Kids should be making software, not just using it - Gove Education Secretary Michael Gove today proposed killing off Blighty's ICT curriculum in September to give it a thorough reboot. Launching a consultation into his plans, Gove suggested that from the start of the next academic year, schools should be able to teach what they want in computer classes. The Tory minister recommended …
Because by the time you've persuaded an actual language to produce interesting output, your 11-year old has fallen asleep. Learning doesn't have to be boring.
As it happens, I was looking for a gentle introduction to programming at the weekend and found Scratch. My 10 and 7 year olds have been playing with it since, by choice. I doubt that would have happened if I'd started them on a "proper" language, but scratch is powerful enough for them to learn from experience about spaghetti code and badly named variables.
In fact, I'm stunned that Mr Gove has stumbled upon such a good idea.
... we used pens and paper. There weren't special classes in "writing" after junior school, and you certainly couldn't get a qualification in "writing".
Once the basic skills were learned, they could be used to write in different languages, make mathematical notations, write up science experiments and so on. WP/spreadsheet/database programs are the equivalent of pen and paper - you use them to make something else possible. After the basics are learned at junior school, it becomes the role of the individual subject to teach the children how to use them for that purpose.
Amen. I'll never stop being bitter about the one A-Level IT Module the teacher gave me and a load of friends 100% in, that came back at the end of the two years from the examiners, all marked down to 60%, with no explanation from the teacher or exam board. It was a web design module, and we'd all gone above and beyond the scope of the module since we all had an interest and actually knew HTML etc. (rather than frontpage/publisher). That isn't to say we didn't do the boring bits to tick the modules boxes too.. we did, which was all the more frustrating.
I was in the first year of national gcse's the main reason we got great grades was that we had the benefit of at least 1 year of the relevant O level syllabus for some if not all of the subjects. Having done both mock O levels and real GCSE's I can assure you GCSE's were easier.
In practise you had it harder because your year had a syllabus designed to meet the requirements of the GCSE not potentially exceed it like ours.
Totally agree with you on the A* grade though. Don't any of these idiots understand about normalising the scores to a bell shaped curve as they did with my degree exams?
Most Reg commentards could quickly knock up a list of languages and tools they like or think are important, or a list of things they wish they'd learned at school, or a list of skills they think will be useful to the pupil or the economy. (witness most of the comments on this thread).
Putting together an actual curriculum that will keep kids from either falling asleep or running riot and will give them a set of skills that are provable in an exam or assessment is a whole different thing.
I wasn't really criticising the idea of taking two years to improve the curriculum, more the bizarre idea of throwing the current one away while you do it.
It's like throwing the passengers out of a leaky boat and asking them to tread water while you build a new one.
that the UK will be able to march towards world dominance in Programming[tm]
Especially games, and entertainment, and a bit of quanting, and stuff like that.
This would be a good thing, if it weren't for one or two minor points - like the fact that programming is useless without both industrial and/or management strategy and design, and that practical programming skills have a nasty habit of becoming obsolete fairly quickly. (Just ask a Flash developer.) Especially if they're toy programming skills, like Sketch - nice toy though it is.
So - programming is better than ICT, but a really interesting curriculum would combine development across multiple platforms with some element of useful, non-Hoxton entrepreneurship, perhaps larded with useful negotiation skills.
Gove's mind - if it can be called that - seems to be trained too mechanically to think strategically. Flash some action words at him, and he's putty in the hands of lobbyists and advisors.
Therefore drama and grandstanding and newsbites, but not so much deep insight into real educational needs.
Also, who's going to pay for the teachers to support this bold new effort - the ones who will be giving up highly-skilled contractor posts for the academic front-line?
(Not the Tories, that's for sure.)
Learning to code isn't generally useful, but learning how to control a cursed, wretched, bastard machine that does ONLY and EXACTLY what you tell it to do may well prove to be an essential skill for the general population in the 21st century. A computer is a reasonable platform to practice this skill.
The only major industries we have left in the UK are Service/Retail, Banking and High Technology. There are others but they are pretty small and unlikely to grow in the near future.
Service and Retail don't actually make any money for UK Plc as you can't export either of them, they only redistribute wealth. Very important to society, and a lot of jobs here but doesn't affect the UK's import/export balance sheet.
Banking creates money out of thin air, but much of it is "economically useless" as Adair Turner said, and doesn't employ very many people and is unlikely to grow in terms of jobs as it's already mostly automated anyway.
High Technology is extremely exportable and the UK really does make a lot from this - things like satellites, electronics designs (ARM is *huge*), and software. Maybe we don't manufacture much of it - but we do design it.
Unless we get a notable number of our children into that high technology sector, the high-tech jobs that companies need will go somewhere else and the UK will be properly screwed.
So yes, we do need to teach our children to code, because then we'll find the x% that are great at it much earlier and can inspire them to be the next ARM. or CGI house. Or something else that nobody has even thought of.
At present we're teaching those kids an "ICT" that is really "How to use MS Office". And that is boring - for a start, there is barely enough to fill two terms, let alone a full GCSE course. To make it worse, the pupils that find it most boring are the ones with an aptitude for real IT - so we're deliberately discouraging the very people we need.
I've said it before & I'll say it again, we need better teaching of subjects like English, Mathematics & Biology, Chemistry & Physics (the three sciences) much more than we need CompSci classes in schools.
Let those who wish to do computers explore it in 'addon' classes during a lunch break or after school or on thier own at primary/high school level. Keep it to the colleges & universities who are much better prepared to be able to teach it in a meaningful manner.
This post has been deleted by its author
"..taught me PASCAL about 15 years ago. I now have a CS degree and I still think it is the right language to teach people how nice and robust programming can be, as soon as the pupil has mastered the syntax details."
Keep in mind that PASCAL was also designed as a *teaching* language and was "dumbed" down compared to the commercial languages of the time. Possibly it's 2 biggest weaknesses are that most people know it from the Borland version, which lacked at least one *major* feature (the ability to effectively do pointers to functions, a key selling point of C) from the full version and a not very standardised library interface (a *major* contributor to the success of FORTRAN).
I agree that it was small enough to get to know well and teach key concepts, while being complete enough (provided your implementation had hooks into the OS) to do serious work.
""Scratch" is an attempt to dumb down programming and will be nearly as useless as Excel macros. "
Possibly. On the other hand the demands of creating a system that children *wanted* to use was a key driver of a lot of the PARC work, Smalltalk and the Alto workstation design.
Making kids stuff is *anything* but child's play (although at least one of them went on to code a flight simulator using LANDSAT pictures).
Scratch *might* get IDE developers thinking a bit more about how they can help developers slot disparate stuff together without endless thumbing through manuals.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
