Stack Exchange
log in

Ben I.

 The Classroom

General discussion for cseducators.stackexchange.com.
1
Ben I.
Feb 19 14:36
Oops!

Sometimes I'm the dumbest guy in the room, and sometimes I'm the only person in the room, and one of those is a strict superset of the other.
Ben I.
Feb 18 14:47
@verbose There is also thesecretmaster, but nowadays he mostly only comes if I direct something to his attention, like like if one of my own posts gets flagged, or if there's something complex that I need a second pair of eyes on.
Ben I.
Feb 3 22:31
That's sensible-ish. There are no truly sensible policies in the face of LLMs. I'm not sure that one could exist at this point.
Ben I.
Feb 3 22:03
@Buffy Well, I'll be. I don't know when they added that in. I like it! It makes acceptable behavior clear, and allows our own meta to override any parts of the document that we see fit.
Ben I.
Feb 3 15:33
I guess the right thing to do is to open a meta about it, which I don't have the time/energy to do right now. I'll get around to it eventually, but feel free to jump in and create it if you have the wherewithal to do so.
Ben I.
Feb 3 15:31
I'm not sure what to do with your AI flag, since AI is banned network-by-network, not across SE generally, and we've not had a meta about AI content.
Ben I.
Feb 3 15:31
@Buffy We got a great question this time!
Ben I.
Jan 22 16:35
If you're supposed to teach more at the OS level, I'd steer clear of anything but the most cursory overview of a processor. There's too much memory management and paging and resource management and threading and process scheduling and such to explore at that layer to really get into the processor in any depth.
Ben I.
Jan 22 16:33
After the learning goals, I'd figure out how I'd know that the kids have mastered what I want them to master, and then I'd figure out how to teach it. (Backwards planning is a basic good practice that streamlines courses towards focused exploration and manageable goals.)
Ben I.
Jan 22 16:31
(I haven't done any RISC programming myself, so I can't recommend which areas to excise)
Ben I.
Jan 22 16:30
If you want a modern architecture, RISC-V seems like a good choice to me, though I would make some deliberate decisions to simplify the curriculum, and make clear to the students which areas are allowed, encouraged for exploration, but not required for our studies, and which areas are core to what you need the students to get out of their exploration.
Ben I.
Jan 22 16:30
Real hardware labs tend to get very time intensive very quickly - you can get them programming, but bear in mind that you'll lose a lot of curricular time and likely won't have a lot of breadth to your course. If you want a survey of one of the layers (either OS or hardware), I'd encourage you to work with simplified versions of things. At the processor level, SAP-1 is very good. For assembly, like I said, I've had great luck with 6502.
Ben I.
Jan 22 16:25
For computer organization / architecture, is your target more hardware, or more like operating systems?
Ben I.
Jan 22 16:24
@Rusi No, we just have the 6502. We've found that it is sufficient for our learning goals.
Ben I.
Jan 11 02:20
I cover both uncountability and uncomputability, though I've never received a proof like that. (Though, in fairness, I only go over the halting problem, so I guess I go not-quite-far-enough to run into that issue.)
Ben I.
Jan 11 02:18
@Buffy As am I, my friend. Happy new year!
Ben I.
Jan 10 12:30
@Buffy The uncountability question?
Ben I.
Oct 10, 2024 13:11
@Buffy In regards to your recent flag, I notice that the question was closed on Academia. I'm not sure it's topical on CSE, and it doesn't seem to be topical on Academia. It's not a fundamentally bad question in an SE sense. Do you think it belongs in either network?
Ben I.
Oct 7, 2024 13:52
I am also (mostly) unafraid of it. It's not ready to take over the world, and at least in the current formations, I don't think that a scale large enough for it to get there is possible at the moment. This is a pretty neat video that talks about neural net architectures, error rates, and scale: youtube.com/watch?v=5eqRuVp65eY
Ben I.
Oct 7, 2024 13:49
@RyderisnotRude. I love me some GPT!
Ben I.
Oct 7, 2024 13:32
Hello, world.
Ben I.
Sep 4, 2024 19:39
We still get occasional great questions, but with no voters, they never make it to HNQ.
Ben I.
Aug 12, 2024 11:56
@Hark!Aquestion! This question should have been a home-run HNQ. As best I can tell, the answers came in a little late (because we're low traffic), and there wasn't a lot of voting.
Ben I.
Aug 7, 2024 00:28
We are mostly fed by SO HNQ traffic, so we're just ancillary damage.
Ben I.
Aug 7, 2024 00:28
@Buffy Yeah, I think people are going to it over stackoverflow. The entire SE network has seen a massive drop in traffic.
Ben I.
Aug 6, 2024 19:35
Maybe jump-start the system
Ben I.
Aug 6, 2024 19:35
I wonder if we could arrange for a HNQ one of these days again.
Ben I.
Aug 6, 2024 19:35
I'm blaming chatGPT and other LLMs for the slowdown in posts, but who knows?
Ben I.
Aug 6, 2024 19:34
@Buffy I think that, with the incredible slowdown in posts, I'll never surpass you. I suspect that you've won.
Ben I.
Jun 18, 2024 19:52
@Buffy This latest question seems right up your alley: cseducators.stackexchange.com/questions/8076/…
Ben I.
Jun 14, 2024 02:06
I would not be just a nuffin'
My head all full of stuffin'
My heart all full of pain.
I would dance and be merry,
life would be a ding-a-derry,
If I only had a brain.
Ben I.
May 19, 2024 02:27
So far, 9 days without a single post is probably our longest dry spell ever.
Ben I.
May 19, 2024 02:27
Thanks for that post, @user85795
Ben I.
May 15, 2024 14:35
::wink wink::
Ben I.
May 10, 2024 12:34
Everyone who's ever chatted here is pingable. You can just reply to one of their messages, and they will get the ping :)
Ben I.
May 10, 2024 12:33
@user70432 He's always pingable!
Ben I.
Apr 10, 2024 14:08
I've had this in my head all day. Anyone here know Sweeney Todd?

Most Honorable Judge Turpin,

I venture thus to write you this urgent note to warn you that the hot-blooded young sailor has abducted your ward Johanna from the institution where you so wisely confined her.

But hoping to earn your favor, I have persuaded the boy to lodge her here tonight at my tonsorial parlor in Fleet Street. If you want her again in your arms, hurry after the night falls. She will be waiting.

Your obedient humble servant,
Ben I.
Mar 8, 2024 14:50
@user85795 Oof!
 

 Discussion on answer by Ben I.: The R…

Imported from a comment discussion on cseducators.stackexchang...
Ben I.
Jan 22 14:43
@Rushi Hey Rushi, I'm happy to continue the conversation here if it's helpful.
Ben I.
Jan 22 14:43
@ScottRowe 6502 is awesome!
Ben I.
Jan 22 14:42
@Rushi I'm not sure what you're asking. We give them a series of assignments that we originally started based on skilldrick (skilldrick.github.io/easy6502), with an emphasis on tracing registers, memory, and the program stack. It's a von neumann architecture with only 3 registers and something like 50 instructions, no floating point, and no built in division. It's great for our needs! And since it was an industry-standard for a while, it's just messy enough to be interesting, even to really advanced students.
Ben I.
Jan 22 14:42
@Rushi Thanks for the clarification - I definitely misunderstood your statement. FWIW, I still maintain that RISC is a nearly absolutely better choice than CISC for school purposes. In our program, our department extended that principle even further, and we teach 6502. It's great for its small number of registers and its total simplicity.
Ben I.
Jan 22 14:42
@JörgWMittag While there may be no effective difference in a computational theory sense, the word transpile typically means high level language to high level language, and compile typically means high level language to low level. Without that distinction, transpile wouldn't need to be a word at all.
Ben I.
Jan 22 14:42
@JörgWMittag I edited my answer to remove my correction.
Ben I.
Jan 22 14:42
@JörgWMittag That was interesting - thank you for correcting my correction! I guess transpiling like that is a cheap way to get the advantages of compilation without having to write a compiler.
 

 Discussion on answer by Jörg W Mittag…

Imported from a comment discussion on cseducators.stackexchang...
Ben I.
Mar 14, 2024 12:18
They must have realized that they were doing something at least somewhat novel, or they wouldn't have felt compelled to bring in a word like that.
Ben I.
Mar 14, 2024 12:16
@Steve I don't think most people would call business administration "programming", but of course we can always trace any idea back to earlier roots in similar ideas.

The word was brought into algorithmic/recipe thinking as a direct result of (and because of) computers - that's just the etymology of the word. The etymological link, I think, is the use of the word "programming" in the arts, where people assemble programs across a season, and in television/radio, where episodes and shows are placed into a lineup. I believe that that idea is what drove them to choose the word "programming" f…
Ben I.
Mar 11, 2024 20:23
@BenVoigt Even if you somehow broaden the definition to now include non-computer "programs" (which would certainly not be common usage), you would still be left with "computers are why we have programming". We coined the word usage when we needed to distinguish programming from algorithms - prior to electronic computers, no such distinction was needed. The distinction of the concept was borne of electronic computers.
Ben I.
Mar 11, 2024 20:23
@BenVoigt Algorithms certainly existed, but what of it? Programming is the construction of algorithms for computers. You can see in the etymological root (a written notice) why they would have chosen the word to carry the idea of encoding algorithms. The word programming was only applied to computation in 1945, in reference to creating programs for automatic computers, which had already existed for a few years. Our modern usage was borne of, and was always linked to, electronic computers. (See etymonline.com/word/program)
Ben I.
Mar 11, 2024 20:23
@ScottRowe I'm absolutely certain that computers are why we have programming and not the other way around. Computing is why we have computers. Programming is the best way we've found to guide computers in their computing.