The Classroom

3:32 AM

@Rusi But when I say students should learn a little about how the computer works, no one will bite. Why does more money make a faster computer? "It just does okay?" Oh well, you wake up and I'll go to sleep :-)

Oh, I just saw more of your message. I don't know C++, unless C# qualifies. But I think language is really overstressed. Data is what matters.

@VictorEijkhout more efficiency comments cseducators.stackexchange.com/questions/7266/…

Should continue here (actually meta is best since can write code)

Because ur straining at windmills @ScottRowe. Nobody (reasonable) denies "students should learn a little of how computers work"

The arguments/differences (between us at least) are more logistic than fundamental : (1) little means how much (2) when ie sequence (3) emphasis etc

3:50 AM

@ScottRowe You complain that we overstate your point, but you also overstate what others object to. No one objects to students learning how computer work. What people object to is the claim that the particular level of abstraction that connected the dots for you must be learned by everyone at the starting line.

Everyone should learn roughly how a processor works at some point if they want to be in this field, and I don't think that there is anyone who would contend otherwise.

I think there's a fundamental quarrel also @ScottRowe : Magic. Do you want to teach magic or erase it??? If you ask yourself honestly you'll find you're deeply inconsistent here

4:13 AM

@ScottRowe "Seeing moves" Yeah that's neat

Should be conjoined with a key statement from Genesis :

And God saw the world. And it was good.

We all see the world. Can we see it as good? Personally ... I find it bloody hard...

Actually "seeing moves" could be taken as the hallmark of functional P: Once the model of the problem is built, the computation falls out effortlessly.

And building a model is less an act of doing; more an act of seeing

Victor Eijkhout

4:43 AM

@Rusi Your efficiency comments are so abstract that I have no idea what to reply to them. Please address my point: print out C25 and tell me how long it took. And tell me why that formula is computed through something with splitting lists. I don't see that in the math.

5:11 AM

@VictorEijkhout ok I think we need to back up some on this

5:25 AM

My basic case is this (kinda) stuff can and should be taught to beginners.

And if that is incomprehensible to the un-initiated I, so to say, break my own case!!

However...

Catalan numbers are a (slightly) more advanced area of combinatorics. And I was showing off (as in showcasing) the succinct lucid power that can be packed into a single line.

But it would be unreasonable to assume that line would be clear without some background explanation

(which I've not done. I did it more breadth-first: ie trailers of combinatorics, language implementation etc)

7 hours later…

12:28 PM

@BenI. Right, I suppose I should stop thinking it is this big issue. It's just that it was probably part of the first few hours of instruction I had when I was 13, and when I was confused later, I could just go back to the simple clear "this is it" picture. Hard to imagine not putting that in at the start. So I have to find out how lots of others learned and where it got them. Too bad I am not an Education Researcher!

@Rusi Ok. Not aware of inconsistency: I have said many times, "Kill the magic. Kill it right away." It will still all be just as amazing, I promise!

@Rusi Well, the only statement that I accept "without sufficient evidence" is: The Collective Enterprise is Worthwhile. (Not to be confused with Camus' famous question.)

3 hours later…

3:37 PM

@Buffy I want to invite a researcher (either graduate student or full professor) who studies how coding works in the brain to my CSTA chapter. Would it be normal to offer some sort of honorarium? We don't have a lot of money.

3:50 PM

@BenI. It would depend a bit on who and where they are. If it is someone local to you, meaning a minimum of disruption to their normal work then maybe not. For someone farther away, where you need to also cover travel, it might be necessary since their time commitment is bigger. I think a lot of people would be happy to do it for you, given the nature of your school.

This would be on zoom, for the state CSTA association

I haven't run the idea of an honorarium past our board yet, so I'm reticent to even mention one to a researcher

Though in principle, I don't imagine that we'd get a lot of pushback from the board about $50 or $100

Probably minimal disruption. I doubt such an honorarium would be needed. I wonder if Owen Astrachan would be a possibility.

Erm, in practice.

Or do you have someone in mind already?

csc.ncsu.edu/people/cjparnin or shashank-srikant.github.io were my top choices, because they both focus on cognitive science and how the brain deals with code

4:00 PM

I don't know either of them. And I think they are unlikely to know of me IRL.

Part of the appeal of the topioc is that it applies to k-8 as much as 9-12 and undergrad

k-8 is the big, important audience right now in [my state], because mandates are kicking in this coming september, and every school needs to provide CS to every student.

Some of that K-8 stuff scares me. I would much rather see programming taught as an adjunct to other subjects (science, say) than as something job oriented. I think the goal should not be professional programmers, but people who can program in support of other things.

A lot of 9-12 too, actually.

If everyone on the planet is a "programmer" then we are all doomed. The world is too complex for that.

Well, when you get a definite answer, please post it as an Answer to my recent, ill-fated Question. I think it's time for me to quit the post of "Teach About the Machine Up-front" Officer. Cheers!

@ScottRowe I'm not even necessarily against mentioning it close to the start, but I think whether or not it's helpful depends on your broader starting strategy.

If my goals and philosophies have led me to start with C or Pascal, it's actually a pretty good thing to mention near the beginning.

If I'm starting with Scratch with younger kids, it's more of a factoid that will probably not impact things much one way or another

If I'm starting with functional programming, it will probably actively interfere with the way they need to develop their thinking at the outset.

I guess I'd suggest that when you want to mention the processor level of abstraction would be more of a tactical question and less of a strategic one. Unless you have decided to start your exploration at the processor level, in which case, it's part of your strategy. But also, that's nuts :)

6 hours later…