Leave a comment

steer December 19 2015, 12:47:13 UTC
Learn To Code, It’s Harder Than You Think

So much wrong in one article:

Given the skills shortage one would expect graduates from computer science courses to have very high employment rates.

Computer science is very much not about learning to code. I work in one of the leading CS depts in the UK (technically refer to ourselves as a dept of computing but same thing). We teach barely more coding courses than a typical mathematics or physics dept.

The Higher Education Statistics Agency found that computer science graduates have “the unwelcome honour of the lowest employment rate of all graduates.” Why is this

Why is this? A massive cherry picking of articles. The author must have searched long and hard to find one stating that computer science was the least employable subject in the UK. It's consistently about mid way up.

There seems to be a ‘double hump’ in the outcome of any programming course between those who can code and those who can’t.

I've taught and marked coding courses since 1999 and have never seen this double hump. There are people who get it and got it when they started. There are people who don't get it. There are people who muddle through -- like any course. I never saw any evidence in my marking for this assertion.

between 30% and 60% of every university computer science department's intake fail the first programming course

Would love to know where this stat comes from. If a course has this high a failure rate you urgently need to evaluate the assessment criteria. No decent UK university would say that a course with a 60% failure rate was correctly positioned in terms of difficulty.

Even programmers with CS degrees insist that they are largely self taught.

Almost as if CS degrees are not to teach programming... which they're not.

Reply

steer December 19 2015, 12:51:46 UTC
Moving to the anecdotal, I found that for the first five or six years I taught programming the numbers passing my module was exactly what I would expect in line with other modules on that degree (indicating I'd pitched the course correctly). Many people went from being unable to code to being able to code. Mostly people who failed did so by simple laziness rather than not being capable -- that is they gave in reports which contained working (or near working) code but which didn't contain write ups surrounding it, documentation etc.

The one exception to this was the year our department had a large influx of new students from a different university which I won't name. Those students took my course and all but one dropped it as "too hard" by week two. The remaining one failed the course. That university (I won't name) did not have the academic reputation the university I was teaching at and the students were not as good. My conclusion isn't that they were "too stupid" to program, my conclusion is that if I were teaching that class to those students I would change the learning aims and goals.

Reply

gonzo21 December 19 2015, 12:54:02 UTC
A friend of mine works at a senior level of computer game programming, and has endless problems with computer science graduates applying for jobs, being brought in, and they discover that very few of them actually know how to program. But they have these shiny degrees. From ~cough~ certain un-named Universities that are apparently just rubber stamping graduates.

Reply

steer December 19 2015, 12:57:33 UTC
ARGH! NO!

Seriously, a CS degree is not supposed to teach you programming. That is not its job. That is not what the course is designed for. They have these shiny degrees in Computer Science that is not a shiny degree in programming. You do not fail because you can't program. Indeed someone with great mathematical and algorithmic talent and a good understanding of process and logical thought who could not program would do far better than a brilliant programmer with no other skills.

OK, if the person has a degree in Games Programming then you'd probably expect that they could program.

So what you're seeing is not that Universities are rubber stamping graduates it's that your friend has fundamentally misunderstood what Computer Science is.

Reply

gonzo21 December 19 2015, 13:02:51 UTC
Actually no that's my ignorance using shorthand, these would be graduates with relevant degrees, so probably as you say, programming degrees rather than just pure CS.

Reply

steer December 19 2015, 13:07:16 UTC
OK fair enoiugh -- there aren't actually very many programming degrees in the country.

Reply

andrewducker December 19 2015, 15:19:51 UTC
There really should be more.

I suspect that a lot of people go to university to do programming, and think that's what CS is going to teach them.

And you'll pick some up along the way, but it's not what they're there for.

Reply

steer December 20 2015, 18:32:25 UTC
There really should be more.

Universities are (I believe rightly) reluctant to teach vocational courses -- it's really not what they are for nor what they would be good at.

Reply

andrewducker December 20 2015, 22:26:01 UTC
Yeah, this is one of the issues with the UK educational/employment system, IMHO. Programming is something which should be taught vocationally. Probably along with a bunch of other subjects.

(There's room for Comp-Sci as well, obviously. But big companies look for degrees from their coders, when they should be looking for good scores from their training courses.)

Reply

steer December 22 2015, 17:59:35 UTC
I must admit I'm not sure what the tradeoff between someone who has the ability to code a particular language or someone who has the analytical toolset for the longer term. I guess it depends how the company is thinking in terms of time span.

Reply

xenophanean December 19 2015, 13:01:01 UTC
Interesting, I've been wondering about this one. As a one year computing student, it was clearly the case that some people learned a great deal faster than others, and as an interviewer I've seen a few people coming out of university with Computing degrees who basically can't program.

That said, I really don't like the article. I think anyone who says "I'm one of the Special People, who has powers which make only me capable of stuff" is probably bullshitting.

Maybe it's not so surprising? Perhaps the logic work needed in coding, is just another talent like artistic ability, just about everyone can paint a bit, less can paint quite well, less still well enough to do an artistic job, and only a very few (and then it'll be a big argument as to who they are) can make really great artists. As for the bad at their fields people? Well, you get people who are poor in every field, and devs are extremely in demand, so it's likely the job market will be sub-optimal.

Reply

steer December 19 2015, 13:06:29 UTC
perhaps the logic work needed in coding, is just another talent like artistic ability, just about everyone can paint a bit, less can paint quite well, less still well enough to do an artistic job

Perhaps you are right. One thing I did find is that some people who were quite talented at programming were dreadful at mathematics. Sometimes at examiners meetings I'd describe someone as a brilliant student and the people who taught theory to them were goggling because those people thought that they were terrible students.

"I'm one of the Special People, who has powers which make only me capable of stuff" is probably bullshitting.

I've certainly come across a lot of people who's belief in their abilities hugely exceeded their abilities and similarly people who were producing code that was as good but believing they were a real struggler. In a way, it's not surprising. For most people programming is continually iterating to work out the bugs until it's "right". If you have self belief that feels like you're brilliantly succeeding at a task that is hard. If you have no self-belief that feels like you're scraping through at a task you're too stupid for. I have a friend who works as a skilled developer, has done for about fifteen years and been paid well and still thinks she's "faking it" and can't program.

Reply

xenophanean December 19 2015, 13:30:02 UTC
"One thing I did find is that some people who were quite talented at programming were dreadful at mathematics."

Indeed so; my father started working as a coder at around 50 years old, after being repeatedly rebuffed for failing the psychometric tests on the grounds of poor maths skills, and not being the "right sort of person". He rose the ranks pretty quickly once he was in though. He's still doing it at 70.

"I've certainly come across a lot of people who's belief in their abilities hugely exceeded their abilities and similarly people who were producing code that was as good but believing they were a real struggler. In a way, it's not surprising. For most people programming is continually iterating to work out the bugs until it's "right". If you have self belief that feels like you're brilliantly succeeding at a task that is hard. If you have no self-belief that feels like you're scraping through at a task you're too stupid for. I have a friend who works as a skilled developer, has done for about fifteen years and been paid well and still thinks she's "faking it" and can't program."

This very much matches my experience. I've had a pretty solid 5 year career, and I still feel like an imposter sometimes. I've also seen quite a range of levels of confidence and ability, and they don't always match (Indeed, there sometimes seems to be an slight inverse proportionality).

Reply

steer December 20 2015, 18:28:42 UTC
I've had a pretty solid 5 year career, and I still feel like an imposter sometimes.

I've been coding for nearly 35 years now -- still learning. But a difference is I feel confident I can learn it. Sometimes it will be days of going around to understand a concept. Or I'll try to do something again and again for several days before giving up and taking another approach.

Reply

newandrewhickey December 19 2015, 20:33:53 UTC
That last paragraph may explain a LOT of impostor syndrome, including how I managed to remain employed as a software engineer by a very large, respected, computer company for five years while as far as I'm concerned "I can't program". It may well also be a small part of why women find getting into software so hard (a larger part is, I think, the cultural stuff). That's actually given me quite a bit to think about...

Reply

steer December 20 2015, 18:40:22 UTC
Well it is worth thinking about... Sometimes it can take days or weeks of struggles and dead ends to solve a particularly complex problem (my worst was a month to find a bug in someone else's logic -- the month was on and off). Some people think it's because the problem was incredibly hard, some people think it's because they're stupid.

Reply


Leave a comment

Up