Open Thinkering


Category: Open Badges

Open Badges is now on the plateau of productivity

Image aCC BY-SA Jeremy Kemp

Today I presented on a topic I’ve been presenting on for around 11 years now: Open Badges. I must have given 200 presentations on the subject, to audiences that number fewer than 10 to the several hundreds.

Over that time, the specification has changed, as have adoption rates, which have gone through the roof. Last year, I reflected in an article for the WAO blog that good things happen slowly, bad things happen fast. It’s taken a decade, as I predicted, for badges to be a no-brainer when it comes to recognising and credentialing knowledge, skills, and dispositions.

We’re no longer in the stage of “imagine a world…” but rather “here’s what’s happening, let’s talk about how this could be useful to you”. In other words, in the language of the Gartner hype cycle (to which I allude in the above post), we’re in the stage of ‘plateau of productivity’.

I recorded this Loom video to ensure I had the timing right for the 45 minute slot I’ve got. While I’m pretty good with timing, I use a lot of slides, and it’s been a couple of years since I presented in person!

I’ll also be throwing in a couple if interactive bits, so I need to ensure I don’t get stuck in the weeds with the inevitable questions about Blockchain / web3. My plan, as you can see in the recording below, is to point to v3.0 of the specification and talk about why decentralised identifiers (DiDs) are more exciting than blockchain, which I consider a back office technology.

So, without further ado, here’s a run-through of my presentation for FERS.

Next week I’ve been asked to speak at Het Nationale Bibliotheekcongres (the Dutch National Library conference) which is taking place in Assen, Eindhoven, and Amersfoort. I’ll be running a session fusing my badges work with digital literacies stuff in the service of discussing digital citizenship. Given than I’m only supposed to be talking for 15-20 mins before participants have 25-30 mins to do something, it’s going to be tight…

Decentralising the description of skills with OSMT

Skills don’t exist. Not really. They’re a shorthand way of describing human attributes and potentials which break down if you analyse them too closely. That’s not to say that defining skills isn’t “useful in the way of belief” as Pragmatist philosophers such as William James would put it ⁠— but rather that they only exist, or represent some “truth”, in as much as they have cash value.

There’s a couple of video games I’ve played for over 20 years, on and off. They’re both football (“soccer”) games, one being EA Sports FIFA and the other Football Manager (these days actually Football Manager Mobile). Both games attempt to quantify various skills and attributes important to the sport.

Screenshot of Football Manager Mobile, taken from

The above skills or attributes are out of a maximum of 20, and they’ve been handily colour-coded so that those playing the game can see at a glance how strong or weak a footballer is in a particular area.

The example I gave above of a video game neglects other things that real-life football teams can and do look for in players they wish to sign. For example, how adept at are they at gaining a social media following? Can they speak to the press confidently? What are they like in the dressing room? Are they volatile?

When we talk about skills in an education or learning and development context, we’re often implicitly talking about them for reasons of employment. Like the footballer represented by colour-coded numbers each out of a maximum of 20, it would make the life of employers a lot easier if they could view job applicants in this way. That’s not to say that they should, it just feels a lot like that’s where we’re heading.

Whoever decides on and controls the numbers is therefore in a very powerful position. They get to decide what is important, provide ways of quantifying those things, and report on them in ways that have real-world outcomes. In this way, it’s very similar to the work I’ve done about digital literacies: whoever gets to decide who or what counts as ‘literate’ holds the power.

LinkedIn is an example of a company (now owned by Microsoft) that would love to provide this level of quantification of human skills. The screenshot below is from a video embedded in an article from last year announcing their ‘Learning Hub’.

Screenshot from LinkedIn video announcing their Learning Hub

Who defines these skills? Well… LinkedIn do. They control the taxonomy. We should be wary of this.

One of the things that really attracted me to Open Badges more than a decade ago was that it democratises the “means of production” of credentialing and recognition. Although there are always attempts at re-centralisation (this week it was announced that Pearson have bought Credly, one of the major players in the landscape) the whole thing, thankfully, is built on an open standard.

If skills are to be a currency with a cash value, then we need to ensure that the skills represented by badges and credentials are also democratically defined. All of which takes us finally talking about the Open Skills Management Tool:

The Open Skills Management Tool (OSMT, pronounced “oz-mit”) is a free, open-source instrument to facilitate the production of Rich Skill Descriptor (RSD) based open skills libraries. In short, it helps to create a common skills language by creating, managing, and organizing skills-related data. An open-source framework allows everyone to use the tool collaboratively to define the RSD, so that those skills are translatable and transferable across educational institutions and hiring organizations within programs, curricula, and job descriptions.

At the heart of OSMT’s functionality are Rich Skills Descriptors (RSDs): machine-readable, searchable data that include the context behind a skill, giving users a common definition for a particular skill. With the open source release of the OSMT, other organizations can now develop and collaborate on individual RSDs as well as on RSD collections.

The power here is in providing common definition for skills within a particular context. This definition, or standard, can then be referenced in the metadata of Open Badges in the AlignmentObject field.

I realise that this sounds rather technical and dry, so let’s look at an example given by Nate Otto during a meeting of the Open Recognition workgroup as part of the Open Skills Network yesterday.

Skills represented in human and machine readable ways

Just like Open Badges, these skills descriptors can be read by humans in words and machines as code (JSON). Let’s look at an example of a skill.

The 'Communicate Time Zones' skill

There’s a lot more I could say about this, as there’s a real balance to be struck here between the flexibility that allows a thousand flowers to bloom, and a level of complexity that could stymie innovation.

One of the reasons that I’m moderately excited about the possibilities is that it slots neatly into that AlignmentObject field I mentioned above which is part of the existing Open Badges 2.0 standard. This harks back to work I was doing while at the Mozilla Foundation, linking Web Literacy skills to Open Badges.

Another reason is that Nate and the good people at Badgr are behind it. Not only have they got great people with the best interests of the ecosystem there, but they’ve also got the technical expertise to make it a reality. The next step is to get many, many OSMTs in existence so that we can decentralise the means of skill description!

Act NOW to prevent the hijacking of the Open Badges standard by an IMS faction!

TL;DR: add your comments to this GitHub comments thread to express your opposition to the Open Badges standard being merged with with ‘Comprehensive Learner Record’ (CLR). An overview of the better path forward can be found in this overview slide deck.

In March 2019, two years after Mozilla handed the stewardship of Open Badges to IMS Global Learning Consortium, I called for a ‘community renaissance’ free of IMS involvement.

I did not think then, nor do I think now, that IMS are a fit and proper steward for the Open Badges standard. Developing standards behind closed doors is antithetical that everything that Mozilla stood for when I was on the original Open Badges team. It leads to power grabs by small groups with interests unaligned to the wider community, and that’s exactly what’s happening now.

Over the past few years, Kerri Lemoie, Nate Otto, and others have attempted to steer a true course for the Open Badges standard towards the wider W3C Verifiable Credentials standard. (The W3C is the organisation responsible for developing standards for the web.) They have done this openly and transparently.

Behind closed doors, a faction of IMS members, perhaps wishing to hitch their bandwagon to the community-driven success, are trying to claim that the CLR is somehow the apotheosis of Open Badges. As anyone familiar with the last decade of the standard’s development will be aware, this couldn’t be further from the truth. Merging Open Badges with the CLR only serves IMS member interests: they would get to remove Mozilla’s trademark, shut down open repositories, and continue to ignore community involvement in the ongoing success of the standard.

As far as I’m aware, there is a meeting at IMS to move this to a member vote THIS WEEK — which does not give those with an interest and a stake in the ecosystem much time to respond. Update: Colin Smythe says “No deadline has been set. I would expect several more weeks of discussion and reflection. For the time being feedback should be to this GitHub repo.”

So please do consider going here and adding a comment. (If you can’t do that, please consider giving a ‘thumbs-up’ to comments with which you agree!)

To be clear, the proposal to merge with the CLR is an existential threat to the Open Badges standard. While v2.0 of the standard would continue to exist as a pre-IMS standard, there would be no future standalone version of the Open Badges specification.

I could make this post much longer, explaining how Open Badges are a great fit for Verifiable Credentials, railing about the US-centricity of the CLR, and complaining about the practices of IMS. But instead, I will end with an entreaty to add your comments to the GitHub thread.

Let’s focus on Open Badges as Verifiable Credentials and keep the momentum going. Let’s ignore the distraction of those wishing to limit the size, scope, and success of Open Badges to only those areas that they know well. Open Badges is so much bigger than one person, one organisation, or one sector.

Image by Patrick Loonstra. Cross-posted to Medium.