Menu:

A Blog, by John Turner, featuring a collection of disconnected musings.
Latest posts
cambridge
spotify
qwop
language_highs
pirates
not_irc
spam_again
spam
gress
ps3s

cambridge

2011/10/10 12:56:24

We’ve moved into the house in Cambridge, boxes are still everywhere.

Figured I’d get this blog running again after it fell down in a mess of broken dependencies previously.

Rowan is 2 years, 1 month, 4 weeks, 2 days, 7 hours, 34 minutes, 32 seconds. He is saying “go” (involving a bit of a game where we say “ready, steady” and he shouts “go!” and then he gets swung or spun around. He’s also possibly saying a couple other words.

Other than the usual MHN work I’ve been working on a game called borpaborp off and on and hanging out with the CB2 indies group on the Tuesdays.


No comments

spotify

2009/02/03 13:47:07

Spotify is amazing.

Ad-supported (short audio ads, every 10 songs or so) streamed music of pretty much anything you want (except the Beatles, they don’t like the internets apparently), or you can pay monthly to remove the ads.

You can also build collaborative play lists, which I’d like to play around with more.

Works on mac and windows, linux under wine.

Poke me somewhere if you want an invite.

Add stuff to my suggestions play list here.


No comments

qwop

2008/12/13 15:06:00

qwop is one of the best games I’ve played on any platform recently.

It’s a game in which the aim is to run 100m over flat terrain. After playing it for probably an hour or so a day for the last couple weeks, my best score has been 46.2m.

It’s challenging, but not in a way that feels cheap, it’s always entirely your own fault when you mess it up. It’s funny, frustrating in a way that just fuels my need to complete it and filled with subtle metaphor. The controls are simple and taking your first few steps is not too hard once you read the help. Then there is the near infinite scope for gradual improvements and refinements of technique.


No comments

language_highs

2008/12/10 13:44:49

I get high on programming languages. I find a new one, read its philosophy and recognise the problems it’s trying to solve. I then start to use the thing, realise how it solves those problems nicely, at least on the surface. I then find out that you can do a range of interesting things with a language, solving a lot of problems more simply than you can with other languages. Later, the flaws start to show. The hacks that make things work under the surface, but which fall apart on edge cases start to come apart. I lose the impression of the language as this perfect representation of computation.

A perfect language would be one in which this hangover never happens, but that’s probably impossible, given the complexity of hardware, computation, mathematics and reality in general.


No comments

pirates

2008/11/20 11:15:19

It amuses me that the British navy is leading an armada of warships to fight the pirate menace. I find it hard not to think of it as a fleet of galleons sailing out from the mouth of the Thames.


No comments

not_irc

2008/11/11 22:39:54

Have quit irc for a while.

Still reflexively trying to check it constantly.

Will come back when that stops.


No comments

spam_again

2008/10/28 13:18:40

I’ve disabled comments for now and deleted all comments from the last month that contained any links. Sorry if I deleted any legitimate comments in the process.

There are a set of comments remaining from spammers that do not contain any links. A page of these is up here.


No comments

spam

2008/10/26 07:43:06

I hacked this blog together in Camping.

An experiment was conducted into minimal comment security. I wondered if any spammers would actually bother automating the “I’m human, honest.” checkbox, simple though it would be, for a blog that is not widely read. Seems someone did. I guess I’ll have to put a captcha up.


No comments

gress

2008/09/09 00:58:11

This is all probably pretty obvious, but it’s going through my head right now. It’s also a simplification, possibly even one of the gross ones.

We can’t really stay where we are as a species, we’re using up too many resources too quickly and causing too much damage.

We can either progress or regress. Progress is risky, might not happen quickly enough, but has the possibility of getting us to a Good Place. Regress will most likely lead to mass starvation and depopulation, but at least the planet will remain relatively usable, as long as we haven’t caused too much irreparable damage.

I like progress myself.


2 comments

ps3s

2008/08/16 15:39:54

I bought a ps3 recently. It is pretty fun. One thing it means is that I am no longer solely Nintendo’s bitch with regards to consoles. GTA4 multiplayer is AAAAAAA++++ would play again. It is a bit annoying how the important and hard things, like the netcode for complex physics and interesting game modes are really well done, but the easy things to get right, like not allowing randoms into games that are meant to be private or not randomly turning off your bluetooth headset which is your only means of in-game communication are not. Yay for run-on sentences. Do not try to read that out loud. If you do then I disclaim any liability of death due to suffocation.


No comments

computer_hat

2008/07/18 12:22:24

My working environment is not very inspirational, or healthy. It is an office on the 4th floor, filled with extremely generic office furniture. The majority of my life is spent sitting in front of my laptop, typing. I would like to be able to write, code, chat, and generally be connected on the move. Not all the time, but I’m getting tired of being trapped at a desk if I want to do these things efficiently.

A thing I want to buy, or make, or help make, is basically a computer hat. The main problem that needs to be solved for this is that of efficient hands free input of text. There has been a lot of work on brain-computer interfaces (see wikipedia for its usual just-about-trustworthy overview). I’d like to see how far you can get with an EEG, a big list of words, and some months training. Otherwise, eye tracking could be an option, or an accelerometer on a finger with some gesture recognition.

Invasive or “semi-invasive” brain implants are not an option right now. I’d also prefer to stay away from voice recognition because it’s still pretty inaccurate and it feels awkward and it wouldn’t work so well in noisy places and people would look at you funny for talking to yourself. Part of the motivation would be to produce a wearable computer that doesn’t make you look completely insane. It will still make you look a bit insane, but I think this could be limited. The head mounted display will be the most obvious thing.

As a tool I would like to use, I would need the hat to be capable of:
  1. running an operating system (ideally ubuntu, but I would settle for embedded linux to start with, Windows XP or OS X would also be fine I guess). #running a terminal of some form, Vim and a browser (links would be just about acceptable, especially now that lots of sites are providing “mobile” versions which are actually usable in links.)
  2. hands free text input (see above, yes)
  3. networking: 3G or wifi
  4. storage: probably a few gigs, probably an SD card.
  5. battery: chunky, would need to last for at least 8 hours of use. Ideally 16. Would have to find the right balance of duration and weight.
  6. display: some form of head mounted display. Either one of those tiny displays that could dangle from the brim of the hat, or the full on VR style goggles. Ideally switchable between the two depending on whether you need to see the outside world or not.
Other stuff that would be nice:
  1. audio: an audio jack and headphones would make sense. Also a microphone.
  2. usb
  3. ethernet because I am amused by the concept of by sitting around with cat5 plugged into my head.
  4. Both G3 and wifi.
  5. Bluetooth
  6. Maybe some kind of silly lcd on the outside of the hat which you could post text or images to that people would see. (probably not)

So, I’d like to see if I could turn the hands free text input into a research project at some point, unless someone else solves it or has solved it already.

Anyway, thoughts on this would be appreciated, as would links people have to existing work on stuff like this. If you’re reading this aggregated elsewhere, I’d like comments on norgg.org so discussion can be in one place.

Edit:

I realised over lunch that the initial prototype of this will probably use a gumstix (probably borrowed from the office when we’re not using it), which has bluetooth and wifi and runs embedded linux. I’d need to get an lcd connected up to one. Input could be done using a wiimote (because I have a wii already and they use bluetooth) and gesture recognition.


No comments

google_knows

2008/07/07 16:02:23

If you use the google apps that I use, it knows:

  1. The contents of your email. (gmail)
  2. What you read about (news, blogs and comics). (google reader)
  3. What you do socially and often where you do it. (google calendar)
  4. Where you live and where you ask for directions to. (google maps)
  5. What films you are probably going to go watch (google movies search)

With this information, google could decide:

  1. Who you should be friends with based on where they are and what you and they like.
  2. Where you should go to socialise with your friends.
  3. When you are free to do so.

So:

You don’t need to organise your social life any more, you should just let google do it for you.


1 comment

stuff

2008/06/25 09:09:02

It says… something about our civilisation that a common problem for individuals is that of having too much stuff.


1 comment

hacking_with_net_hack

2008/06/23 15:09:14

Whee.


No comments

malcolm_trail

2008/06/20 14:13:09

Science:


No comments