Date : Mon, 03 Apr 2006 23:10:54 +0100
From : Jules Richardson <julesrichardsonuk@...>
Subject: Re: Basic & BBC Basic
Andy Armstrong wrote:
> On 2 Apr 2006, at 23:42, Jules Richardson wrote:
>> The OO side of it seems to be a total disaster though. Shame there
>> isn't much of an alternative when it comes to server-side web stuff
>> (Perl's just complete spaghetti, ASP isn't portable etc.)
>
> I'd say Perl's my favourite language - but I tend not to advocate it to
> anyone else.
:-)
Perl just makes my brain hurt. But then maybe that's what a lot of IT
graduates brought up on a diet of VB think about OO languages, scripts etc.?
> It's certainly not a suitable teaching language.
No, but I'd rather hope that IT students are made aware of the fact that it
exists, *is* widely used, and taught what its strengths and weaknesses are.
>> Oh sure, they all need to evolve. I suppose the trick is in getting
>> the basic framework right such that things *can* be changed without
>> fundamentally "breaking" the language. That's where I don't like C++
>> as it seems it was built on some fundamental bad choices which make it
>> a nasty language to use for a lot of situations.
>
> I really don't like C++ - it feels as if it's stolen the worst aspects
> of the languages that have influenced it. Objective C is nice but it's
> hard to do anything useful with it unless you're developing for Mac OS.
> Actually ObjC + Cocoa is lovely.
That's a problem - the 'best' languages tend to be the niche ones that just
aren't widely available on all platforms :-(
>> It was as though when faced with something unknown they went into
>> meltdown and couldn't see how to tackle the situation. A company's no
>> good if it can only cope with 60% of any problem and then makes a mess
>> of the remaining 40% (which seems to be what's wrong with a lot of
>> software)
>
> I guess across the industry it's probably more like 90/10 (cope/can't)
> and it may be that companies can get by with a few real programmers and
> a bunch of people who can bolt stuff together. Dunno.
Hmm, maybe so. I've always been involved with 'cutting edge' type work, so
maybe it's just been more obvious there how brainwashed a lot of grads are.
> There really is a lot for a would be developer to learn these days -
> probably way more than you can reasonably cram into a CS course.
That depends on how the course is taught I suppose - give a student a good
grounding in assembler, a procedural language, OO programming/design, a
functional language, and that probably gives them a good basis from which they
can learn anything.
ISTR being taught 68k assembler, C, C++, Eiffel, Miranda and Occam at uni -
plus most of the machines were Unix and so I had some exposure to Perl, shell
scripts etc. along the way. Given that, it's been a doddle to pick up other
languages since as the syntax may differ but the basic structure has always
been similar to one of the languages that I'd been exposed to.
The same goes for other aspects of IT - application designs, GUI design,
networking etc. - get the framework right and it's easy for people to adapt.
These days though I'm getting the impression that all the courses are about
specifics - not because that's what the industry needs, but because that's
what gets more wannabe IT 'professionals' through the course with good grades.
Specifics are much easier to spoon-feed to people with good results - it's
just killing the industry when these people get out into the real world.
> I'd
> like to see a stronger culture of life long learning in the industry.
> I'm not talking about going on courses to learn about the latest API or
> whatever - I mean proper development of their craft.
Yep, I agree. Mentoring works really well, particularly when it's informal.
Never been one for courses much myself as everyone naturally works at their
own speed and courses tend to force everyone to work at the same pace.
>>> One way of looking at it is that the world needed more programmers
>>> but teaching people to program to a decent standard was a bit hard -
>>> so we redefined 'programming' to make it easier.
>>
>> It's a nice idea - providing that efficiency, disk / memory footprint
>> etc. can be sacrificed. I suppose I just see "one size fits all"
>> solutions as being horribly wasteful!
>
> Same here. Of course there's may be an element of us just being grizzled
> old curmudgeons who remember when every byte was sacred and haven't
> quite adapted to just how ludicrously cheap once scarce resources now are.
:-)
But 'ludicrously cheap' tends to be a misleading thing given the resources
that go into making stuff or the cost in human labour to do so. It's only the
end user that tends to see a saving, whereas someone or some thing always
seems to suffer behind the scenes. I'd much rather see more effort going into
repairing and reusing what we have than into making ever-cheaper (for the
consumer) stuff. Not just computers, but all consumer products.
cheers
Jules