Comment by caseymarquis | Hacker News Reader

Comment by caseymarquis | Hacker News Reader

I think you could probably build an entire course out of following a button click into a browser, and dissecting all the layers along the way.

What happens when you hit the mouse button? Let's break open the mouse and find out. This leads to a pretty simple circuit, until you hit the MCU doing USB communications. What is an MCU? It's a small computer with a processor, memory and electrical inputs/outputs. USB is pretty complex, but you could emphasize that there's a power line, a return, a data in, and a data out line. I'd probably skip electrical encodings and the details of USB protocols at this point. The computer is the USB host (server), and the chip on the mouse is the client. You could then open up the PC, follow the path through the hardware to the OS. At this point you get to explain what a driver is. This leads to application code and OS APIs. This leads to the coordinate system in the OS and monitors/touchscreens, which finally gets you to the browser, html, css, javascript with ajax (browser inspection tools), http, https and certificates, tcp, ip, ethernet frames and wifi (wireshark), network cards and their management, looking at a network card, dissecting an ethernet cable (crimping one), taking apart an access point, switches, the gateway, the internet, ISPs, servers, databases...

I think the point would be to emphasize that there are no black boxes, give a good overview of how things work, and hopefully find a couple kids who really like some portion of all this and help them start some more advanced projects.

I think this is way too much content for a single course. But well, it's an interesting idea, it just has to be filtered so the students learn something in enough depth to apply it.

I don't really think they need to directly apply it? I've never had a direct application for biology, but I'm really glad I learned about it. It informs my thinking about things like funding research on new antibiotics.

This is more about demystifying IT than anything else. It takes years to be able to use all these technologies to build things, but if you cover one layer per class and you get 30 classes, you could cover a lot of topics.

I took a law class in high school (New England has pretty good public schools), and the whole class was basically: Let's talk about something interesting in law today. Having an IT version of that would probably go a long way toward educating the general public on how all their stuff works.

The bigger thing for me would be finding the kids who love learning about it, and getting them the opportunity to go further.

Selecting for individual kids and dumping the rest is what clubs are for outside of school though, not actual classes. Anybody can make a course that only a few will benefit from, the trick is designing a course that has proven results for all students taking it otherwise why even waste their time.

"One size fits all" is not a realistic goal.

Ever heard of "No Child Left Behind"? The government provided a set of minimum requirements everyone should meet, yet provided no metric gathering beyond the minimum, no funding for advanced studies. Result was a huge decrease in funding for gifted students and there's been several studies on the negative effects.

In my opinion and personal experience, it was awful for any gifted children and all the gifted programs were drastically cut.

One size fits all classes are not the answer. Kids are too varied in aptitude.

I think the course described above might be a better fit for an intro survey course at the college level. Or as an upper level elective course for high schoolers.

You can select for people for are actually interested, and you’re really just introducing the students to concepts that are worth there own class to get to the level where you can actually apply them.

Reply on news.ycombinator.com