Skip to main content

Workshop: Multiplier Event 1

Schedule​

  • Date: Thursday, October 27, 2022
  • Time: 9:00-14:00 EEST
  • Place: Room PR105, PRECIS Center, Faculty of Automatic Control and Computer Science, University POLITEHNICA of Bucharest

Initial Presentation: 9:00 - 10:30​

The team and community members will host talks regarding the purpose and aims of Open Education Hub. We will go over the following topics:

Each talk aims to take 10-20 minutes, followed by 5-10 minutes for questions and discussions.

Coffee Break: 10:30 - 11:00​

This will be an opportunity for more in-depth and informal discussions regarding the topics presented so far.

Demo of Open Education Hub: 11:00 - 11:30​

In this part we will showcase how to create a repository under the Open Education Hub organization using our templates and how to contribute to it via pull requests and in accordance to the Open Education Hub methodology.

Practical Workshop: 11:30 - 13:00​

Participants will clone and experiment with our first class: Operating Systems. They will build the lecture slides, run the demos and practice exercises and provide us with feedback based on them. For this, we ask participants to bring their own laptops. (Wireless) internet access will be provided.

We also encourage participants to prepare a few ideas about the repository they might want to create at some point within Open Education Hub. Please consider topics such as:

  • general theme and subject of your repository
  • main ideas / points you would like to deliver to future students

During the workshop, participants will put their ideas into practice and take the first steps towards having a full-fledged class that uses modern technology for efficient teaching.

Open Discussion over Lunch: 13:00 - 14:00​

Lunch is on the Open Education Hub team. As always, lunch will be a pretext for more open-ended discussions regarding the Open Education Hub project and the guests' further involvement with it.

Conclusion​

Takeaways​

  • The discussion focuses on our proposed methodology.
  • We follow the 5Rs: retain, reuse, revise, remix, redistribute.
  • The content should be open, easy to use, easy to update, easy to contribute. Most questions from the audience are related to how can they use our template repository, if it fits their needs and if the rules very strict (e.g. if they can use similar technologies to create the content).
  • Most task should be automated so the educator would spend their time on creating quality content, rather than spend it on setup actions which are usually small repetitive tasks, e.g. grading assignments, synchronizing grade books, creating exams, etc.
  • We provide a template repository that anybody can fork and customize according to their course.
  • Most questions from the audience are related to how they can use our template repository, if it fits their needs and if the rules are very strict (e.g. if they can use similar technologies to create the content).

TODOs on our side​

  • Update the OER Template Repository README with more explanations and intermediate steps.
  • Continue working on the infrastructure: integrate quizzes with automated assignment validation solution (VMChecker), make VMChecker more visible.
  • Make the entire project visible inside and outside our faculty.

Meeting Notes​

Project Overview​

Find the entire discussion from our meeting below in text format or on YouTube in audio format (Romanian only).

Speakers:

  • RD: Răzvan Deaconescu
  • RN: Răzvan NiÈ›u
  • TD: Teodor DuÈ›u
  • LB: Liza Babu
  • SW: Sergiu Weisz

Audience:

  • AO: Andrei Olaru
  • DN: Dan Novischi
  • BB: Bogdan BocÈ™e
  • GC: Gabriela Ciuprina
  • HG: Horia Geantă

RD: We want to see how realistic our idea is.

RD: Our content is often on a wiki, but is not easily reviewed / updated.

RD: I had problems. When suggestions arise, people send the fix to you, and you have to deal with it. We want to fix this.

RD: We don't have a collaborative way to do it.

RD: Folks from Singapore are using our content as it is, they don't have a what to update it for their needs.

RD: A good example of how it's supposed to be is the Operating Systems 2 repository and resulting site. It's a repository based on the Linux kernel repository that has our deployment infrastructure on top. Everything works in a VM, it is easy to use, update, and create snapshots. Students can use it as it is, and can come up with fixes via pull requests.

RD: This was our motivation.

RD: We usually receive feedback from students that we have outdated content, information that is repetitive or even mistakes. They do not have write access to OCW (our wiki) and they cannot be granted write access because OCW does not have a method for reviewing materials.

RD: Usually content is private or, if it is public, it's very difficult to update.

RD: It's not easy to create super high quality content, you need a lot of iterations.

RD: We want to automate as many actions as possible, everything should be in one place: in a Git repository. We want us, teachers, to spend more time mentoring students, rather than doing repetitive tasks that can be easily automated. Spending time on repetitive tasks makes us waste time, focus, and will to educate students.

RD: Usually content is difficult to update, it is non-digitalized, there is no such thing as versioning, it's difficult to distribute. GitHub makes it easy to contribute, easy to review, easy to report problems, and easy to reuse.

RD: We use a lot of technologies such as Markdown, reveal-md, draw.io, Git, ffmpeg.

RD: We want the evaluations process to be fully automated. We want to provide an infrastructure where students can practice.

AO: Where is the Linux Kernel Labs repository?
RD: Here it is, on GitHub.

HG: How much does your work so far depend on the fact that the content is in text format?
RD: A lot. We want the content to be, ideally, full text format.
DN: So this means you want people to ditch their old PowerPoint Slides / Google Slides / PDFs for this.
RD: Yes. It will take time and effort to migrate things and get used to it.

GC: How can we integrate equations into the new slides format?
RD: We have a solution for this. Our partners from Iceland are mathematicians and they face this situation as well. Markdown allows you to insert LaTeX sections.
GC: So if I already have then in LaTeX format, it's going to be easy for me?
RD: Yes, this is what they did.

DN: You mentioned some tools. Are we limited to those? Can other tools be used?
RD: Anything is fine. Ideally they should be open source (or free).
DN: Markdown is used to create other materials as well (e.g. Jupiter Notebook). I have a lot of such notebooks with code and explanations. Is there a way to publish them?
RD: There are 2 repositories: one with your content and one with the infrastructure that contains a publishing framework of your choice. We currently use Docusaurus for the Operating Systems class. For Linux Kernel Labs, we use Sphinx. Our partners from Iceland use TutorWeb. The storage format is the same. We chose Markdown because it's easy to use. Other formats can be used as well.
RN: Then, there are converters translate files from one format to another. Pandoc is an example.
DN: So depending on each course, the presentation can be personalized, but the storage format is basically the same.
RD: Precisely. We want to make things work. Teachers that adhere to our methodology should be worried about this. The one who publishes a course should do maintenance for the content.

RD: Students consume the information, and they can evaluate themselves because they have lots of drills and quizzes there.

RD: We would like to stimulate students using SmileyCoins, an educational cryptocurrency. The idea comes from our partners that provided SmileyCoins to children from Kenya that solved math problems. It can be difficult because students might end up doing this only for compensation. We have to think a lot about it.

RD: Not only teachers, but trainers and external people can use the materials. They can take the infrastructure and use it.

RD: Ideally the entire content would be in English so it can be international.

RD: We will test our ideas on 3 lectures: Operating Systems (Romania), Security Summer School (Romania), and Computing and Calculus for Applied Statistics (Iceland).

Methodology​

RN: We want to have guidelines, principles, rules so that the content would be interactive, easy to automate, easy to contribute, easy to access.

RN: The educators don't have to be the source of knowledge, the information is already there. The educator should rather help with soft skills.

RD: The educator should also correct bad habits.

RD: The student has extra questions, the educator should be there to provide answers.

RN: We use a versioning system named Git. We define all rules there.

RN: Any person with an internet connection can read the content, solve problems, and contribute. The purpose of this methodology is to tell us how to contribute.

RN: It also offers guidelines on how to create high quality content.

RN: A course has the following components: the lecture, the practical sessions (labs, seminars, discussions, debates, etc.), drills (a big database with exercises), assignments / homework (much more complex exercises), exam / evaluation.

RN: Anybody can evaluate himself, not only students enrolled in a course.

RN: The lectures should be interactive. For this, there are quizzes and questions.

DN: Is the methodology very strict?
RN: It's up to the content creator (educator), but we create these guidelines with what we consider useful and good.
RD: We want to provide everything that is necessary for the educator to do its work. The educator can adapt everything according to his style, his students, his course.
RN: The methodology does not teach the educator how to teach. It teaches him how to create high quality content.
BB: The methodology should suggest things to the educator, not to force him.
RD: Yes.

RN: The methodology will also suggest how to create good slides (e.g. replace any piece of text with an image, if possible).
RD: Iterating helps. These slides are really bad, but there's something. We can make them better. We at least have something to start with. It's more difficult to go from 0 to 1, rather than go from 1 to 10.

RN: During practical sessions, students will be "forced" to read the entire content in order to solve exercises and quizzes. Before this, the labs were usually structured with the theoretical part at the beginning and the practical part at the end, so students would just dive into the exercises without reading the theoretical part.

RD: The TA should answer questions.

RD: If you answer the same questions for 10 students, the answer might need to be in the content.

RN: For courses that are technical, exercises should be graded automatically.

RN: Ideally, the practical sessions should not be part of the final grade. The practical sessions exist to help student learn, not to grade student work.
RD: This is a suggestion / opinion.

RN: TutorWeb generates questions based on previous answers. If a student answered correctly a difficult question, then the student will receive a more difficult question.
BB: Does it use AI?
RD: We don't know, it's a good idea. It can be integrated.

RN: Assignments should be automatically graded as long as it's possible. We shouldn't spend extra time on this. You should ideally score anything between 0 and 100 points, not only 0, 50, or 100 points.

RN: The exam should also be public, since everything is open source. This is not very common, but all the questions and answers should be public, the exam is in a controlled context, student wouldn't have access to the repository while taking the exam.

RN: Ideally, the exam should be practical, especially for programming-related courses.

RN: We would like to offer cryptocurrency for contributors and for elegant solutions to quizzes / assignments. SmileyCoin is a cryptocurrency that doesn't have much value, but can be used by anybody. You can use them to buy educational materials, entries to summer schools, books, etc.

RN: In Iceland, students can use them at the university cafeteria.
BB: It's offensive to offer food to those that are more intelligent.
RD: It's complicated, people would try to violate the system.

Content​

TD: Our methodology is based on open education resources.

TD: The 5 Rs - retain (everybody has a copy of the repository), reuse (educators can use the materials created by other educators), revise (anybody can create pull requests for improvements), remix (trainers can mix the existing content with their own for another flavor of the material), redistribute (the information can be easily shared). This is what we want.

TD: We use GitHub. This is the most popular platform that integrates Git. We keep everything here. It's simple to contribute and review changes.

TD: Quizzes are stored in Markdown format and are published in multiple formats (after transformation) - Moodle, TutorWeb, etc.

TD: A quiz has at least 3 sections: the question itself, the possible answers the feedback (given in case the student doesn't answer a questions correctly). This is similar to Moodle. The advantage is that other sections can be easily added: difficulty score, tag, keywords, etc.

TD: We need a conversion method from Markdown to other formats. We are creating a tool for this.

TD: There are 2 types of images: SVGs (handmade, diagrams) exported from draw.io and screenshots (static, difficult to update). We prefer SVGs because these can be modified without much effort.

TD: For videos we have screencasts (asciinema, mp4, .gif) and handmade (a set of SVGs converted into GIFs using ffmpeg).

Community​

LB: Up to ow, the information was a commodity. I have this commodity, I give it to you, the transaction ends here.

LB: For example, I have a watermelon, I give it to Răzvan, now he has a watermelon. We didn't learn anything new from this. I'd rather show Răzvan how I harvested this watermelon so he can do it on his own. Răzvan will naturally ask questions, I won't know the answers to all of them, so I have to do my research. We will both learn something from this experience.

LB: Now, information is hard to access, private and usually one-way.

LB: We want it publicly available, both-way, we need a big community outside of university.

LB: We currently don't have acquaintances in other countries (apart from Iceland), so we have to grow the visibility of this project.

LB: Ideally, in the future, we would have a large number of repositories, each having maintainers, contributors and users.

LB: We have multiple resources - Discord, GitHub, site, Google Groups, Twitter, LinkedIn, Facebook, Reddit, YouTube, Instagram.

BB: You said that there should be a sort of guideline of what is acceptable for each repository, who is responsible for creating this?
LB: The maintainers, they know the base idea of the repository and know what is acceptable and what is not.

Infrastructure​

SW: A problem in our faculty is that educators spend too much time with setup actions. These things do not help us educate students better. They only take time, so we have less time to spend thinking about how to educate others better.

SW: The difference between universities and Coursera is that we have a mentoring component.

SW: Our goal is to automate the setup actions as much as possible. The educator creates a new instance for his course, updates it every semester with minimum effort, and publishes the content into his desired format.

SW: We want to create a simple method for educators to do so - for both technical and nontechnical people.

SW: We want to reuse as many already existing solutions.

SW: We absolutely need to avoid getting into a situation similar to VMChecker that is maintained only by Răzvan. There is a single point of failure, any problem has to go to Răzvan, and it's complicated.

SW: Keeping the infrastructure up, generating homework, grading homework, checking quizzes are all things that take time.

SW: An ideal course is described using a GitHub repository, it's open source, students can access materials easily, and they can contribute and correct any problems they might find. We have an example of this model in the Operating Systems course.

SW: The content is all on GitHub, in Markdown format.

SW: Anybody can create a fork and update the content.

SW: If this were to be on OCW, it would be difficult for me to update anything there. I don't have write access to the Operating Systems class on OCW. I can copy and paste the text (without the wiki format), but I cannot see the source of the page.

RD: The problem with OCW is that it's both a storage and presentation engine. Ideally, they should be separate engines.

SW: This tutorial is available here.