Composable Education
Sunday, July 9th, 2023
#education #composable-learning
It's summertime and I didn't send my kids to Summer Camp as I normally would.
I have more flexibility during the summer, so I wanted to see if I could start to create some self-directed education with them.
This hasn't been easy.
But I think it is the future.
Normally, camp consists of one week options I can pick across a range of different vendors and disciplines. I haven't done that yet. I usually send them to the YMCA and, within their program, select for the kids their own interest areas.
But increasingly I want them to self select and, to start, teach themselves. Between ChatGPT, YouTube and the public library (Libraries, Homeschooling and the Future of Learning), I was hoping they would find their groove.
But I haven't yet, so I need to keep figuring it out. Here are my thoughts on why this is important.
The Future of Education will be Composable
When I mean composable, I mean literally a "choose your own adventure" where courses can potentially be stitched together across providers.
This is unbundling college, potentially, and the legacy providers are definitely going to stand in the way.
What does college provide as a bundle?
One is a consistent brand. The brand conveys an assumption that the quality of the courses are good. Not just the individual courses, but the bundled "program."
Meaning, if I graduate from Yale with an English degree or from Stanford with an engineering degree, that bundle means something because there is some rigor in the requirement.
The value, then, becomes either a) the selection of courses is important to prove understanding or to actually make use of the education; b) the "trust" we place in the brand that the selection and the educators are worthwhile.
So, for English, the bundling wasn't that rigorous. It was x-number of courses that qualify towards the degree, with some mandatory requirements around the classics and writing. Pretty basic.
For Engineering, it was a little more rigorous in terms of the sequence, areas of specialization (advanced mathematics, physics, electrical engineering cores, and then some set of specialization areas). And it makes sense that the degree or bundle conveys some proof of either quantity or base requirements.
But a couple of things start to fall apart in terms of the quality of education.
The brand is supposed to give us some assurances that the individual teachers meet those requirements. But it's not the case.
A great school like Stanford can have crappy teachers.
In fact, there might even be some disincentives.
Schools, even those with large endowments, still want to attract research grants, and those are based on quality researchers.
Students, however, mostly care more for quality teachers -- teachers who are engaging, interesting, and probably entertaining.
This doesn't always coincide.
But why not?
If we treat students as customers, shouldn't they want schools that are known for great teachers?
They should, but the review process is even more opaque than something like Amazon.
For starters, I don't think it's that common to have public verifiable reviews, and even then, there are some issues with quality reviews (meaning, someone who is a poor student and got a bad grade just slams the teacher; conversely, a teacher makes the course too entertaining and too easy to get an A to get high ratings.)
Second, the research process isn't that rigorous. Even if a student heard about a set of great teachers (and was qualified and self-aware enough to know what a good teacher was), there's no guarantee they would get the teacher.
So the selection process is poor enough between consumers in the B2C model that universities don't really have an incentive to ensure quality teaching. They get a bigger bang for their buck with better marketing (brochures, rankings, brand-building) than they do with better instructors.
On the research B2B side, the signal is clearer: which professors bring in more grant dollars. Those that do have high product market fit.
So that's a bit of a problem.
In a composable environment, students can pick the instructors, subjects and have more choice.
The challenge is whether there can be a reliable "standards body" which says whether something learned is "good" or not.
The brands of the legacy providers are transferrable when unbundled.
For example, if I studied computer science and took some classes from MIT, some from Harvard, and some from Stanford, I might be fine. Even though I am not getting the "bundled" value from the brand, the brand still convey some value to individual courses.
I think it can also go another direction, more atomic (and more accurate): down to the professor. For example, Andrew Ng, who is famous for his Machine Learning, could teach a course independent of which University, and it would carry far greater weight than many similar courses from top name schools.
Brand value may also start to shift out to corporations. Google, for example, if they offered a course on distributed computer or search algorithms would probably be highly competitive. Now, would they do so is another matter.
What, then, is the purpose of education? And here we need to think carefully.
Much of education is perceived to deliver value through its conveyance of skills (and ultimately the ability to get a job).
On the other hand, legacy schools have often focused on the value of "liberal arts" and, for a while, this has worked because many people hiring for prestigious jobs also had a "liberal arts" background and so valued the same.
But if this liberal arts background, for example, is a smaller number of graduates each year as non-liberal arts take on more jobs and hiring positions, perhaps this is not a winning battle anymore from the perspective of being hired.
In this use case, I see corporate-based courses potentially gaining quite a bit of ground. The challenge will be that corporate bands of a shorter half-life than university's. Stanford, Harvard, Princeton and Yale have all been considered high quality schools for year; and perhaps the brand is over-hyped against the intrinsic value.
However, IBM once had a durable brand, and I wouldn't consider that one someone's recent resume as a strong signal to hire in tech. So there's that challenge if the brand is based on corporate brand.
On the other hand, if they had a course that taught foundational skills in computers, I probably would value IBM. Perhaps I am swayed by their brand which did try to convey some steadfastness and trustworthiness.
It seems many schools have associated value because of their brand and trustworthiness. This is a good thing. There is value in institutional knowledge and values that are relatively insulated from market forces.
Question is what is the right level of insulation?
As noted earlier, winning grants and attracting tuition-paying students is a market force.
Professors on tenure, however, does provide insulation, but is it too much? I think some relative safety to research and write about the things that aren't swaying with popular opinion is important and valuable.
At the same time, having some public value matters.
I think finding some way to show the usefulness and value of scientific research that drives both innovation and curiosity while rewarding market developments (without making short-term profit-seeking the driver) is certainly worthwhile.
I think Stanford has done a great job of encouraging research that ends up being commercial. As things get more complex and competitive, I think profit-seeking will need more and more of this deep research and investment to find those breakthrough. Minimum Viable Research is probably not going to be a thing, ever.
So perhaps to address the composability, there needs to be a few things:
a) Trustable and verifiable reviews;
b) Clarity on the value (life skills, marketable skills, others)
c) Value to others
For my kids, the first part of this "composability" is atomic interest. Can they pick something (at this point, anything) that seizes their attention?
Can we stitch that together into something comprehensive against some kind of normative standard? Math, science, movement?
Right now, I am breaking things down into Reading, Research, and Writing.
But they definitely aren't working their critical thinking skills in terms of math, science and logic.