Back

Notice that "integrating functions" is an outcome, a specific skill. A nice one that implies you know a lot of math... maybe. But not all math curriculums end there. There is a robust debate in math education about if we obsesses about The Calculus too much, everyone understands that doing some integrals isn't "knowing math."

I think most CS educators understand something similar but there is much less consensus about what it is that we are teaching if not "how to code."

For me the major topics of a CS education for the general public are:

* Computer Hardware
* Encoding and Decoding
* Logic and Control Structures
* Iterration
* Objects & Functions
* Databases
* Ethics and Applications
* User Interfaces and Design
* Computer Networks
* Computer History

This list keeps changing every time I revise my courses which is every year.

I added functions with objects, but students learn about functions almost from day one.

**File Structures need a clear home.

@futurebird Computer History is straight-up relevant to my computer-touching job regularly, in part because we've barely advanced computers in 50 or so years, and in part because knowing the history of a system helps you make educated guesses about it's present.

@lampsofgold @futurebird I've been sharing a lot more computer history with people lately, partly because chip design is being so creative right now (well, and that my dad was big in the metrology business), and partly because I am so inspired by this recent paper: https://docs.google.com/document/d/1tgZ86ugijXPUOo_lD3kLfQBPbneecUNzZim1GOgkC1M/

@slowenough @futurebird Seeing the thesis I definitely agree with much of it, though I find the crypto sponsorship and subsequent filter unfortunate. This reminds me of the community media manual (and seems like evidence of that work's thesis also) https://communitymedia.network/

@lampsofgold @futurebird For me the main point was to make visible the patterns, and the structural changes IETF made starting in the 1990s, that have helped them to make great specs while drawing on broad input, and avoiding anyone or any one group having control over the organization.

I was one of the developers of the Group Works deck, one source of the patterns they highlighted. This is the best analysis of *anything* I've seen using it. https://groupworksdeck.org/ #patterns #ietf #governance

@david_chisnall @futurebird For people who *are* going to be practitioners, the history bit is also the most valuable part. Pity we fail so hard at teaching it or paying it any attention!

One day people will say, "Oh! This system could work so much better if we break it up into this huge number of services that all talk to each other," all because they won't know any history. :-)

@hydropsyche @futurebird The best science teaching I had focussed entirely on the scientific method. Here’s a theory, what experiments would you design to falsify it? Start with some of the exciting nonsense the ancient Greeks believed, the gradually move forward. By the time you get to Lamarckian Evolution, you can immediately think of experiments. Most importantly, when someone presents a new claim, your first reaction is ‘this is how I would test it’, and if they haven’t tried that test you are skeptical. I don’t really care if students understand Darwinian va Lamarckian evolution, I want them to think, if they’re ever in a position to need to know which is less wrong, how they would test it. And then hopefully look up the experiments that were already done and see the results.

@david_chisnall @hydropsyche @futurebird It's funny because this is exactly the same skill that someone needs to have to be any good at efficiently debugging a complex system. (1) What do you know, *for* *certain*, about the system? (2) Formulate a theory as to why the system is behaving the way it is behaving. (3) Design a tractable experiment to falsify your theory, etc.

@hydropsyche @futurebird Indeed, this is something that has to come from a curriculum design with specific learning objectives in mind, it's not something that an individual teacher can fix.

@Jennifer @futurebird

Yo, COBOL programmers are bad ass. And in very high demand.

@futurebird Some of this examples were:

1. The Benham disk illusion (this is to prove you are fallible).
2. You have a creature that moves up and down a line. Can you create an algorithm that replicates this movement exactly?
3. Why the 80 column limit exists?
4. The history of COBOL, and what actually is structured programming and its real ethos?
5. The history from Babbage to Ada up until Turing on Turing machines.

I think these are interesting and give you a solid foundation onward.

@glitchypixel @futurebird Also non-computability (halting problem), NP hard, etc. There are things computers are bad at, so don’t trust claims that involve efficiently solving such problems.