The state of open educational resources in 2017

I spend the last couple of weeks exploring the open educational resources (OER) landscape and wanted to write down my thoughts and observations about the field. The promise of an OER “revolution” that will put quality learning material into the hands of every student has been around for several decades, but we are yet to see OER displace the established publishers. Why is it that “open content” hasn’t taken off more, and what can we do to make things happen in the coming decade?

It’s easy to become over-optimistic about the possibilities afforded by new technologies in the educational space. It’s important to keep in mind the tremendous friction associated with adopting new tech and new ways of teaching and breaking into the established educational practices.  Let’s “aim low” and set realistic targets that can easily be achieved, even in the face of resistance from current stakeholders in the system. What is the common core of useful tools and practices that everyone can agree on that we can achieve in the next three to five years?



Improving access to education is a win-win for everyone. In fact, education is so important that it doesn’t make sense to have commercial interests mixed in with the production of educational content. Whenever possible, students, teachers, and schools need to be in active control of the content, rather than being passive consumers.

The ability to continuously update and improve the content will lead to higher-quality content in the long term. That is, of course, assuming the barriers to collaboration are sufficiently low to allow for productive collaboration to occur.



There are two distinct categories of educational materials we can identify.

  • Content: text, exercises, images, videos, audio files, simulations, interactive demonstrations, links
  • Structure: curriculum, course plan, lecture plan, prerequisite structure, and topic guides

Existing OER efforts [1,2,3] are predominantly focused on the production of content items, but there are also recent efforts that provide entire curriculum including course structure, lesson plans, exercises, etc. The common core state standards for math provide an overall structure so that content from different sources can be used interchangeably. In general, it’s not clear that content from different OER sources can be easily remixed and combined.



Let’s review the main players in the “education space” and point out the roles they play.

  • School boards administrators oversee the operations of the “education machine” and are most interested in quantifiable results and metrics. They are interested in new technology and resources that will improve educational outcomes, but would be very slow to act, because of the red tape involved in large administrations.
  • School administrators operate under a lot of financial pressure and school board administrators breathing down their neck. They must also balance parents’ demands, keep their teachers happy, and make sure “operations” continue smoothly.
  • Teachers are the foot soldiers. They are faced with the daily tasks of keeping students learning, policing the classroom, and surviving on a limited teacher salary. They are usually not very technologically savvy, an won’t instantly adopt new tools, unless they can clearly see the benefits.
  • Students are the patients in the system. The whole point of the system is to support the students learning process, but they are given very little choice, and have very little control of their learning.
  • Parents often want to get involved in their kids education. Some parents have the ability to help their kids, others less. Some parents are willing to get involved in school projects and contribute their time.
  • Content producers like book authors, editors, and publishers are responsible for getting “educational products” shipped. From textbooks, to exams, to exam prep courses. The old players in the game (textbook publishers, educational content providers, exam sellers) are large organizations that make their profits through large deals with governments and school boards. The actual content authors are not necessarily concerned with the quality of the content they produce, and are often far removed from classrooms.
  • Technologists working in EdTech startups hope that bringing technology into classrooms will improve learning outcomes. The profit motive leads to a lot of technological innovation, but selling products and content to schools is notoriously hard so very few EdTech startups have made it big.
  • Governments usually want what is best for the population, but they are so far removed from the action that they are unlikely to know the best direction to move in. They are also likely to be lobbied by special interest groups.
  • Private donors often fund educational initiatives and non-profits. These people have their heart in the right place and want to encourage new efforts to improve the status quo. The ad-hoc nature of their efforts can often lead to duplication of effort.

Observe that the “main actors” in the educational system—students and teachers—are only a small part of this entire pipeline, and have very little say about how things are organized. The need to standardize the educational system has taken away the power from the teacher, and the need for one-size-fits all education removes student’s joy of learning. This sucks.


How can we fix this?

Who will build the educational system of the future? What will it look like? How can we put teachers in control of the educational content? How can we make students take ownership of their learning? These are important questions whose answers are far from simple, or obvious. The best I can do at this point is point out some high-level themes and directions that I believe could lead to useful results.

  • Better textbooks. I’m totally biased about this topic, but I think learning any subject benefits tremendously from following a structured sequence of lessons—something like a book. I’m talking about printed books, but it’s possible in the future the book will evolve to new formats. Regardless of the medium, the key benefit of the book-like “main text” is to lead the student through a path and, like a story, with a beginning, a main part, and an end.
  • Better support materials. Video lessons, exercises, images, demonstrations, and other “bonus material” are excellent for reinforcing the material students will be learning. It’s totally up to the student to pursue the subject at any depth they choose.
  • Project based learning. Instead of being passive “receivers” of information, students can work on personal projects related to to the subjects they’re supposed to be learning. Such projects will help them become autonomous learners.
  • Better authoring and collaboration tools. If we want the vision of quality OER produced by the community to become a reality, we need to build the best tools possible for authors to collaborate. Teachers must be able to submit typos and fixes. The best place to store educational content is github. Github and the pull request model is a proven technology. Perhaps a domain-specific layer on top of github could make things simpler for authors?
  • Incentives for authors. This is the hard one. Why would a teacher or author dedicate hours of their busy life to create or improve OER content? The situation is different from software, where people make contributions to fix bugs and solve their own problems. It’s comparatively more work for someone who wants to use the book to customize or edit a book. And if they edit it, what are the incentives to contribute back? And for the main author who receives contributions, why would they incorporate them?
  • Free software that runs on school premises through which students can access educational content. If it the “OER platform” is sufficiently easy to use, then school administrators will have to put in only one new system, which will act as conduit for all OER, and other resources. If a single “school server” system authentication, authorization, and student data, then this will greatly simplify things. The fact that it’s free should make it very competitive with commercial offerings.
  • Converter between different OER formats. It’s clear that every OER repository out there has useful content, but how can we use a mix of content from different sources? We need something like pandoc for OER.

I hope that the open source model will prevail and OER will be adopted as the predominant source of educational material in the future. I think putting the content authoring and editing tools in the hands of teachers, parents, and students is the important focus point that will enable all the rest. Paraphrasing the lyrics of Rage Against The Machine, it’s not about books, but about the machines for producing them.


Obstacles to OER adoption

The reason why OER are not used most often are as follows:

  • Content quality. Many people thing OER resources are of inferior quality compared to commercial materials. Certainly OER has “high variance” with some content being good and some being bad. In general, whenever OER content tries to reproduce the pedagogical approach of mainstream textbooks (teaching procedures instead of explaining and focus on repetition), the OER will be as low quality as the mainstream material.
  • There is a perception of insufficient resources. It’s possible to find 70% of what you would want to for your class, but be missing the 30%, which makes it impossible to use the material.
  • The resources are difficult to find. There are OER search engines and many repositories, but it remains that every teacher who wishes to adopt OER for their course must invest lots of time to find resources. This could be improved, with ready-made course packs and prepared/curated lists, but it will require lots of work, and constant updates.
  • Switching costs. The OER model is new so teachers and administrators are not used to it.
    • Lack of “vendor” to buy from, and ask for support.
    • How does CC licensing work?
    • Will there be updates and corrections?
  • There is no common format to allow mixing OER from different sources.


Next steps

I’ve come to the realization that getting quality OER into the hands of kids around the world is a much bigger task that I previously thought. It’s not like people who worked in the field before were stupid and didn’t use the right approach. It’s just really hard! The obstacles are not only technological, but also social, and psychological. Is multi-author collaboration on textbooks even possible? Every author has their own voice and approach, so it’s not guaranteed their “voices” can merge without outstanding amount of coordination.

When faced with big challenges, the best strategy is to “bite off” small pieces. For the coming weeks, I’ll focus on the “typo fixes” workflow, and write a prototype for anonymous readers to contribute fixes to an existing text. Let’s see if it is possible to hide the complexity of git, while at the same time keep the power of github pull requests. Easy does it. One step at a time.







Linear algebra final push

Last week was a watershed moment for the linear algebra book. My editor finished her final pass of edits to the text and passed the ball back to me for the final touches. All the remaining tasks have been placed in a Trello board and now all I have to do is act on them to finally get the job finished.

It’s difficult to describe in words what a relief finishing this book will be for me. This book has been my life for the past four years. Here is a time lapse video that shows two-years-worth of commits in two minutes.

I want to thank all the readers who have supported me throughout the years. Thanks for all the feedback and suggestions, which made the book better. A big thank you to all the gumroad readers for your financial support and your patience—I’ve been “finishing up” the applications chapters for three years, yet not a single reader has threatened to come and break my legs because of the delays. Y’all are awesome!



Request for feedback

In parallel with the edits of the linear algebra book, I’ve also been working on a new look for the book covers.

Which version do you prefer and why? You can reply via the comments below or send me an email.

Improving the math chapter

The goal for the NO BULLSHIT guide to MATH & PHYSICS was to make a concise textbook that teaches university-level calculus and mechanics in a nice “combined package.” The math fundamentals chapter grew out of the need to introduce the prerequisite material that many students often lack. I didn’t want to be like “y’all should remember this math from high school,” because if you don’t remember the material such comments would not be very helpful. A review of high school math would be more helpful.

Over time, I kept adding and improving the introductory math material in Chapter 1 until it reached the point that it’s a pretty solid little intro to high school math. I was very proud of the fast paced flow of explanations which manages to cover a lot of material (70% of high school math topics) in less than one hundred pages. Many readers also praised this chapter, saying how useful they found it as a review of high school math topics.

Recently I’ve been hearing from several readers who say the intro chapter sucks, and the book sucks, and by extension I suck. If it was one or two reviewers I could have dismissed this feedback, but now I realize there is a clear and consistent message in the readers’ feedback: Chapter 1 sucks as a first contact with math. My effort to “cover” all the high school topics in a fast-paced narrative like in the free mechanics and linear algebra tutorials is probably the worst thing to do for absolute beginners. I can totally understand why a reader who is not familiar at all with sets, algebra, and functions will have a rough time in the opening pages of the book. In the words of a reader, the book “goes from 0 to 60 in the blink of an eye,” which might be a good thing for a sports car, but not for a math book. It doesn’t help that I say “anyone can learn math from this book, regardless of their mathematical background” in the marketing copy. I need to do something to fix Chapter 1, and soon.

So what am I going to do about it, then? Write, of course—what else can a writer do? I’m going to prioritize the basicmath project and write the best sequence of introductory math lessons that ever existed! I’ll then use these explanations to beef up  Chapter 1 to make it a solid foundation. I think adding 20–40 more pages will be enough, so the book won’t get that much thicker. It’s not just about adding though, I think Chapter 1 could use better organization, flow, and clarity of explanations.

Interestingly, the basicmath project overlaps well with my planned social media campaigns that will push the message “learn math; math is useful,” as well as the math lessons by email. February is gonna be very mathematical!

Math for lawyers

A reader recently suggested I should write more articles that introduce math specifically for different audiences. Math for doctors (although I’d hope they already know math!), math for artists, math for musicians, etc. I like this idea.

In the spirit of procrastination and given the three other tasks I have to do in currently open tabs, I’m going to now dedicate the next hour to writing a sample post that introduces math to people with a “legal mind.” Don’t worry, dear readers—it just takes an hour to write a blog post. Reading it will only take three minutes.


Yes. I’ve said the dirty word. Everyone’s dreaded topic. So factual and unforgiving—you’re either good at math or you’re not. Okay. No, no, no. We’re not going with the usual narrative today. Let’s deconstruct this thing that is math, and see what it’s made of. Perhaps it’s not so bad.


Axioms and rules

Math is very ordered. It’s a bit like the French system (civil-law). All of math can be summarized as basic axioms on which everything else is built. Think of math as a set of rules that people have found to be generally useful in the past two thousand years. Like, send a spacecraft-to-mars useful. Once you know the dozen or so rules for working with numbers and expressions, and the dozen definitions and observations about geometry, you’ll have access to some of the “best stuff” that human intellect has to offer. I guess what I’m trying to say is that A) math is not that hard to learn because there is a finite set of basic rules to learn, and B) once a reader learns the basics, the reader is granted a nonexclusive, royalty-free, perpetual, irrevocable, transferable, worldwide license, to make, use, enjoy, benefit from, teach, share, offer for sale, sell, reproduce, include in, distribute, modify, adapt, prepare derivative works of, display, perform, and otherwise exploit math.


Math has a lot of common-law aspects to it too. Theorems (big results) and lemmas (little results) are like cases that mathematicians have proved. A proof is a bit like a trial, where the mathematician tries to convince the jury (usually consisting of fellow mathematicians) that some new mathematical fact is true. Unlike a real-world courthouse that depends on a judge’s judgment at some point \(t\) in time, a mathematical proof is always on trial. If the mathematician’s proof is solid and can be followed to the basic axioms, it’s very unlikely it will ever turn out to be wrong, so it’s common for mathematicians to simply cite theorems as if they were “ruled upon” case law.


The pitch

So here we are talking about math as if it’s some cool new thing, but we all know that math is difficult to learn and probably not that useful in every day life. Yes, perhaps it is so, but the point remains that certain math—let’s call it the useful part—has been around for thousands of years and hasn’t been proven wrong. Basic math is well understood, highly useful, and totally empowering.

Do you want to be part of this math thing? If so, check out the books.


Linear algebra problem sets progress

I’ve been working on the problem sets for the linear algebra book non-stop for the past month. It’s a lot of work, but also very rewarding. I’m going through online resources and looking for inspiration by reading exams and books to find illustrative exercises and challenging problems. This leads to a lot of  learning and reviewing of ideas along the way…

Finalizing 50 pages of problems and solutions is not an easy task, luckily I have good coffee to temporarily boost me, and more importantly access to the SymPy live shell. In combination with TeXShop, writing up more than one solution per hour becomes possible.  Check this screenshot if you want to see the author at work . Using as a URL shortener, you can put an entire problem solution as a URL:

So far I am yet to find a useful linear algebra problem that I haven’t covered in the book. I’m pretty happy about this because when writing a textbook from scratch you never know what topics you might miss. I’m only using the formulas from the book, and able to handle most problem types… Linear algebra? we got you covered!

Call for student assistance: are you taking a linear algebra class this term? Do you want to solve all the exercises and problems in the book as “extra practice” for your final exam? Get in touch with me ASAP (my_first_name at minireference dot com), and I’ll send you the PDF of the entire problem sets. It might take you up to a week to go through all of them, but by the end of it you’ll be able to handle even the toughest  linear algebra final.