Monthly Archives: November 2013

Kano: ICT education, easy as Pi!

Posted on by

I stumbled across the Kano Kickstarter project this evening, and felt compelled to take to this blog in order to say what an excellent idea this really is!

There has been a lot of negative media coverage over the state of ICT education in the UK, and from my perspective this seems fairly justified. As far as I can remember, none of my ICT teachers in high school actually had any qualification in the field, and only one or two had any relevant experience to bring to the classroom. The majority were almost completely unaware of anything other than the allocated syllabus, but it only took one particular teacher (who has influenced my career path much more than anyone will ever realise!) with a passion for programming and the subject in general to get me hooked. It’s worth noting that this inspiration didn’t come from the taught subject matter itself, it was extra-curricular activities that really got me started in the field. ICT will continue to be a niche subject until the curriculum is updated to actively engage kids, rather than subjecting them to endless lessons on dry subjects on network architectures and database schemas. I may be biased as a kinaesthetic learner, but I think that the best way to get kids to engage with and learn this subject is by getting hands on.

The Raspberry Pi foundation have done a fantastic job bringing a cheap (£30) computer in reach of everyone. I own a couple myself for general tinkering and hacking about, and can honestly say it’s the main reason why I started playing with electronics, and gave me the confidence to start on a whole raft of new projects (such as this) which I never would have considered before. However, selling the raw pieces as they do, this machine seems weird and scary, outside the reach of the majority of educators and parents. While this is not a failing of the foundation itself (as I think they’ve been slightly overwhelmed by demand from the hobbyist sector), it’s crying out for someone to take this excellent system and package it in a more friendly way. Enter Kano.

Kano appears on the face of things to be a very simple project – they’re packaging up the Pi with the majority of peripherals needed to run it, and crucially they’re including kid-friendly instructions on how to get the whole thing working. Their use of the phrase “Simple as Lego” really struck a chord with me – that’s exactly the right way to approach this kind of teaching, by letting the kids play, hack around and figure it out themselves.

I really hope that the guys behind Kano take some of the money they’ve made from this project (it’s already 3 times over their target as I write this, with another 27 days left to run!) and take these kits into schools at a lower per-unit cost for education uses, just to make them a truly irresistible purchase for any ICT department. I genuinely believe that giving kids access to this kind of kit as part of their curriculum will not only educate them, but it’ll help inspire a future generation of hacker nerds – and that’s no bad thing in my view!

Managing music with beets

Posted on by

In a previous post, I talked about how I use Subsonic in order to make my entire music collection available over the internet to either my phone, or any computer via its web-interface. I’m still using Subsonic to achieve this, but had one fairly major gripe left with the set-up – I was managing the library on disk manually. Subsonic is great for editing ID3 tags on individual songs, but it relies on files being in sensible per-album folders in order to populate its library, something which I very quickly got fed up of doing manually.

I’ve known for some time about the Musicbrainz project, which maintains a database of all music releases, and has a number of applications built on top of it which will scan, tag and move your music collection as desired. I recall using Musicbrainz Picard back in the day to sort my library before I moved to hosting music on my own server, but never found anything similar that I could run on my Ubuntu server, until now. Enter beets.

Beets is a program that will manage your entire music library, allowing command line access, and interfaces directly with the Musicbrainz API to tag tracks appropriately. Plugins for beets also allow you to update genres according to Last.fm, download cover art (which Subsonic will quite nicely pick up!), and even acoustically fingerprint unknown files to figure out what they are!

Installing Beets was as simple as following the instructions on their getting started guide, however importing my existing music proved a little more tricky, and I had a few false-starts at doing this. I eventually found that the easiest way to do this was to move my existing collection into a separate folder, and set up beets to sort my collection back into the original place. The ~/.config/beets/config.yaml file I ended up using looks like this:

directory: /media/music
library: /media/backup/beets/musiclibrary.blb
import:
    write: yes
    move: yes
    resume: yes
replace:
    '[\\/]': _
    '^\.': _
    '[\x00-\x1f]': _
    '[<>:"\?\*\|]': _
    '\.$': _
    '\s+$': ''
art_filename: cover
plugins: fetchart embedart lastgenre

Put simply, my music lives in /media/music (an NFS share), with my library file on a separate backup share. When importing files, I want them writing (moving) to their new location, so all I need to do is run beet import New\ Album/ and the files will be tagged and moved into place (with most dodgy characters removed – unfortunately some special characters still seem to slip through). Album art is also downloaded into the new folder as cover.jpg, embedded into the files themselves, and also the genre field is populated using Last.fm. A nightly scan configured in Subsonic picks up the new files and adds them into the library, making them available to listen!

The next step for me is to integrate importing into the same process as my automatic sorting of TV shows, as I currently still need to manually import newly downloaded tracks. However, even this is a massive improvement on the tedious processed previously needed for getting OCD-quality tags on new music!!