On Sat, 2018-05-05 at 11:57 +0000, KingJoffrey via Digitalmars-d wrote: > On Saturday, 5 May 2018 at 11:25:45 UTC, Russel Winder wrote: > > […] > > What is the pedegogy here, what are the justifications. > > That the languages being taught to undergrads, must be pervasive.
Not needed. The language must be good and fit for purpose not necessarily pervasive. For example Clean is an excellent language for teaching functional programming. That most professional programmers have ever heard of it doesn't undermine it's suitability. Conversely, Spark Ada is pervasive but would you use it to teach programming? However, I think it is right that exiting undergraduates have experience of Python, Java, and other languages used in the world of work, but they do not necessarily have to be the teaching languages. > That the languages being taught to undergrads, must teach them > about low-level types, and higher-level types. Definitely. Abstraction is at the core of programming. But different programming language have different type systems. So Ceylon, Haskell, OCaml, and Lisp are language for learning this sort of stuff. > And a whole lot of other stuff..(e.g, open source, cross > platform, have multiple compilers, standardised, formalised, > etc..) ..I can't be bothered typing any more ... Clearly the technology of programming in a professional way is important. But this is not specific to the programming language. And if we add IDE support as an integral part of professional programming, D fails to meet the criteria. > > In UK we have Scratch then Python then ??? This is working > > tremendously well to get large numbers of young people > > programming far better than most professional programmers. > > > > Don't teach undergrads how to play with toys! > > They can play with toys in junior high school, sure. Do not underestimate the quality and quantity of code written by 14 year olds with a Scratch and Python background. They put a large swathe of professional programmers to shame. And this with school teachers who are not at all sure they know what they are doing. CAS, Code Club, and other organisations have helped massively. More importantly though in a situation where new undergraduates already know Python, universities have to have a whole new approach. This is a revolution to traditional university CS education and training. Most universities in the UK are having to completely rework the whole curriculum ad way of working. For the better. […] > > C and Java are not extremes. Lisp, assembly language, Haskell, > > Erlang, these are extremes. > > C and Java, in comparison to each other, are at the end of each > extreme. In that C is a portable assembly language and Java is an object-based language supporting some forms of polymorphism they are very different, but they are far from extremes. > Those other languages you mention are mostly irrelevant (at least > on a grand scale), and certainly non-pervasisve. They have no > place in undergrad course. Erlang still runs much of the international telephone system, and a lot of share trading systems. Haskell is used for a lot of quant work in finance houses. Common Lisp and Clojure between them run quite a lot of Web systems and a lot of front-end share trading systems. They are very relevant and any undergraduate who has used them at all is probably sub-standard. > > > > D could be a postgrad interest perhaps. > > > > No, this would be a bad idea.We can debate this elsewhere. > > That should depend on the interests of the postgrad. > > At some point, we really, really should let them choose to focus > on what interests them. That is what options and projects are for. It all works very well in the UK. […] > you forgot 'mobile'. I guess a lot of people are interested in mobile, which these days means Swift and Kotlin. […] > > Python should be banned! Cut of it's head!! Hopefully this is a troll, and you don't mean that. Python is used by a huge swathe of data science and especially the quants who more or less run the finance system of the world. […] > > As someone intimately involved with a university for many, many > years, I have to wonder whether teachers are the problem, rather > than the solution ;-) I can't speak with direct evidence for the last decade but certainly till then far too many academics in the UK couldn't really program well at all. The whole university system militates against programming as something an academic is good at. Fortunately there seem to be just enough academics who can program well, at least in the top universities, that the programming courses do actually get taught well. Of course with programming moving from university to school (6 to 18) the problem shifts from academics to school teachers. -- Russel. ========================================== Dr Russel Winder t: +44 20 7585 2200 41 Buckmaster Road m: +44 7770 465 077 London SW11 1EN, UK w: www.russel.org.uk
signature.asc
Description: This is a digitally signed message part