What will "classically trained" look like for computer science and digital literacy?
This might be a weird framing but it's been bugging me for a few days. Many fields have a concept of classical training -- this is most common in music but applies in the humanities and many other areas. For example I do a lot of CAD work for my job, but I received what I would consider a "classical education" in design...I learned to draft by hand and physically model before I was ever allowed to work digitally. I got a lot of value out of this approach and it still informs the way I work today.
A lot of people view computers and technology as modern and almost anti-classical, but as the tech industry matures and the internet moves from something shiny and new to something foundational to our society, what will the new classicism look like?
Thanks for reading my question.
Even today I would argue you could have "classical" training in computers where you start learning at the assembly level and work your way up to Python instead of the other way around.
It's an interesting thought experiment, but it assumes that technological progress will 'level off' and we'll end up with some modern-ish tech level for the next 100 years or so (long enough for 'classical training' to emerge). I don't think it's likely that this will happen, because it seems that technological progress is exponential.
If it does happen, the 'classical' education will be heavily influenced by whatever force stops the exponential progress of technology. This would probably be some sort of AI-winter like event, but even more significant, as it would involve the stagnation of AI, along with hardware, software, and other subfields. Since software doesn't rot, I could imagine a word where technology leveled off requiring much fewer tech workers, with the few remaining simply maintaining existing software solutions. This would be a very different field, probably more like history or archaelogy.
Don't worry, that too will come to an end. Eventually the free ride will end and we'll have to spend higher and higher percentages of our energy stores sourcing new energy stores.
When that comes technological progress will be hampered somewhat significantly.
Maybe, although I suspect renewables will just continue to get cheaper, and there's still a lot of unexploited energy sources (until we build a dyson sphere at least). If things get bad enough we as a society will probably just have to get over the stigma we have towards nuclear energy.
And if we really start to run out of energy, I predict that a lot more investment will be put into alternate energy sources, i.e. fusion (which means the future of energy is ultimately unpredictable, because there's no way to know how productive this research will be). I don't see energy being the bottleneck anytime soon though.
I don't think it's settled what the heart of computer science is. For example, you can take a mathematical approach where it's all about computing values and I/O is hardly discussed at all. That approach is exemplified by something like The Structure and Interpretation of Computer Programs where you program in Scheme, but alternatives might be Haskell or maybe even a proof language like Coq or Idris.
Programming interviews tend to assume that you need to understand algorithms, which would be a different sort of mathematical approach.
Alternatively, you could say it's about making gadgets, and start out with Arduino or something. Then, programming plays a small but essential role and electrical and mechanical design are also important.
Or you could assume the hardware is given (or maybe you start with a virtual machine) and learn assembly language and writing an OS.
You might compare with the different approaches to learning music. Do you start out learning to read sheet music or delay that and work on learning to play by ear and memorization? Depends on what your goals are.
Lambda calculus.
I was quite tempted to just leave it at that two word answer, but the more I think about it the more I wonder if "classical" and "fundamental" can be used interchangeably here? My understanding is that they're very closely related but not quite synonymous if you're talking formally about the worlds of music or cookery, but then if we're expanding the meaning of the term to other fields, especially more modern ones like computing, it seems like pretty much by definition there isn't that weight of tradition to separate the two ideas.
I think drafting and modelling by hand is conceptually equivalent to learning computer science (using a strict definition of the term) from the mathematical principles on up. You're unlikely to need to think about Turing completeness or context free grammars on a daily basis, but that foundational understanding can significantly change how you understand the things you do work with regularly, and why they are the way they are.
@a_s_k were you asking about computers or actual computer science? Computer science, it is said, has as much to do with computers as ham does with hamsters.
I think the term classically trained would be replaced at that point by humanly trained.