I recently came across the RabbitHole DAO, which is exploring the approach of Proof-of-Learn.
I’ve barely gotten my feet wet and still have a ton to absorb, but the basis of the approach to create an ‘on-chain resume’ is this:
- ⭕ Earn XP / tokens for completing tasks
- ✔️All current tasks are on-chain activities and are therefore consensus verified somehow (depending on the chain protocol)
- 🎓 If tasks are properly structured, then by completing the right amount (in the right order, perhaps) you can be said to have “completed a course”
Here on Hive, the good ol’ @hivebuzz system demonstrates this. As you do more Hive things, you earn badges that are automatically verified by the blockchain. So if you wanted to prove to someone that you’re a veteran Hivean, you could point them to your badges.
My goal in this post is to expand this model beyond just web3 and try to imagine how this could be applied in a broader, interactive, decentralized system of educators and learners.
Below is a loose set of ideas and components that could contribute, and I’d welcome any and all thoughts to help crystallize the concepts. 😃
| Note |
|---|
| Most of this thinking applies to adult learning, and not kids. There’s no reason to think that the concepts here couldn’t be applied in public or private schools for children, but I suspect that most early adoption will be in higher education. |
Proposed overall approach to decentralized education
- We want people to obtain, retain, and then use knowledge
- In the web3 broad vision, institutions would have less power (not no power) in dictating what knowledge people obtain and how they do it
- We need a way for people to own their knowledge and experience credentials
- We need a job market that accepts Web3 validated credentials
| Key assumption: Rabbithole model is valid |
|---|
| For the thinking below, I’ll assume that the idea of on-chain task-based activity as the valid, bedrock system that underpins everything. |
(A) 🔝 The role of curation: surfacing the best tasks
Who decides what tasks should constitute a “course?" How do you structure a credential that a third party will take seriously and that holds real value?
A DAO could come together and build out courses. I imagine that’s how Rabbithole does it today. (I believe they call task-sets "quests.")
But I think a curation model could work as well, maybe in tandem with DAO governance. Imagine a Hive-like system that elevated quality task-sets to the top of a list.
(B) ✅ Ensuring quality with subject matter experts
I think this would be a big area of pushback from traditional education proponents:
“Sure, some random people on the Internet voted this “course” to the top, but how do you really know that the tasks inside constitute real value, real learning? You need an expert to verify!”
I think this is where Creator economies (such as Rally.io) could intersect. Some people will build up a reputation for having course building expertise entirely online and anonymously. Others can use their existing non-web3 reputations to get a jumpstart (e.g. professors, tutors).
You could use the value of the creator’s personal economy or some sort of community rating (ala Hive reputation) to give them more weight in evaluating certain courses.
(C) 🖇️ Tokenizing off-chain tasks
This needs to be answered in order to expand the educational system past web3 topics and at least into digital.
If the topic of learning is “public speaking” how do you translate associated tasks on-chain? Can you? Should you? There are definitely ways (upload a video, somehow verify that that is you), but there will be limits on just how well certain tasks translate into the web3 space.
VR and AR tech may be an eventual bridge to this, if the tech becomes accessible / affordable to people.
(D) ✒️ Space for manual evaluation and verification
The blockchain can only give consensus that something happened, but not necessarily on if it was done “well.”
Lots of task completion has a subjective, highly human component to it. You need experts or intended recipients to judge your skill and say whether or not you completed a task in a satisfactory way (and who decides what is satisfactory?)
This will be another key hurdle, and it is related to point (B). There should be some thinking around how we incentivize verified experts to manually and fairly evaluate certain tasks that can’t be easily ranked.
This could become a key job in the future. Here’s one way I can imagine it happening:
- Learners earn tokens for interacting with the system, like we do here on Hive
- If they need a completed task/course manually evaluated, they can bid or pay those tokens to evaluators (because the to the learner, the credential has more potential value than the token)
- Alternatively, the learner’s company or DAO might sponsor them instead, paying the token cost
(E) 🏛️ Space for traditional education to interface
There should be a lot of thought on how to gently and meaningfully onboard both traditional education and web2 edTech into this world. It is likely that a hybrid web2-web3 approach would exist as the world adjusts to the new technology and possibilities in this space. It may even end up that the existing educational structures are better positioned for some kinds of learning (CPR training comes to mind).
(F) 🔐 Security for the masses
For wider adoption of this, we will probably need a better system for securing on-chain resources. A wallet is a single point of failure in many cases… lose your key, and now you not only lose your $$ but your entire educational resume and credentials? Right now it’s not really possible for someone to “steal” my university degree as it's guarded and guaranteed by a big fat centralized institution.
My general hope is that this is a topic being tackled from many directions in the web3 world. So that by the time decentralized education is more mainstream, so will better security options for wallets or digital identities.
💡 Closing thoughts
I’m excited to get to know the RabbitHole DAO better, and see if any of the above aligns with what they have in mind. I think that curation based systems like the social side of Hive could play an important role in decentralized education. The sidechain capacity of Hive could also potentially support some or all of these ideas… food for thought!
📰 Other resources
- Web3 will change the way we learn • Medium article by Akshay Shinde
- Breaking Down Education in a web3 World. An interview with Brian Flynn, Co-Founder and CEO of RabbitHole.gg • Substack article by David Zelniker
- Kernel • "A custom web3 educational community"
- Introducing the EthernautDAO • Medium article re: training the next gen of Eth devs.
- RabbitHole.gg • "Learn how to use and contribute to decentralized apps, earning tokens based on your on-chain activity."
- Proof of Learn platform raises $15M in round of funding led by New Enterprise Associates • Mysterious new DAO/startup? Also see Crunchbase.
Would be nice to have something like this on hive ^^
I think that the second layer on Hive could definitely support some of these ideas. I'd be great to have a new kind of educational dApp to help drive more value to Hive.
Yes it could be awesome! Or if someone could post jobs there ^^
So I have a lot to say about this and I will try to contain my enthusiasm and word count to a manageable number.
This does excite me and sparks off with concepts I have been mulling over for a long time. Because you have opened the door, it instantly snapped together with a whole series of new concepts that I have just learned over the past 4 months he on Hive.
In my viewpoint, to build this you would not need to re-do a lot of knowledge that has been tried and tested and as such I believe that bridging the gap between traditional education and this web3 tokenized education to be the most crucial.
Our main focus, when considering this project should be nothing other than competence. That is what drives the world forward in a sustainable way. Without people who truly understand what they are doing and who can complete the real life tasks successfully, there really is no value in having a system that teaches anything.
I would design a system that would attribute subject authorities with various soul-bound NFTs ro represent their credentials as well as their successful projects. Each NFT needs be assigned a value to carry weight within appropriate fields of study.
I would need to get more granular to explain it properly but the idea would be for a professional with a Doctorate in electrical engineering and 15 years experience designing and building successful systems to have a vote weight when validating the course information of a piece of education dealing with electrical engineering.
An electrician and a mechanical engineer could also vote on this subject, but you should not be able to collect tokens infinitely and all of a sudden someone with minimal knowledge and proven competence can outvote the Doctor with years of success in his resume.
Similarly, when this Doctor votes on a course regarding architecture then his vote weight must be modified based on the relevance of his knowledge. Hid Doctorate only barely touches the subject matter, while his practical experience counts very well into the subject matter due to the cross-disciplinary skills gained by installing electrical systems into buildings.
When the same Doctor has to assist in verifying a course regarding cardiology, his credentials nor experience count for very much and his influence within that subject is counted as a mere educated opinion. Someone with no proven education nor practical validation in any field would count even less in this regard.
There is a lot more nuance to this but this would be just a starting point to give a value of professionals in the web3 education world. It would get easier the more a system is built to include and test the various factors that would make this system work.
I could go on, but let me give you time to read and give me your opinion on this.
Love the article!
Hiya! Appreciate the enthusiasm, it clearly shines through :)
You've done a solid dive on some of the nuances surrounding point B—subject matter experts. As I understand, you'd advocate chunking up everyone's "expertise profile" into as many subcategories as reasonably possible, so that their impact on the system is proportional to their actual knowledge in any given range of topics.
If the way forward does cross this bridge, I can see a lot of tension forming between keeping things manageable/simple (e.g. fewer categories) and striving for precise representation (e.g. as many categories as are needed).
Knowledge in general is a fairly subjective topic, so I imagine a compromise would emerge 😊
Indeed. But in that case we can simply follow the nature of physics and nature.
It is just as ambiguous in many situations as to what constitutes the correct conditions for something to happen. But in every natural system there is a Range within which it happens and there are ranges on the extremes that it does not.
Whatever number we place on a person's skill, there is a safe range in which we can allocate to them.
I happen to work with a number of technicians. I can probably rate each one with an ambiguous set of numbers based on what I think his competency is and that competency would be derived from what I think of him and what I know of him. If I don't really spend time with a technician or don't like him, then I would have curated and ordered the technician's competency incorrectly.
But if we replaced my opinion with the statistics of his success, then that value would be far less subjective.
Sure, we dont use that system. We just know if we send this guy for that job that the chances are high or low of success and move on.
But if it were MY business I would totally have these statistics generated and use it as a way to manage my business, train my staff and avoid mistakes.
This is a fun philosophical discussion too, because we're straddling some questions of humanity—for millennia, we've relied on our gut feelings and intuition, and much less on precise measurement.
Along comes computation, and in a few short decades many people now have access to a gargantuan amount of computing power (few of these people maximize that potential). Now here you and I are, discussing hanging up most of that gutfeel and intuition and replacing it with a deep dataset of measurement. And we're not just riffing, we're having the conversation on a platform that is made of the tech that would theoretically support our ideas.
I'm mostly curious (versus for or against) about how we achieve harmony with a deeply measured existence. Once we get into it, can we get out? Will those who are uncomfortable with it be able to thrive without it?
Intense stuff. :)
Indeed. So there will always be those people who will choose the "gut feel" over data and analytics, especially from a computer.
Those algorithms are generally based on someone's expertise and this has directly led to other humans making remarkable advancements in many fields.
It goes along with the premise of not reinventing the wheel. The microbiologist does not create and invent his own microscopes, nor does he test and verify biochemistry experiments when conducting his breakthrough in study. He must work off the shoulders of others.
Thus we already do live in that kind of society, its just not digital yet and there are no quantifiable links to one part of the experiment or task and the other, yet.
Failure or perception of success in laboratories often are influenced by a multitude of factors no matter how hard we try eliminate them.
It takes time and repetition to prove or disprove something.
Finally some people still need to use their intuition to decide on whose shoulders to stand and what factors to remove, add or alter when an experiment fails.
In a different context we could align it to a baker or chef that follows a receipe. He may decide, like I do, to make an alteration. But he hardly begins my grinding his own flour. Or growing his own wheat.
So coming back to that question: Once we delve into this, can we come out?
If my above statement is true then we never really come out of it unless we are self sufficient gurus that do everything ourselves and become perfect hermits... then we do not have a society that needs other people at all.
You are right. This is fun. 😁😄
On the path to the singularity, it would seem. 😉
I also wonder if this digitization-of-all-the-things will have a short term impact on the general population's mental health (for good or for ill). We are now seeing some of the repercussions of web2 social media starting to air out on downside of that idea.
I can envision a future where more granular, quantitative insight helps, and also one where it hurts. As long as our brains have not yet evolved to rely on machine input, we will be old-school pattern seeking creatures that oversimplify and see stories everywhere. I feel like a data-driven reality will need to work hard to put up guardrails.
Reigning this discussion back on topic though... perhaps a big role of web3 education can start with a more pointed global curriculum in understanding the machines that underpin our digital existence.
There should be a global indication to help understand the context of the internet and group psychology as well.
I find that individuals are often both clever and kind and that masses of people are often neither.
We all want to be treated as humans but often lack the insight to treat people as human once we have no clue who they are, which happens when you move and react as a group.
Thank you for mentioning our work @jfuji.
BTW, it would be much appreciated if you support our proposal for 2022 so our team can continue its work!
You can do it on Peakd, ecency,
https://peakd.com/me/proposals/199
Thank you!
Proposal signed! Keep up the great work :)
Thank you for your support @jfuji, much appreciated!
🍾