional reputation, and sharded identity.
  • Profiles on Project MoodleNet should include badging, and could also make use of microformats (XFN).
  • The best application of crowdfunding would probably be around recognising the curation of resources.
  • Clint's team curates their collection of open textbooks, and culls stuff that's not used. "Curation is under-rated", he commented.
  • The OER Commons team was approached by Clint about custom integration a couple of years ago. One of the biggest challenges, however, was getting content in and out of Moodle (common cartridge format?)
  • Open Courseware has a huge challenge because of the difference in courses even with the same name at different institutions. Might be easier in Europe with Bologna process?
  • Name 'MoodleNet' is positive in that Moodle has a good brand and it increases legitimacy. However, it also reduces the scope for those not using Moodle to think that it's for them.
  • The JetPack metaphor is a nice model but the key will be to convince admins - how can you make their life easier? (conversation led to potential use of IPFS to reduce burden of institutional hosting)
  • Clint is going to introduce Doug to a colleague who has been using badges extensively with Moodle.
  • What about formalised professional development with Project MoodleNet? How would it link to events (online/offline)?

Tom Salmon (28th February 2018)

Tom is working on cross-disciplinary research in technology, education, and development. His areas of expertise are the governance of education systems, quality and language of instruction in education, technology and ICT integration and teacher development. Tom's profile can be found here.

  • Tom's work in South Africa around teacher CPD and open textbooks.
  • Disparities of access depending on the different scenarios outlined in the Project MoodleNet whitepaper.
  • ORCID, a system that "provides a persistent digital identifier that distinguishes you from every other researcher and, through integration in key research workflows such as manuscript and grant submission, supports automated linkages between you and your professional activities ensuring that your work is recognized." Something similar could work well for Project MoodleNet. It provides a 16-digit DOI for you as a researcher and a directory where you can be found. Allows researchers to tie their identity to published papers, etc.
  • People pushing back against 'proprietary stuff' not only in west but in developing contexts, too. Timing is important - example of open textbooks being accepted in South Africa at a crisis point where a budget shortfall would have otherwise meant no textbooks. Tom made the point that people need things that are more federated and open than proprietary products.
  • Example of an academic who had developed 30 flashcards on how to supervise a PhD. Everyone thought they were great, but when he approached publishers, they wanted to turn the flashcards into a book. Could potentially work better as crowdfunded resource?
  • Key thing in social networks is recognition. Not necessarily focused on a particular object or resource that an individual has created, like when people subscribe to YouTube or Twitch channels. Also discussed supporting teachers staying in the classroom.
  • Project MoodleNet could allow users to filter based on their interests - e.g. resources / professional development / social updates?
  • South Africa has seen a massive growth in Google Educator Groups, which are supported by SchoolNet. They call them 'communities of practice' but unsure as to how much sharing goes on in them.
  • Tom knows the person who set up SchoolNet, so could connect Doug to her, along with perhaps Mark Pegrum (mobile learning / digital literacies) who is at UWA, Perth, Australia.
  • These days, need to talk open and walk open - emphasis on diversity.
  • Project MoodleNet will probably need national champions who can contextualise for local relevance.
  • Could also talk to Tim Seal at the Open University (UK)

Jim Groom (28th February 2018)

Jim is co-founder of Reclaim Hosting, an independent web hosting company focused on the higher education community. Previously I was the director of the Division of Teaching and Learning Technologies and adjunct professor at the University of Mary Washington in Fredericksburg, Virginia. His blog can be found here.

  • How much of Project MoodleNet will be about users being able to control/share their own content and push it to different spaces? (e.g. Known)
  • Instead of being an OER repository, how about connecting to repositories, and allowing users to track the number of uses/remixes, etc.? (c.f. DS106 assignment bank)
  • Ravelry is as social network for people who knit and crochet - focused on shared objects (patterns) which in turn give particular users status
  • Should talk to Alan Levine, Martha Bertis, and Tom Woodward
  • If social media influencers (famous for being famous) are on one end of the spectrum, then social networks based on useful, remixable social objects are at the other end of it.
  • Should stay away from term 'OER' as it comes with a lot of baggage ("the idea of open has become a bit pedantic")
  • What's the angle when sharing on social networks? Need to start with a couple of excellent examples.
  • Dashboards could link with the idea of a personal API
  • Professional development through social networks could be enabled through dashboards by browsing places through a reader which is enabled through RSS and APIs.
  • WordPress metaphor (JetPack connecting .org/.com) is a good one - need to avoid cognitive confusion through problematic terminology. Might need to rethink calling it MoodleNet?
  • Good idea to talk with Kin Lane about dashboards. Two things: how do users customise? how do they pull things into MoodleNet (c.f. IFTTT - taking stuff you and your network already have and bringing it into one place)

Project MoodleNet resourcing (11th January 2018)

Present: Sander Bangma, Martin Dougiamas, Rohan Hardie

  • Current status of Project MoodleNet, MD mentioned that we need a map of the system (architecture)
  • Resource implications of GDPR (+Damyon)
  • External options
  • Timezone issues
  • Backfilling of roles

Gavin Henrick (8th January 2018)

Gavin is Business Development Manager at Moodle. His personal website can be found here.

  • (useful feedback r.e. slides upcoming presentation being given at All-Hands meeting on 17th Jan)

Sander Bangma (8th January 2018)

Sander is Open Source Development Coordinator / Program Manager at Moodle. His LinkedIn profile can be found here.

  • Development team uses JIRA for issue tracking. All code on GitHub in public repos.
  • Need to discuss resourcing in light of GDPR time pressures.

Tom Murdock (4th January 2018)

Tom is Education Advisor and leads the Education team at Moodle. His personal website can be found here.

  • Thinks the white paper is on the right lines about multi-faceted identities.
  • Great interest in approaches such as Blackboard's xpLor, designed by Moodlerooms, to share content.
  • An OER repository should:
  • include content that is licensed from the start and tools we offer should work within the constraints of the license
  • be trustworthy (c.f. GitHub) and it's better to have a versioned home rather than copies
  • be interoperable (not just Moodle quizzes, but wider resources, via LTI)
  • There's opportunities to put Project MoodleNet in other places that already have great reach (e.g. Google Apps marketplace)
  • How are we going to prevent bullying and misconduct within the social network offered by Project MoodleNet?
  • Should the social network be focused on people or on content? Something like the latter (e.g. Pinterest) creates conversations around OER as social objects. Allows for multiple, sharded identities.

David Mudrák (4th January 2018)

David is developer at Moodle HQ. His Moodle profile can be found here.

  • Suggests talking to Mark about Moodle authentication systems
  • Suggests using an external system as the primary source for user data which then exposes itself via various standard protocols (LDAP, SAML, OpenID, OAuth2 or whatever exists these days) and provides authentication services to Project MoodleNet.
  • Thinks that it's generally easier to integrate systems X, Y and Z with a standard auth protocol (as a solution will probably exist already) than integrate each of them with Moodle auth.
  • Thinks that developers need the list of software candidates Project MoodleNet will use and integrate with ASAP. Thinks it's correct to have started with expected features, but unwise to go that deep and design the details of auth mechanism without knowing all or most of the software stack that will be involved. For this round, suggests specifying that there will some kind of SSO functionality.
  • Thinks Google Apps might be a useful analogy when explaining Project MoodleNet to people.

Martin Dougiamas (13th December 2017)

Martin is CEO and Founder of Moodle. His Wikipedia page can be founder here.

  • Wants Project MoodleNet to be 'extremely open'
  • Focus on functionality for educators, need to reflect this in 'Scenarios' section of white paper
  • Be more specific in Scenarios - what are they trying to achieve?
  • Reputation: Project MoodleNet should encourage positive behaviours such as making/editing resources, and encouraging educators to help their community/world (badging?)
  • We should be enabling educators to get paid for the work they do, as currently it's only partners who are paid. There's a whole 'gig economy' layer below that (e.g. quick consulting where Project MoodleNet provides the reputation system, method of payment, and calendar)
  • JetPack analogy needs unpacking / explaining / diagramming
  • Focus on the user experience - what does it allow educators to get done?
  • Don't include 'approach to design and build' and 'enabling technologies' in white paper
  • Project MoodleNet could (eventually) interface with Learn Moodle and MoodleMoots (suggested / advertised). Local meetups, too?
  • Pseudo-anonymity is a good option, should people require it.

Project MoodleNet team (21st November 2017)


Helen Foster (15th November 2017)

Helen is Moodle Community Manager. Her Moodle profile can be found here.

  • Moving MoodleNet project documentation from user docs area to dev docs.
  • People like group icons and badges, which are awarded by helping (some automated, some manual)

Helen kindly put together a list of things that work well:

  • Educators and developers mixing in the forums
  • Groups with group icons in the forums - Particularly helpful Moodlers, Core developers, Plugin developers, Testers etc.
  • Useful ratings - used to calculate PHM group membership
  • Badges
  • Moodle jobs database
  • Answering questions by posting links to the documentation
  • Moodle translation site
  • A clear policy on what is and isn’t allowed with regard to advertising in the forums plus making the effort to enforce the policy
  • Giving trusted people moderator and editing rights
  • Communities for different languages
  • Forums for different areas of Moodle e.g. themes - though not all areas of Moodle need a separate forum

...and things that are sources of frustration / not working:

  • People don’t have any idea about the tracker and find it bewildering to use
  • Having a different Moodle Docs wiki for each version of Moodle - causes problems when people do a Google search
  • Moodle Buzz - used to work a long time ago but nowadays nobody adds anything
  • sharing courses and content
  • Less active communities on attract spammers
  • Newcomers confuse with their own Moodle site
  • People think they need to register their site for others to access
  • Crowdfunding database - set up but never really used
  • Front page of!
  • site policy - nobody notices the link in the footer of each page and you’re not required to agree to it when you sign up for an account
  • Tags on
  • Confusion with and

Juan Leyva (15th November 2017)

Juan is Mobile Lead at Moodle. His Twitter account can be found here.

  • Whether we need a new app specifically for MoodleNet (undecided)
  • UX around searching for Moodle installation - vs. typing in URL of organisation's instance
  • Blocks within the mobile app and example of course completion being available via 'course option' in app
  • Ways authentication / login currently works (OAuth / tokens) and whether it would be possible to login to a self-hosted Moodle site with a MoodleNet account
  • The appear of Stack Overflow and the importance of building / maintaining reputation within a community
  • Well-formed questions, moderation, and serving up answers to common questions within Moodle installations
  • Improving search (which is poor at
  • Creating progressive web apps vs native apps for MoodleNet functionality - e.g. integration with Siri / Google Now / Alexa
  • Pursuing small experiments (e.g. 'MoodleNet Help' app)