I’m the co-founder of BigBlueButton, an open source virtual classroom we’ve been building since 2007.
About three years ago, we integrated tldraw into BigBlueButton as our whiteboard. It’s been an excellent upgrade over our old, simple whiteboard — tldraw is a fantastic project.
I'm also the CEO of Blindside Networks, the commercial company behind BigBlueButton. We have growing by the traditional open source business model: we offer hosting, engineering services for acceleration of features, and support contracts.
I understand the motive behind tldraw's change of license. Open source projects often get asked two contradictory questions:
1. Can I use your work for free?
2. Can you guarantee that you’ll be around in 5 years?
You can’t answer (1) without a solid plan for (2). Licensing changes are one way projects try to answer both of these questions.
We are no stranger to license changes, we recently rewrote the entire back-end of BigBlueButton and moved away from mongoDB to PostgreSQL + Hasura.
For us, moving to tldraw 4.0 would mean:
- As Blindside (the company): buying a commercial license — that’s straightforward as we are also a commercial company.
- As BigBlueButton (the open source project): it would require every organization running BigBlueButton to obtain its own license key to tldraw.
There are pros and cons here. We want a world-class whiteboard in tldraw based on a sustainable open source project, but we also want to keep BigBlueButton’s community deployment model simple.
Curious how others in the HN community have handled integrating source-available components into open source projects. How do you balance sustainability with accessibility?
Answering those two questions depends a bit on the "you" in the question. If "you" refers to the open source project/code, then both can be answered with a resounding "yes". If on the other hand "you" refers to you individually or as a company, then the first can be answered with a resounding "no" and the second a solid "maybe, that depends on how much you are willing to pay us for a 5 year support contract" (though you should probably word it a bit differently when talking with potential clients).
As far as working with source available components, suggestion one is to look for others int he community that you can cooperate with to maintain a fork, and option two, if you really can't get the community to support a fork, is to make it a plugin/optional component, preferably with an API so that other solutions can be integrated as options, or at least a fallback to the old version that was open source.
Seriously in this specific case I'd cut a final version with V3, then going forward include v4 in main with a note about the tldr licence in readme. The licences are affordable for anyone who wants to use your project, and if they can't they'll have to make do with that old unsupported version of your project.
How is $6000 per year (upfront) affordable? And that's the pricing for startups with <10 employees, surely large organisations / schools (which seem to be the audience of BigBlueButton) will pay much more. What about non-profits etc?
At this point tldraw v4 seems to offers no significant improvements over v3, which does everything I want from a whiteboard. I think most people who can live with the previous watermark will just stick with v3 permanently.
If you've got lets say a 6 person start up your wage bill would be ~$1 Million a year, $6000 feels quite reasonable. I agree that this seems pitched towards "we're making a saas" style startups and that for direct end users this is more difficult however if you can perswade tldraw to price this based on the IT staff your school has then if you've got a group of 5 schools each of 2000 kids, you probably have a IT team of 8-10, $500/month is $0.001c per child per day.
> We have growing by the traditional open source business model:
I know this is not relevant to the thread, but could I pick your brain on this model? I'm looking at launching a product soon and I've been struggling with how I might monetize it in a sane manner that works for customers and for the business.
Really, that was your question? I think I can see how you got into a position where you are launching a new product soon and you don't know how you're going to monetize it
Sometimes a conversation is easier than a brain dump in a text only space. Which is why I said 'can I pick your brain' rather than write a massive wall of text of things I've considered, research I've done, product/market fit, etc.
> (perhaps frustratingly for the student), force them to iterate through something to get an answer.
IMHO, I think feeling frustration is the whole point -- it's how our brains rewire themselves, it's how we know we are learning, and it's how we build up true grit to solve harder problems.
As we want to "feel the burn" in the gym, we want to "feel the frustration" when learning.
I re-read the abstract and they tried two different modes of ChatGPT-4, "base mode" and a "tutor mode". The tutor mode helped students more, but it cautioned at the end:
> Our results suggest that students attempt to use GPT-4 as a "crutch" during practice problem sessions, and when successful, perform worse on their own. Thus, to maintain long-term productivity, we must be cautious when deploying generative AI to ensure humans continue to learn critical skills.
I think the caution is the use of AI to circuit the real learning, even if AI is in a tutor mode, to avoid building up true grit.
Ultimately, in writing this article, my hope was that a student would read it and get angry, angry that over use of AI - using it as a crutch - is actually having a negative impact on their learning, and they would resolve to using it only for more efficiency and effectiveness, not a substitution for the true learning.
I was thinking of Richard Feynman’s approach to learning when writing this article. He was a genius, so I didn't want the analogy to be unrelatable. However, he really enjoyed understanding the first principles and that enjoyment gave him such a solid foundation. He put in the necessary hours to learn, and what a remarkable life he enjoyed because of it.
Say what you will about Oreos and other processed foods, but they do actually contain calories. They are legitimately food.
Here's my experience as a professional educator: AI tools are used not as shortcuts in the learning process, but for avoiding the learning process entirely. The analogy is therefore not to junk food, but to GLP-1, insofar as it's something that you do instead of food.
Students can easily use AI tools to write a programming project or an essay. It's basically impossible to detect. And they can pass classes and graduate without ever having had to attempt to learn any of the material. AI is already as capable as a university student.
The only solution is also hundreds of years old: in person, proctored exams. On paper. And moreover: a willingness to fail those students who don't keep up their end of the bargain.
Oreos are food, but only good in controlled quantities. During covid, many of my co-workers cited putting on extra weight as they were unconsciously snacking on junk food when working at home. It was just to easy to have another bite when the plate of food was next to their mouse.
For learning, I think having an Oreo cookie (using AI) is OK once in a while, especially if your hitting a wall and can't get through, but it's a really, I think, a very steep slippery slope that leads to avoiding the learning process altogether.
I remember as a co-op student spending three days solving a particularly subtle bug in a C-based word processor. My grit was rewarded. On day three, I vividly remember staring at the code and the solution just popped into my head. That was one of the most formative experiences in my earlier years as a developer and feeling of elation never left me. I worry that AI will take these moments, especially early in ones career.
Our brains have not changed in hundreds of years, and I agree that the in person experience is actually the best. Humans learn best from humans. I'm trying to learn French, and Duo has been sad for a few weeks due to my absence, but its not having the same effect on me if it were a human French teacher was was sad with me.
Regarding failing students, I personally had to take summer school twice and still ended up failing grade 12 and repeating the entire school year. Why? I was too focused on computers and nothing else. In retrospect, taking summer school and repeating grade 12 actually helped me catch up at time when the stakes were low. If I hadn't, I would have definitely failed later in life when the costs were higher.
It doesn't even have to be on paper. That computer science exam can be done on a monitored university computer. The only bit that needs enforcement is not using outside resources, to show that a particular knowledge and how to apply it is actually in one's head.
Here in czechia they still do oral exams where you sit in a chair in front of an instructor and they ask you questions which you have to answer by speaking. I don't think there's any better way to show actual content mastery than that
Jo, bohužel takový postup nefunguje tak dobře při větším počtu studentů. I kromě toho problému, jak najít čas na osobní výslech každého studenta, musim navíc vymyslet pro každého úplně nové otázky; jinak předchozí student prozradí obsah zkoušky následujícím. Asi v Česku máme menší třídy?
I was talking to a high-school English teacher recently about building oral exams using ChatGPT voice-mode. Current models would struggle to provide a uniform experience across students, but it feels like it's within near-term reach.
I think htere is a future for AI tools in learning, and I think they could be hugely valuable, the problem is that we aren't teaching anyone (or learning as a society) how to use these tools well, and using them well will require discipline.
In education today there's a lot of focus on knowledge and testing, and therefore it's fairly natural for AI to be used to just answer questions instead of as a learning aid. If we had a focus more on understanding, I'd hope that use of AI would be more exploratory, with more back and forth to help students learn in a way that works for them. After all, if LLMs are basically just text calculators, every student having a concept explained to them in exactly the language they need would be pretty amazing.
It's a good one! I'm a lifelong fan of the leveling-up techniques you're talking about and I found they're essential when working with AI agents, especially.
I had the epiphany that all of the "AI's problems" were problems with my code or my understanding. This is my article[0] on that.
I'd like to echo the impressiveness of tldraw. At the BigBlueButton project, an open source virtual classroom, we built tldraw into the core. It has saved us countless development hours as we stopped trying to build our own whiteboard and instead stood on tldraw's (very) wide shoulders. We've never looked back.
I am the Product Manager for BigBlueButton. We have been building out the project for the past 12 years to focus on one market: online learning.
With Covid-19, it has been a pretty intense 2020 for the BigBlueButton project
In 2020, we’ve seen some large deployments of BigBlueButton (100's of servers serving thousands of simultaneous meetings). Such deployments do not deploy a single large core machine; rather, they deploy many individual BigBlueButton servers in a pool (see https://github.com/blindsidenetworks/scalelite). We don't have BigBlueButton structured where you can split across the components on separate servers: the recommended deployment is each server is a self-contained BigBlueButton installation.
There is a large community of users around the world using BigBlueButton. We, the core developers, have been focused on ensuring that BigBlueButton is solid and scales horizontally. We released 2.2.0 on March 29, 2020 and the current release, as of Nov 11, 2020, is 2.2.29. Those 29 iterations reflect a strong desire by the core development team to respond quickly to the demand and make BigBlueButton the best possible solution for virtual classrooms.
Regarding the packaging, behind the scenes there are a bunch of scripts that I wrote using fpm to build the packages, and it's all non-standard, has grown organically over the years, and needs to be updated.
We had planned to clean up the packaging in 2020 and include it in the core it so others can build and build upon it, but all that got put aside when Covid-19 hit and there was so much demand for the software. Instead, we focused all our development resources on improving the system for end users (security, stability, usability, and features).
We are looking forward to providing standard Debian packaging scripts for BigBlueButton in 2021, alongside making continuous improvements to BigBlueButton.
Our goal remains to make BigBlueButton the best virtual classroom system for online learning.
I am the product manager for BigBlueButton. While we implement most of the capabilities you would expect in a web conferencing system, we focus on giving the instructor many ways to engage students for learning. Being open source has enabled many schools around the world to setup and run their own BigBlueButton servers. Thanks to our community, we're localized in over 25 languages, provide a pure HTML5 interface, and have been deeply integrated into many of the most popular learning management systems. Our road map will continue of focus on the teacher/student engagement. Needless to say, Covid-19 made a lot people take a closer look at BigBlueButton. We've been working on it for 10+ years now, and we're very determined to make it the most effective platform for virtual classrooms and build upon our community.
Some of the folks in the Debian community have been trying to package it up in Debian and found it a bit challenging that some of the dependencies of BBB work only with specific Ubuntu versions. Would you please help address that and help them move forward with packaging it up and include it in Debian? Thanks again.
We're in the process of moving away from an internal build system for the packaging to having the debian package scripts as part of the repo. This work is underway for BigBlueButton 2.3 (the next version) and beyond. Once we get them released, it's going to be a lot easier for others to build and contribute to the packaging.
I grew up on Byte magazine. Love the covers and the articles on programming. My parents had bought me a TRS-80 Model I with Level III Basic and learned programming by typing in programs (and then debugging them) from Byte and other magazines. Did anyone ever type in the Scott Adams Treasure Island adventure game from the December 1980 issue? https://archive.org/details/byte-magazine-1980-12
OK, have to say this is pretty cool. Reading about how to do something is good, but ephemeral. Applying what you are learning (and making mistakes and figure out what went wrong) is a far better way to learn. Kudos to the author of this tutorial.
I must have ready ZMM at least seven times (so far) in my life. Back when I was taking an undergraduate, I read it the first time and was inspired to take as many English courses as I could. I wanted to be a technical writer. After a series of co-op work-terms in the field (the companies loved a tech writer who could also program), I landed a full-time job as a technical writer in a large telecommunications company. I would read ZMM on the bus to work for inspiration. Pirsig could write with such clarity that I tried to emulate him in my writing (as I'm sure all poor writers do). I eventually returned to programming as it was my first love. The job as a tech writer definitely improved my writing skills, and reading ZMM definitely improved my life.
Pynchon worked as a technical writer for Boeing, and Vonnegut worked as a publicist for GE. You can see how their appreciation of tech shaped their future work. You're in good company!
About three years ago, we integrated tldraw into BigBlueButton as our whiteboard. It’s been an excellent upgrade over our old, simple whiteboard — tldraw is a fantastic project.
I'm also the CEO of Blindside Networks, the commercial company behind BigBlueButton. We have growing by the traditional open source business model: we offer hosting, engineering services for acceleration of features, and support contracts.
I understand the motive behind tldraw's change of license. Open source projects often get asked two contradictory questions: 1. Can I use your work for free? 2. Can you guarantee that you’ll be around in 5 years?
You can’t answer (1) without a solid plan for (2). Licensing changes are one way projects try to answer both of these questions.
We are no stranger to license changes, we recently rewrote the entire back-end of BigBlueButton and moved away from mongoDB to PostgreSQL + Hasura.
For us, moving to tldraw 4.0 would mean:
- As Blindside (the company): buying a commercial license — that’s straightforward as we are also a commercial company. - As BigBlueButton (the open source project): it would require every organization running BigBlueButton to obtain its own license key to tldraw.
There are pros and cons here. We want a world-class whiteboard in tldraw based on a sustainable open source project, but we also want to keep BigBlueButton’s community deployment model simple.
Curious how others in the HN community have handled integrating source-available components into open source projects. How do you balance sustainability with accessibility?