Feeling out the boundaries - a few scenarios
For each scenario, please ask yourself:
- Does it feel right?
- Does it help us succeed at the big picture level?
Scenario: Google MOOC
There is a lot of interest in massive open online courses at the moment. Yet early experiments are not taking advantage of good pedagogical practices, or sophisticated technology to support community building and peer-learning. Google Research contracts P2PU to create a prototype MOOC using existing Google applications (combining Google+, Google Hangouts, Google Docs, and Google Groups). The project team includes researchers from Georgia Tech who are experts in large-scale computing education, and a dedicated engineer who manages the integration. The course focuses on teaching strategic game playing skills and is launched with a lot of attention / co-branded P2PU. The technology used is available for use by anyone without charge, but it is not licensed under an open source license. User activity data is shared with the project team for analysis, but not made available to the general public. As a result of the project, P2PU is engaged by Google Research to convene and support a groupd of university partners that are interested in launching their own MOOCs.
Scenario: Platform or Partner (was "Schools") [updated the scenario to use hypothetical school]
P2PU partners with Open Journalism Foundation to create School of Journalism, a peer learning community focused on journalism. OJF has many enthusiastic software developers who volunteer to support the project and set up a separate instance of the Lernanta software. They make modifications to the code and ask P2PU's Tech Lead for help with problems they run into. Some of the changes they make flow back into the Lernanta code base. After the first year of the partnership OJN decide to switch to a different platform - a micro-blogging system - because it offers the core functionality they found to work best with their users.
Comments:
- The end of the scenarios is imaginary. right? :) seems so
- The whole scenario is fictional (just replace OKFN and School of Data with something else)
- Bekka -> I guess this scenario hinges on how much of our focus goes into this School of Data, and whether OKFN (or any other partner) pays for the time our staff put in to the projects. Bearing those variables in mind, I think it's a (potentially) good option for us, but it need not be 100% of what we do. Other configurations for P2PU could/should still support these kinds of partnerships
- Vanessa -> If the organization goes another way (platform that isn't P2PU) is that considered a failure? How does a lab facilitate "productive failure"? Will we have a lessons learned phase of each experiement? Is the collaboration itself what we're looking to achieve, or the outcomes from it?
- Nadeem: Those are the questions that pre-occupied me for a while. Personally dont believe it would be a failure if the aim of the Lab is to experiment and collaborate with others (or indeed help them get off the ground). My difficulty is with the notion of a single platform into which we co-opt everyone, the reality is that other projects will evolve along lines we cant control. But that doesnt mean the collaboration wasn't worthwhile.
- (Hope nobody minds if I insert some notes after the call, couldn't make it -- Stian. I think we need to think about what success looks like. Think about something like YCombinator, whose job it is to spin off as many successful products as possible. It's a company incubator. Would it make sense to think of ourselves as a learning project incubator? Maybe having SOSI or School of Open Data "graduate" and have their own platform (could be a forked Lernanta, could be anything else), their own admin, funding totally separate from P2PU would be a huge success? (I'm not saying I think so, but it's an interesting model to consider). After all - the goal of P2PU is to enable peer-learning and experimentation... (Of course, there is another aspect - YCombinator has a financial stake in companies being successful, this is something we'd have to think more about in terms of funding and sustainability of P2PU etc).
- Agree Nadeem. Tho "Scalability" and "Productive Failure"--are these at cross purposes? Supporting both is a tough needle to thread
- +1 agree this is a really interesting tension.
- +1 great question.
- I'd love to hear from more people if it feels like failure if OKFN uses a different platform even if we still collaborate on many aspects of the project. And why it would feel like failure.
- Dirk -> It doesn't feel wrong from their side, they should use what suits them best. But we should aim at providing them with enough freedom and enough support. An interesting question to ask in this scenario would be "What did they loose when they switched to a micro blogging system?" If the answer is nothing, then maybe Lernanta was never a fit for them. If the answer is something like the connection to peers we need to figure out how to offer them that without them needing to use the rest (API)
- Zuzel -> I will consider it a failure if the developement of the customizations is done completedly separated from the development of lernanta:
- We should think about how to transform lernanta to allow the kind of customizations they need, so it is easier the next time we have to partner with other pleople that also need to customize the software
- Making quick modifications that can not be extended and are too specific will not help much +1
- If the time between doing customizations and merging back into lernanta is too long it is likelly that it will not be possible to accomplish
- Piet -> To me, this is exactly what an open, learning organization does. They create, they share, they explore with their own peers (viewing OKFN as a peer org of P2PU) and learn together what works and what doesn't. Some knowledge transfers between orgs, and in the end, everyone gains. If a particular P2PU individual doesn't feel he/she is getting a lot of value out of the interaction, then that individual will limit his/her effort there. To me, this scenario has less to do with 'Schools' and more to do with organizational-level interactions.
- I am not attached to the P2PU platform. I'm attached to the P2PU community. +1 - agreed, i think this is really important to remember.
- jessy -> wouldn;t want to see all our dev/cultural efforts go into a single project. it doens't feel WRONG but it also doesn't feel scalable. agree with dirk that it is good if others want to use the software.
- Philipp -> to clarify this would just be one of our many projects
- yes, but IMHO this wouldn;t scale to "many" projects (as presented :)) What are the reasons that this would not scale? What if many projects like this came with funding to support (or there was a big pot of funding to support many projects like this) <-- then i guess many of our resources would go towards it, and correspondingly not other things? i suppose i'm assuming the community stays roughly the same size but if we had tons of funding we could hire bazillions of people -- then it becomes a question of how fast you can scale things culturally (google hiring 100s of engineers each week etc.)
- John -> Why does OKFN want a separate platform? +1 A possible better solution could be to make P2PU a central place for profiles (ie p2pu.org/username) and allow other projects to hook in to those profiles via an API. I'm really not keen on the rest of the specifics above.
- I don't think the platform should be the focus of this scenario though - it's about the evolving partnership, isn't it? Philipp -> For this scenario, the platform question is core - let's drill into that.
- If that's the case, the specific details in the example are extraneous...
- I like the idea of the partnership, but the details of this example is quite specific. "Install their own" etc (Yes, it's intentionally specific so that we really engage with the possibility of a partner using a different platform)
- Stian: I think we can also think about what value P2PU can offer to other experiments that don't feel like they fit on the "full" P2PU platform (which I think will be many)... Let's say I wanted to do a research project about new ways for groups to collaborate and I wanted to iterate very quickly with a bunch of technology that would neither be ready nor necessarily appropriate for "all of P2PU" ... are there technical advantages for me to work with P2PU - could P2PU offer common signon, help with badge issuing, storing portfolios, linking learning on multiple platforms to a common learner profile / portfolio, promotion of courses, a peer community that is interested in reviewing the pedagogy and provide support etc.. Or is our message to others: unless you can convince us that we should modify our core platform to fit your ideosyncratic needs - which will happen on our schedule, because we have thousands of users to think of and can't just push your changes without thorough review etc - then we really have nothing to talk about...
- QUESTION -- Do we think our platform should and can (realistically) serve the needs of all our partners. Or does success feel like people using many different platforms? Does success depend on what our partners use?
- Many Platforms or Our Platform
- John -> I don't think that our sucess hinges on the use of our platform, but it is tied connecting all of the different parts in an understandable and meaningful way +1 +1+1
- What does that mean concretely? It sounds right, but the devil is in the details.
- can we feel like we are having success if we are collaborating with others who are using many platforms
- Alison -> at the moment it feels like, in schools, some types of activity is taking place on other platforms, so in answer to the (original) question (can our platform support everything?) I would say no, +1 +1-- this phone call is platform for collaboration and learning, P2PU needs to think of itself as platform/tool agnostic+1
- @alison i assume you mean "no" that it shouldn't try to be everything
- alison: correct.
- note that not being able to do everything does not mean we can't do many things
- nadeem: id go further slightly and say that it isnt practical to try to be everything ... what would our focus actually be?
- Is this like Philipp's earlier question--does it "feel" right--if the features are in line with our mission and values, could we consider them? And also, is this a definition question? Each time a school makes a feature requirement, should we compare them against P2PU core features set?
- This scenario goes more into lernanta as an opensource project. Why will people want to contribute code other that to learn about django? One main reason for wanting to contribute will be that they can customize the software to fullfil their short term needs (even if after a pilot they switch to another platform). If the process has improvements to the software as a result then it is worth (if those improvements are in the same scale as the efforts we put into the collaboration)
- Stian: It's also a question of how tightly bound the platform (Lernanta) is. I have no idea about the underlying technical complexity, but I would be very interested in enabling something similar to Facebook apps - where someone can experiment with new functionality, new ways of providing information, interaction, integrating outside services etc -- without having to get it into the main code branch (affecting the stability of the whole platform) etc... I think an API might be a first start - I was very impressed by John's automated TED talk stuff (did he ever build it?), and would love to see more of those kinds of things!!
- Jane:
- what is the separate instance for? To have the school running on entirely different software so they can customize look and feel? Then it seems like they are kind of using P2PU's resources to develop something that very tailored to a specific product, and many not reflect P2PU brand. Is that ok?
- As for the larger point around finding the P2PU platform no longer fits their needs and going off to do their own thing or using something else (I'll ignore the nonsensical "micro-blogging system" bit --r eally?), I don't see a problem with that. Using P2PU for a year is a long time. Lots of companies start and die within that time. I agree with Nadeem there is still value from partners using P2PU for a time and then evolving on to other things. As long as there is a mutually beneficial relationship going on during the period they are with us.
Scenario: Thought Leadership
- P2PU is interested in the question of giving certification for skills that lack good certification (e.g. web developers). We organize a workshop and write a small white paper with Mozilla. Based on early ideas we build a first prototype using OSQA and pilot in School of Webcraft. The work takes off and gets spun into a much larger independent badges infrastructure project hosted at Mozilla with support from MacArthur Foundation. P2PU steps back from the core project. Through the process we developed a reputation as thought leaders for assessment and alternative certification, but the badges infrastructure project grows beyond P2PU and is mostly associated with Mozilla and many other organizations that get involved.
Comments:
- Nadeem: Scenario sounds to me like an "experiment" that worked? transition into something bigger, and that P2PU got credit / recognition for. To me that feels both "right" and speaks to our big picture: Makes us relevant?
- an experiment that transitioned.
- Bekka -> But we need to learn how to grab some of the glory for ourselves too, because it can bring in money and recognition. +1+1+1+1 big time ("big time"? haha)
- Who is aware of our involvment outside of the people who were originally involved?
- In this "scenario" the core open education community, but not the broad public. The New York Times writes an article about badges and it mentions Mozilla but not P2PU (I think it actually did mention P2PU, but let's assume it didn't)
- funders like Hewlett and MacArthur.
- I read the scenario as we actually got the recognition ;-)
- At the same time, our flexibility allows us to pivot, try other things. Badges may be going in a way we might not necessarily agree with, and we're free to let it be and move on/try other stuff. Mozilla now has to answer a lot of the motivation questions associated with badges. We can take our lessons learned and do something else right away.
- I like the flexibility and also thats going to speak to individuals motivations. We are a "volunteer" community many people are interested in research or ideation but not necessarily in the productisation of whatever comes out of that. We could find that Lab volunteers/innovaters ( cant remember the phrase Philipp used earlier ) find this appealing and once the idea is proven or garners wider interest others can take it further.; or we focus on a specific part that we find more interesting.
- projects do end, and that is healthy, but i would want to see us writing up more formal lessons learned for ourselves/our community, possibly publishing "old fashioned" academic papers (or science superhero comics! just saying) and documenting our thought leadership better.
- Let's assume that this is not always possible. History gets written by the leaders. The early innovators often get overlooked. (Not saying that's 100% the case, but want to make the scenario hurt a little)
- But I like where Jessy is going with visually demonstrating our ideas. John's assignment that we create stories is an opportunity to do some visual documentation.
- Stian: I agree with this, for example the early example of publishing our reasons for choosing a specific CC license was very positively received. There have been some whitepapers etc around badges, but mostly from Mozilla (because they funded staff etc I guess) or joint... I think we could do a much better job at gathering experiences, best practices, insights etc, and making these easily accessible... Of course this is a lot of work :) But it's incredibly useful. (Maybe we can give out badges for it :)))
- Seems to me like none of us object to this scenario in theory, we just want more to come out of it. But we can see ourselves doing this kind of thing in the future, only better.
- How can we get more to come out of it (let's stick to this concrete case / be as concrete as possible)?
- publication
- more experiments at the fringes - what do you mean by this? in this particular case?
- good question... i guess i just meant at the edges of what is a badge, how they get assignment, automatic or manual, what they look like, etc. but also perhaps totally alternative approaches like reputation systems, payment (radical i know), er... physical tokens?? authentication chains?
- alternatives to badges, different types of recognition/accreditation, testing theories that mozilla and others come up with, etc.
- Has P2PU documented what it got out of the badges work? Perhaps more reflection internally would lead to a better feeling of "getting something out of it". Writing up another white paper could be a useful strategy to compile thoughts, lessons learned, future directions, etc. Sharing that paper more broadly can also have an impact on the badge community.
- Sounds a little like having to let your children leave the house? Yes
- The best part of this scenario from a tech perspective it that we are able to use what they are building (the obi) in this new more bigger phase of the project
- [This is a different scenario - really interesting scenario, but brings out separate issues!!!] This scenario doesn't end here. School of Webcraft still sits as a major section of the learning community. Now that Mozilla has moved on it has no leadership, and the P2PU leadership does not have the bandwidth to take it on in full. I would not want to see a scenario like this happen again. +1 - we need to think about the "legacy" stuff like this in future partnerships.
- Jane: I think it's fine, but we need to document everything better. Lab for research around accreditation and assessment - here's the portfolio of the work we've done and our findings with these various partner organizations/groups in the past. So that we'll be appealing to future partners who want to come in and build off this prior work for their own Schools or research or whatever. Example: School of Open will be doing certification around open skills. We'll be looking at the Mozilla/P2PU partnership as a starting point. In fact, Creative Commons is choosing P2PU to work with because of the School of Webcraft model that was kicked off /the whole badges discussion, and how successful we perceived it to be (regardless of whether it was or not).
Scenario: Consulting
- JP Morgan hires P2PU to redesign its professional development offerings. P2PU will work with JP Morgan staff to design a series of challenges and hosts a separate learning space for JP Morgan on our server. JP Morgan's global community participants in courses. The courses are not open. JP Morgan is interested in the possibility of using technology to scale PD offerings to their employees. For P2PU this project offers a source of consulting income but also an opportunity to test our social learning models in a professional development context.
Comments:
- Stian: I think if this was a good source of funding for other parts of the project, and we were able to get tech development/research/etc done that could benefit open parts of the project, there's nothing wrong with this model. We'd have to think about how to do it organizatorily of course.
- But even if our resources are paid for, we would be focusing attention on providing services to a commercial client. Does that feel right / does it really help us achieve our big picture goals? Also, what do you mean by resources are paid for? we can get all the funding we want, but if we only have two developers they can't do JP morgan deal + everything else P2PU is doing, no matter how much they are getting paid. Resources include time and humans. The reality is that a significant amount of human resources will be diverted from other, actually "open" projects on P2PU to help JP Morgan with their closed professional dev courses. Unless there are other real payoffs and benefits for P2PU and open community.. .just something to think about and weigh carefully. (Jane)
- I definitely think that P2PU can and should test out the consulting scenario, and I actually have some potential projects to propose, but I wonder if the "lab" is the vehicle for this. A month or so ago I started a discussion about this on this pad: http://pad.p2pu.org/p/consulting +1
- To clarify - the discussion about "the lab" is not for P2PU to set up a lab that is separate from the non-lab P2PU, but for P2PU to think of itself as a lab - where everyone experiments / prototypes / builds things.
Niels: Like this as well. instititutions and large companies benefit from knowledge transfer between employees. A private peer to peer uni could help out peer learning withing organisations.
Scenario: MOOC
- P2PU designs a concept for MOOCs. Stanford University already offers MOOCs and has developed a set of tools to support them. P2PU approaches the Stanford team to discuss collaboration. P2PU sets up the Stanford software on our server and designs a MOOC around Social Innovation materials. It interfaces with Stanford's automated assessment engine. The course is featured on the P2PU home page, we run a survey with participants afterwards, and publish a small report. Based on the experience we decide to encourage more P2PU community members to try similar projects.
- +1 (Alan) I like this scenario and can see it being a bridge to folks who develop other home-grown courses or tech that we'd love to use.
- Another scenario we've discussed is remixing MOOCs... we take the Stanford AI course and, with their cooperation, remix it to do A, B, and C testing of different strategies for pairing social interaction around it.
Comments:
- Stian: The Stanford MOOCs so far have been extremely content-heavy (ie. the main "value added" from Stanford/Coursera etc has been very high quality material produced by people with very high standing in the field)... all the social stuff has come at the fringes (I think they are incorporating a bit more of this in the new platforms, but still a lot of discussion happens on outside forums. I think it would be a better fit for how P2PU traditionally saw itself "as the social wrapper for resources" to think of the MOOC itself as a "resource" (albeit a synchronous one), and think about ways in which it could support learners, for example by letting groups of learners easily set up learning groups, find other learners at similar levels of understanding, do class projects together, etc... This is similar to how we imagined working with for example Open Learning Initiative in the past. +1
- Jane: Setting up Stanford software on our servers is no trivial matter. A lot of these scenarios actually... I see as being a significant drain on tech resources. Are we planning to hire ten more web engineers/developers if we do become a lab? Just asking. Which makes me think - a lab is open for experimentation! But a lot of these scenarios I find we are weighing pros and cons and benefits to P2PU. Maybe we should come up with a guideline of types of experiments that belong in our lab, in line with our core values? ie. You have to be open. JP Morgan wouldn't fit there, unless they somehow incorporated opennes, however small, into something they were doing with us.
Scenario: P2PU Sandbox
- P2PU hosts and supports two websites. One is the stable production site that is used by most courses. The other site is more of a sandbox - it allows contributors to develop new features, run live courses to test them, and gather evidence for what works and what doesn't. The P2PU development team moves tools from the Lab into the production site. The way Lernanta (our software) is structured makes it easy for contributors to plug new modules into it. Our development team spends a significant amount of their time helping partners build tools that plug into Lernanta. We develope few features ourselves.
Comments:
- Stian: +1 for making it easy to plug new modules in. Ideally this would be done in a way that there was only one website (but plugins wouldn't make the whole website unstable - you could easily filter experimental courses and more traditional courses etc)
- This is great. Don't we do this already?
Scenario: Better Online Courses
- Based on the first rounds of courses P2PU realizes that scaling traditional course model online is difficult and may not work. Want to make learning peer-based, engaging, hands-on. Developed a challenge model and platform to support it. Ran a first prototype with Webmaking 101 set of challenges and analyzed the results (80% of Webmaking 101 participants would recommend to friends). Learning challenges feature is turned into an embeddable widget. New challenges can easily be created on p2pu.org and then embedded in other sites. The learning challenges model scales to hundreds of thousands of users.
Comments:
- Yes! This is how we started. This is how we should keep going. We want to improve online, peer-to-peer education!
- Niels: I like this 'embed' function, as this would make it way easier for people to get p2pu in their workflow. p2pu keeps track of your performance in all challenges on all kinds of different online environments. Feels great.