Interesting Links for 08-02-2015

[andrewducker]

• Samsung: watch what you say in front of our TVs, they're sending your words to third parties
  (tags: samsung tv surveillance )
• Is 50 Shades of Grey abuse?
  (tags: FiftyShadesOfGrey abuse bdsm )
• If you don’t genetically engineer your kids, you are a bad parent. It’s that simple.
  (tags: genetics parenting fiction scifi vaccination funny )
• Girls

Comments

Maths exams

[woodpijn]

I'm confused how this can happen, when maths is probably the most objectively-graded subject there is. I clicked through to the article to try and find out more about the methodology, but it said very little (and the actual study is paywalled).
I could understand it happening in a humanities subject: some kind of bias affecting a teacher's assessment of how well-structured an argument is. But maths is usually either right or wrong: if biased teachers are marking right answers wrong or vice versa, that's more than sexism, that's serious incompetence at the job and possibly even fraud.

Re: Maths exams

[andrewducker]

Math is only right or wrong at the very simplest levels. Beyond "2+2" I was always asked to show my working, so the teacher could see (when the answer was wrong) whether I was understanding concepts (and making mistakes) or misunderstanding entirely.

Teachers, therefore, have to make a judgement about whether the student understands, or not, and assign marks through that - and that's entirely open to bias.

Re: Maths exams

[woodpijn]

Maybe it's different in Israel, but in England (and I assume in Scotland) there are strict mark schemes specifying exactly what you have to include to get the marks. This actually used to backfire on me, because I'd get the right answer, but omit to write all the trivial steps, so lose marks.

Re: Maths exams

[andrewducker]

How strict? I'm intrigued, and would love to see some examples.

Learning to code

[woodpijn]

Well, that's one of many ways the route to learning to code can go. It doesn't match my experience. For me, things got pretty much monotonically easier, coinciding with the rise of the internet and the improvement of Q&A sites. My biggest struggle was with the basic concepts, because I was quite young, and because I had nothing but the QBasic help manual from which to try and figure out what a function was and why you might want to use one.

I'm not sure whether he's claiming his "four phases of the typical journey into coding" are inherent to coding, or just typical for the current culture and the current state of the internet (to the extent that he's doing either, as opposed to just trying to drive traffic to his learn-to-code site).

Re: Learning to code

[andrewducker]

It seems to be how most people I know go through it - without a big project to make you want to do more a lot of people get stuck, and going from "I see how a 'for' works" to "I can structure a large project reasonably well with hundreds of classes" is something that most people seem to find hard.

It's also something I've seen people struggle with when moving from university to work. At uni you rarely code with more than a couple of other people, and usually by yourself - where I work you'll have multiple teams all working on the same project (and connected projects), and managing that (and the associated systems, testing frameworks, build servers, version control systems, etc) is a lot to take in.

RE: Re: Learning to code

[lilchiva]

I am currently in this no man's land. I was just bitching about it even.

Re: Re: Learning to code

[andrewducker]

It's hard. Easier when you have support, and something to work on, but still hard to change the way your brain works to make it intuitive.

Feel free to moan about it here (if you're not moaning about it on your own LJ)

[starbrow]

That caption set about Bilbo really captures one of the things the movies got very, very wrong: he's just too nice and too happy to be there