Episode 53: Analogien, Metaphern und Studien

In Epsiode 53 kommen wir nach etwas längerer Auszeit wieder zusammen und entsprechend dauert es etwas länger zu besprechen was wir seit der letzten Aufzeichnung gemacht haben. Kapitelmarken helfen denjenigen, die sich dafür nicht interessieren. Eine Besonderheit dieser Episode ist, dass wir uns vornehmlich mit der Machbarkeitsstudie für eine Plattform für Hochschullehre und dem von uns eingereichten Kommentar befasst haben. Feedback und Anregungen gern bei uns im Blog.

Was wir trinken

Markus: Kaffee und Wasser
Christian: Kaffee und Wasser
Das Trinken von Bier übernehmen inzwischen die Kolleginnen und Kollegen von Bldg+Alt+Entf
Marke 1:25


Marke 7:20

Was wir gemacht haben

Marke 17:05
Marke 27:35

Was Markus gemacht hat

Marke 37:40

Was Christian gemacht hat

Marke 52:30

Was wir gelesen haben

Marke 1:52:55

Was wir tun werden

  • Beiratstreffen OER-Infostelle
  • OER-Camp West 

2 Kommentare bei „Episode 53: Analogien, Metaphern und Studien“

  1. Hallo Leute, danke für eine sehr spannende Episode. I’ll write in English since it’s a bit faster.

    First, here is what I was doing while listening to your podcast (images and statistics :)) https://imgur.com/a/HaukCTY. Maybe you could ask people to take a picture of where they are/what they’re doing while listening to the pod. I found it interesting to imagine the guy who was exercising.

    Anyway, I haven’t had time to look at the Machbarheitsstudie myself, but I found the concept interesting. I’m curious about what they defined as the needs or gaps in current platforms… Was it more about a discovery platform, in that case, you don’t really need to host the content yourself – class-central.com is a great example of aggregating metadata and ratings of courses from many MOOC platforms. (And if the platforms were open, and allowed deep-linking to material without signing up, one could do much richer indices, something like the Saylor Foundation curricula https://www.saylor.org/, or the “knowledge graph” I saw recently for Computer Science, where they showed knowledge pre-requisites in a graph (to understand graph theory, you should understand automata, to understand automata you should understand binary etc – with each topic linked not just to specific videos, but specific locations in videos (this lecture, minutes 10-20))…

    If the goal is to have a platform to host content, again the question is what is the need? France was mentioned, they have France Universite Numerique – but this is hosted on EdX Open – China also has an EdX Open instance (as well as many homegrown MOOC platforms), we now have a national EdX Open instance in Switzerland (hosted at EPFL, financed by SWITCH) etc. If one is not content with the current state of EdX Open, Moodle, the quite interesting new ELMSLN (http://www.elmsln.org/ – see the very energetic Brian Ollendyke https://www.youtube.com/user/btopro), one could spend some millions contributing to these platforms, benefiting German institutions and the world-wide community…

    Also curious if there was much discussion about epistemology and pedagogy in the document – I know you felt it had a very static approach to education, but even if it doesn’t treat AR and AI, cMOOCs are nothing new, but fundamentally different from xMOOCs (and many other things – there have been radical distance courses by email from the 1970s!)… The idea of having a platform that let’s you author courses is not that interesting to me – although certainly useful as a base, but here I think EdX, OpenStax, or perhaps the new ELMSLN stuff (or just a MediaWiki!) is perfectly adequate. (If not, why?)

    What’s more interesting to me is tools that allow students to create, share, discover, collaborate, discuss, act, participate… And where I do think there is a massive need, is coming up with better standards for interoperability of these tools… We have LTI, but it’s extremely limited (doesn’t support data in from other tools, data out, configurability (although there is apparently some support for this through deep-linking, but I don’t see many platforms supporting it), live streaming learning analytics, live collaboration with multiple students, etc.

    The tool that we’re currently developing aims to be a platform where it’s easy to author new tools supporting all of the above (https://github.com/chili-epfl/FROG/wiki), however I don’t want to reinvent the wheel, and I don’t want to have to recreate the hundreds of PhET physics simulations, or all the other great tools out there… However, if I embed PhET activities in FROG right now, they become IFrame black boxes, that cannot be deeply integrated into the pedagogical flows that we are authoring… So we are experimenting with richer protocols for integrating activities – I laid out some early experiments in the last part of this talk: https://www.youtube.com/watch?v=5IHsXCBcbaI … inspired by H5P activities, which streams learning analytics to the parent website, we’ve been experimenting with allowing embedding of FROG activities in other tools, etc.

    I know you are skeptical to SchulCloud, because of some of the bombastic statements from Dr Meinel etc, but it seems like they are also trying to integrate a lot of existing tools – but I don’t know much details (will be visiting them soon, maybe I’ll find out more).

    Anyway this is the kind of area where invest could have a big impact… Another thing we are trying to do is to make it easy to take algorithms and trained ML models, or rich visualizations, which usually end up in papers at conferences like LAK, EDM and AIED, and actually make them directly usable by students. A simple example is here https://www.youtube.com/watch?v=jI8A4Mu_qrQ (start at minute 5), where we have four students providing different answers to how a university should apply sustainability – and they are automatically grouped by the semantic similarity of their answers, to first discuss in similar groups, and then in opposing pairs (a kind of natural jigsaw design). To me, this ability to both bring research into production, but also giving teachers access to cutting-edge algorithms that are open source, with data hosted locally, in the age of big internet companies, and hundreds of ed tech startups, all with competing products that don’t integrate, and where the algorithms are black-boxes and proprietary, is very important.

    Looking forward to the next episode, and another long walk!

    1. Hi Stian –

      First of all: huge thanks for adding your thoughts here, much appreciated! And I like how you manage to add a sense of aesthetics and beauty (as proven with your image) to our podcast, which is often quite the opposite to that.

      My impression is that the overall goal of the platform currently is not to be left behind by other nations who operate their own platforms. Whatever that means, I thought it was hard to find a vision for a concept of a German platform. As in: “Spain, UK, USA and France have their own platforms, now we want one, too!”

      There’s also the notion to enable students to take courses online at other unis and transfer credits to their home university. This could be interesting and actual progress, but might not necessarily need a central platform.

      And I agree, there’s plenty of open source solutions that enable universities to author and provide courses, they could join forces to do that today. But only few German unis are doing that, so where’s the need?

      My reservations regarding the Schulcloud are multi-layered and refer to governance, but also to the neglection of pedagogical concepts as inherently woven into any learning environment, the idea of creating an app store rather than a learning environment and, of course, the hubris that comes with claims such as “there has been no tech in schools before us”.

      Thanks for all the links to your current projects, I have to dig deeper into these and hope that we’ll find a chance to chat soon, maybe next week in Hamburg.

      Take care, Christian

Schreibe einen Kommentar