to this. I think that at Falcon we can put together an interesting presentation. As Content tagging is a feature that has been built from scratch, we could talk about it in general, about the architecture, all the different layers (from openedx-learning, to the frontend in the MFE course authoring), all the features that it is going to have and what is expected with it.
Another idea is to talk about the different versions of the storage backend of content libraries, modulestore (v1), blockstore (v2) and learning-core (“v3”), and how a migration between versions could be made. I touched this topic in a very superficial way in a theoretical way. Maybe someone else has more knowledge about this.
I like the idea of finding a way to talk about Listaflow at the conference, but it is important to make sure that it does relate to the conference topic. So if we were to do a presentation on Listaflow, it should probably be couched in how it’s used for the Core Contributor sprint planning. @Ali has recently been doing some investigative work on how the CC sprints are going, of which Listaflow is a part (though a significant one.)
Last year I submitted a talk about using Listaflow for all-remote teams-- I tried to frame it in terms of the fact that so many education teams are now global and remote-only or remote-first, and how Listaflow can help with that, but the reviewers noted that they weren’t sure it was quite relevant enough in the way I submitted it. So, if we do this we will want to make sure we have a solid pitch.
@rafay I also like the idea of giving Listaflow some limelight at the conference!
FYI Here is how it is currently being used for Core Contributor sprints:
Listaflow sends Core Contributors an email telling them it’s time to complete their sprint retro
They complete the Sprint Retrospective and Planning checklist on Listaflow
Dean (from ABC Online Courses) posts a summary of the feedback on the forum (here’s the latest summary)
This process may have changed by the time of the conference (we are hoping to improve reporting in Listaflow so that step 3 is not necessary. There may be other improvements by then too).
My ideas of topics I would like to research and present:
A presentation about Grove. Either of:
a. a demonstration/tutorial of how to use it for deploying Open edX, possibly showing some interesting use cases
b. an overview, including what particular features are more interesting
A presentation on XBlock development. Either of:
a. a demonstration of how to develop a new XBlock
b. a technical overview of how XBlocks integrate into the platform
(Sub-options depending on which one best fits the presentation time and format.)
A presentation about bringing type safety to the Open edX platform
why typing is helpful
how type checking works in edx-platform today (backend)
how to add more type hints and type checking
how to use type hints in your IDE
how to add type information/checking to frontend code (like we’ve been doing in Course Authoring MFE)
how to use those type hints in your IDE
how to add and enforce an OpenAPI spec for REST APIs
Join in any talk about Modular Learning, Tagging/Taxonomies, Copy/Paste, Content Libraries, etc.
Lightning talk: React Query - why you’ll love it and how to use it
a talk about how we can make XBlocks more mobile-friendly and MFE-friendly and/or Building XBlocks with React. (possibly focusing on neXBlocks proposal)
(Not me but I would love to hear:) a talk about how the move to Kubernetes has substantially cut our number of fires and improved uptime. Probably a lightning talk.
(Not me but I would love to hear:) a talk about our new k8s-based sandbox system
That sounds like something that Dave and/or Kyle from Axim would be likely to speak about :)
1.Talk about how we achieved automated multi-site, multi-theme MFE setup using Grove and the road forward.
2. How the Open edX retirement service works, how to set it up, how to un-retire a username using a plugin that we developed. (We already have a blog post to supplement this).
3. Autoscaling using Grove and Tutor. Different types of autoscaling (nodes vs pods), challenges we faced, and the road ahead using Harmony.
3. How authentication in Open edX works. How to configure third-party auth sources, how to use Open edX as an auth source for external services.
4. The XBlock runtime in Open edX and how we simplified it. (through BD-13)
+1 from me. This could be a great way to introduce the new sandbox system to the wider community, and show PR authors how they can debug and configure their sandboxes.
We presented a talk to introduce Grove in Lisbon. I think it might be a great idea to present a follow-up talk about the challenges we faced since, improvents that we have made, and the benefits that we are reaping (including the reduction in the number of fires).
Alternatively, we could structure this as a talk about Open edX deployments at scale using kubernetes, with Grove being a case-study, along with the challenges and benefits. (and also introduce Harmony)
These might be great topics to present jointly with community members outside OpenCraft.
Specially, we could collaborate on the mult-tenant setup talk with EduNext since they developed eox-tenant in the first place.
First, quick reminder that when I was ok with the talks topics suggested and approved them directly, I have clicked on the “like/heart” button for your post. If so, you can consider your talk proposal approved If not, I commented below to iterate:
@navin These topics sound like they have a lot of good base material for a talk, but you are missing giving an angle to it. You want to avoid simple “howto” or simple documentation/workshop material, which just walk people through how to do something (or how we did it) and instead give a title that will be thought provoking, or present a specific opinion, for example. As is, unless I would want to replicate what was done for those three topics, there wouldn’t be much reason to attend those talks? You need to pick people’s curiosity with something more specific or opinionated.
@yusuf Your topics have a bit the same issue - there is good talk material in there, but it needs to more directly pick the curiosity of the potential attendees, make them react when reading it. Maybe even a bit “clickbaity” (but with content to back it up).
For example, for the first one, maybe something more along the lines of “Forget about microsites! You need Django micro-apps for your next Open edX feature”. Or for the second one “A feature without test data is like code without tests - it holds you back”. You can probably come up with better titles, but maybe it gives you an idea of what I mean?
@maxim As noted above, target more talks than workshops (or lightning talks) - talks are harder to come up with but actually don’t end up taking as much time preparing as workshops, which are deceivingly time-consuming. Talks are also more interesting and bring a larger audience, in general.
It might be worth submitting something about grove, but it might be something to talk about under the lens of working with the rest of the community on supporting Open edX with Kubernetes through Harmony, which is going to be more useful and relevant to the community at large. Our experience with Grove can still be a way to give a concrete experience with Kubernetes & Open edX - especially if we give concrete numbers/stats about using it at scale.
Same issue as mentioned to @navin and @yusuf above - there is likely a lot of good materials in there, but you need an angle for the talk - otherwise it’s just a HOWTO/workshop. Maybe use the experience to give your opinion about whether using multi-tenants in Open edX is good today, or not? And if not, why?
+1! For the angle here, presenting the features that have been added recently in the platform & the general roadmap would be good (though we need to be careful to not be promising to do the work - maybe an additional angle would be “join the effort to improve Open edX theming” & maintenability improvements to expect from putting work in it?)
A lot of potential topics here - though they would all need to find an angle to not just be a HOWTO.
+1 to this topic - there has been a lot of work done, so it might be worth presenting the features that have been developed, and the roadmap. @braden I assume Jenna will be presenting something about this? It could be worth discussing a shared talk with her?
@maxim Sounds like a good one! And you could pick further the curiosity of potential attendees by taking a slightly tongue in cheek tone, like: “Courses Archeology - Follow the adventures of a developer uncovering the ancient ruins of courses archives from long-gone Open edX instances. Learn more about how past civilizations kept their knowledge, and about the dangers awaiting explorers who want to access their riches.”
Then the talk is going to be pretty short, isn’t it? :) Remember, as mentioned above, no original material development for the conference - there won’t be enough time for that, the talks themselves are already a lot of work.
@tikr Then maybe that material could be incorporated into another broader talk? Like for example the one about importing old courses, or anything related to content reuse, to provide an alternative approach/perspective?
@tikr Btw, for you in particular, it would be really interesting to get your experience as driving the OSPR process, along with Michelle (maybe a shared talk?). Some stats, details about what has been changed/improved, what challenges remain, where people in the community can help… It would be a good one to have strongly opiniated, especially about being blunt and direct about what needs to improve. I’m happy to contribute to that talk too, or review it.
@rafay It’s a good idea, though +1 to @Fox 's remark that it would be important to tie it to the Open edX project - which is definitely doable, as it is used for the Open edX Core contributor program. And rather than just presenting the software (which can be done in any case), adding an angle/opinion on top could help – for example something like “No more meetings! How the Open edX Core Contributor program and OpenCraft got rid of meetings - using software like Listaflow for doing sprint updates, retro and planning asynchronously”.
@Ali On my side the ticket to prepare the talk topics was in the sprint starting today, so I haven’t had a chance to post proposals yet - but I’ll do so today or tomorrow.
@artur See my comments above about grove, and generally for both those options they are too much a tutorial/howto - see also my comments about that above. Though XBlocks could provide a good topic, but maybe with a specific opinion/argument about them? Tutorials about them already exist, so it would need to add something to past presentations and existing documentation.
@kaustav Those have good base materials for a talk, but an angle/opinion needs to be added - like for others above it’s currently too much of a straight “howto”. The last one though has a bit of that in it:
or at least it could easily add an angle, using the excuse to present and explain the internals of Open edX – and the result being simpler/cleaner to understand, allow to use that newfound simplicity to get the audience to better understand how the platform is architectured.
@paulo See my comments above about Grove talks – a good topic, but with a few things to integrate. The angle of (2) load testing and scaling could stand out on its own, and actually be an occasion to talk about Grove and our experience, but from a point of view more likely to be interesting to the community as a whole (which doesn’t use Grove).
My talk proposals for this year, by order of prefence - the first and last one are carry-overs from last year, while the second one is new:
“An inside look at the governing body of the Open edX project: the TOC”
What is it, what can it do for me? Who does what?
This talk will review the history of the project governance, and how the TOC was introduced. We will present the type of activities that it follows, the decisions it takes, and who its members are. Also, we will introduce how it impacts the community and the Open edX as a whole, and how the community members can interact with the TOC. And as well, how you could very well join it soon.
It could be a collaboration between TOC members, presented as a team - I’ll suggest it to the group during the next TOC meeting
“Open Source Masterclass (In a Nutshell)”: Would you like to start contributing to Open edX? If you are a user or community member of the project and would like to get more involved, making your first contribution can be a daunting prospect - no matter whether your intended contribution is technical, or not. But fear not! This talk will propose a gentle introduction to the topic, and equip you with what you need to maximize your chances of success. It is based on the Open Source Masterclass MOOC, as well as the rest of the Open edX onboarding courses.
“Learning from open source communities” Open edX is one of many open source communities. What can we learn from others? Through a review of other large open source projects and of the literature on the topic, we identify where the Open edX project differs compared to the established practices, and what we can all do to improve our community.
I would probably submit all three, but let me know if you are interested in taking on one of those, or participating to it.
I see what you mean, I like the idea of making it more “clickbaity” and story like to draw more people in. Here are some things I’ve been thinking about to reframe these topics based on your suggestions. Let me know what you think:
Breaking the Open edX Monolith/Mono-Repo - add your next Open edX using Django micro-apps
Explaining the pains of dealing with edx-platorm’s large mono-repo and why we need to navigating away from
Introducing the power of django micro-apps and how/why to build major features with them
Separation of concern - the django micro apps should be solely responsible for a feature, providing a way to interact with it’s functionality once imported (api/rest)
Well tested isolated code - a lot easier to achieve vs large projects/repos
Faster/Parallel development - developers don’t step on each others toes
Show-case current example using openedx-learning django app, and how development is currently done on it
Feature Test Data - invest upfront and iterate fast with confidence
Many teams disregard spending time/resources on good test data generation that covers various scenarios because its “not the best use of time”
However it has many benefits:
Going through the action of thinking about what test data to generate often leads to discovering issues early
Once it’s done, everyone on the team can reap the benefits, and more time is saved vs have everyone manually create test/sample data to test the feature
Enhancing the sample data is often easier when its automatically generated, introducing things like randomization, or even more sample data is alot easier, and everyone benefits
Showcase how we did this with sample-taxonomy-data and how its being used by the whole team for continuous development and testing of new content taxonomy/tagging features.
@antoviaque I’m thinking the second topic might be too short for a full talk, so I think it leans more towards a lighting talk. How long are normal talks vs lighting talks in general in the conference?
@antoviaque Makes sense. I’ve decided to set aside the other two ideas and focus on the multi-tenant setup, as it appears to be generating interest both in this specific forum post and within the broader Open edX community.
Host & manage multiple Sites within a single instance of Open edX OR Navigating the Multitenancy Maze in Open edX for Resource Efficiency
How can we run multiple sites in Open edX?
Customize each tenant with their own themes and own list of courses.
Manage all tenants from one place.
Let me know if it still feels like a “howto” instead of a talk.
Remember to submit your talk proposals before 13 December (next Wednesday) to make sure you qualify for early feedback and a chance to resubmit. Submit your talk proposal at this link.
@Ali +1! And don’t forget to get a from me on the talk topic first. I would highly recommend prioritizing this work this week, as that could take some back & forth, and you will still have to write-up the talk proposals further to submit them (cf the fields form the form @Ali pointed at: Open edX Conference - 2024: Call for Proposals @ Sessionize.com )
@yusuf The talk about breaking the monolith is great! It might be worth reaching out to Dave Ormsbee to copresent if with him, he’s the specialist of the topic in the community - maybe with some help from @braden ?
For the second one, it’s true that it could be tricky to fill a full talk about it, unless you had plenty of interesting examples or stories to illustrate it. Depending on whether that’s the case or not, it could still be good to submit it as a full talk - there is usually a chance to requalify talks as lightning talks. I think you could also still make the title a bit more catchy, if you want to.
@navin I do think that it still sounds very much like a howto still, sorry How about something like “Multi-tenancy vs separate instances - which approach works best with Open edX?” And then in the description you can mention that you will give the pros and cons of each approach, explain scenarios when one or the other works best. And last, during the talk, you can actually spend a portion of the time on the points you have listed:
I’ll propose this one, since I’m still heavily involved in Aspects. Cristhian from eduNEXT has been the biggest Aspects contributor this past year, but he’s not able to attend.
@antoviaque Great! I’m leaning more towards this talk as well. The other one I feel would be too short as I don’t have many examples except the one I mentioned. It would be great to copresent with Dave and @braden if you’re up for it? What’s the best way to reach out to Dave to see if he is interested?
There are plenty of interesting ideas in this thread. The only thing I can think of adding to this is from my experience of dealing with edx-enterprise. Taking a step back from the “enterprise” aspect of it. It is primarily about packaging of content in interesting ways (mini-lessons from sections - we have a section-to-course plugin), grouping, sequencing (pathways, programs), customization of learning experiences (user-specific learning paths)…etc. However, we still haven’t done much in way of implementation yet. We would have done some work by the time for the conference and would probably be in the middle of it. So, this will most likely be more suitable to for the 2025 conference than 2024.
to this. I think that at Falcon we can put together an interesting presentation. As Content tagging is a feature that has been built from scratch, we could talk about it in general, about the architecture, all the different layers (from openedx-learning, to the frontend in the MFE course authoring), all the features that it is going to have and what is expected with it.
