The educational aspect of the mod reminds me of visiting one of those 'pioneer living' re-enactment places. The game is setup to force you to bootstrap up from being a cave (person) through farming and livestock and basic metal working, to actually achieving something that resembles medieval tech. The ores are actually named what geologists call them, you have to actually prepare the leather somewhat like in RL. Things aren't exact, but for a game that's just a mod it's pretty close.
Oh, and never forget the tedium of applied statistics that is Forestry bees/trees (breeding and Punnett squares). :(
As you point out, producing a new /type/ of educational product is very difficult. There need to be experiments in that space and some of the ideas will work better than others.
Contrast this with that recently linked story about how electricity did in factories. At first it wasn't that disruptive. It was changing the model of how a factory worked that allowed the true potential of electrification to shine. Today, at /best/, the most effective use of computers is as office software and assignment submission tools. In that use model a student might be better served by a cheaper or more powerful fixed station in their room and a portable 'notepad' (or large cell phone) for field use.
Agreed, with the proviso that "office software" includes some sort of computational environment. It bugs me that "school math" is so radically different than how adults -- possibly including mathematicians -- do math in their work and lives. Don't get me wrong, I breezed through school math, but I rarely use it now.
I open up a Jupyter notebook. It's not doing math for me. It's doing computations that are guided by my math knowledge, which is often coupled with empirical data.
Many people use Excel, including those who claim to be bad at (school) math, yet they are capable of doing useful computational work, even if it is relatively simplistic.
I'd like to see students learn math in more of an exploratory fashion, aided by real world software. For instance I think that probability and statistics would be a lot more intuitive if it were approached through modeling via random numbers. Problems that are hard to grasp via equations might be easy to grasp by modeling to predict an outcome.
I don't know if any youngsters have learned probability from there first. Would be interesting to hear.
1. Most of the software was glorified flash cards, a complete waste.
2. The medium strongly influenced the message. The sheer effort required to create software that supports a varied and interesting curriculum was (and probably still is) staggering. As a result, the software was typically one-dimensional, e.g., a single lesson structure varied by changing the parameters.
I've seen my kids use "educational" software today. In my view it suffers from the same problems, though it's graphically more sophisticated. A history lesson consists of looking at a screen, clicking on items, and reading the text that pops up for each item. A math lesson consists of choosing steps in a derivation from a menu of options.
Most of what they do using "educational" software is busywork, or drill work for the standardized tests.
On the other hand, I've also seen them use what I call "real" software, which is just general purpose stuff intended for a wider audience. For instance they write papers and engage in genuine collaboration using Google Docs. There's some really killer real software out there, that kids could be turned loose on, such as Jupyter/Python, Arduino, Scratch, any of the Office apps, Blender, etc. Also, kids need to be taught how to do real research via the WWW, especially the challenge of separating fact from bullshit.