I’ve had to learn C to do this, which has been a painful process for me and all the other people I’ve been badgering for help. (Thanks to the guys and gals on #edevelop and ##c!) However, I’ve now got to the stage where I can successfully draw a planet on the background, so I’m fairly close to getting something that’s usable, if not yet massively functional. There’s a few loose ends to tie up (not leaving loads of files laying around, for instance), and it appears to be leaking memory rather badly (give me garbage collection any day), but once those are slightly more under control I’ll put something up for people to try.

Now, at sea the weather can often get a little rough, so like all good vehicles, this ship had windscreen wipers – six of them.

Very sensible, and controls for the wipers were placed all the way across, so anybody (including us land-lubbers) could clean the windows if we felt the need. At least, that’s the theory: in practice, we just asked the officers to do it for us. Why? One look at the control panel will enlighten you (click to enlarge):

This control is impossible to understand without training. It is possible to control all six wipers from the panel, which is presumably why there’s duplicated symbols on some buttons. Of those symbols, I could hazard a guess as to what some of them might do; the zero probably switches off the wipers, but how about the spot? What does the light bulb do? It doesn’t illuminate the panel at night – I tried. The horizontal lines are meaningless too – if I had to guess, I’d say they were something to do with wiper speed, but these wipers have only one speed. Their function remains a mystery. As for the duplicated buttons, they apparently control different sets of wipers, but there’s no indication of which ones. The officers had learned the mysterious art of controlling the wipers, but all admitted that it would take too long to teach us. If we wanted the windows cleaned, we just asked.

Now, some of the controls on the bridge are incredibly complicated, and need to be – ships aren’t simple things to work with. But if the windscreen wipers can’t be made understandable, or at least easy to explain, then there’s something amiss. Whoever Wynn Marine are, their interface designers leave something to be desired.

GDM’s configuration is in the file /etc/init/gdm.conf. At the top of this file are two commands that specify when GDM starts, and when it stops. In Lucid Lynx (10.04), these lines look like this:

start on (filesystem
          and started dbus
          and (graphics-device-added fb0 PRIMARY_DEVICE_FOR_DISPLAY=1
               or drm-device-added card0 PRIMARY_DEVICE_FOR_DISPLAY=1
               or stopped udevtrigger))
stop on runlevel [016]

All we need to do, then, is to alter the start criteria. The easiest way is to replace the multi-line start command with this:

start on runlevel []

The runlevel approach tells GDM to start or stop when a given runlevel is entered. And if we give it an empty list on runlevels, it never starts. Simple!

Here’s a relatively easy method of writing a CSV file with a header. First, make sure you have two variables – one containing the numeric data for your file, and the other containing the header line as a string. I’m going to call my numeric data csvData and my header header, just to be original. Then I can use the following code to write both the header and csvData directly to the output file:

outid = fopen('out.csv', 'w+');
fprintf(outid, '%s', header);
for i = 1:length(csvData)
    outLine = regexprep(num2str(csvData(i,1:37)), '  *', ',');
    fprintf(outid, '%s\n', outLine);
end
fclose(outid);

Now you’ve seen it, I’ll explain how it’s done. In the first two lines, we open a channel to a new file, and write our header as the first line. Next, we loop through each row of the data set, converting it to CSV format and adding it to a file. Finally we close the file to tidy up.

The generation of the outLine variable requires a little more examination. First off, my data set has 37 columns, so I grab them all from the ith row, and run num2str on it (There is undoubtedly a way to work out how many columns your data set has programmatically – answers in the comments please!). This gives us a single string containing all the columns, each separated by a number of spaces. Finally, we use a regular expression to convert one or more spaces into a comma – this gives us our data in the desired CSV format, ready to be added to the file.

Note: This code was developed in Matlab 6.5. I assume that the newer versions work in much the same way, but I don’t have the resources to try it.

Update 6th July 2010: Walid suggests an alternative to my code in the comments, which apparently is much faster. I’ve not had a chance to check it out myself, but feel free to give it a go.

For the most part this isn’t a problem, apart from an occasionally annoying delay while the networking is sorted out before you can get down to some serious work avoidance on the Internet. However, if you want to have a headless computer on your wireless network, for use as a server, MythTV box, or a grunt for distributed rendering, you don’t want to be logging into this machine to get the network going whenever you switch it on. Fortunately, it is possible to get your machine to connect to a wireless network on boot without having to log in. Note that you’ll have to hard-code which wireless network you’re connecting to, but since this is a headless server and presumably somewhere in a dusty corner, that shouldn’t be a problem. So how’s it done? Well, here goes…

Before we start, I’ve made a number of assumptions on your setup, so if your system isn’t covered here, you may need to do a little more research elsewhere. Sorry, about that, but I can only advise on what I’ve done with my system.

• You’ve already got a working wireless network card in your machine. If you need help setting up drivers for it, I’m afraid this isn’t the place. I used the stuff on this page to get my Netgear WG311v3 working.
• You’re using WPA security on your wireless network. I’m not getting into the arguments here.

First off, make sure you have the wpasupplicant package installed, which will handle the security side of the wireless networking for us. (You are using WPA security, right?) It’s almost certain that this is installed, but we’d better check now before we cut the machine off from the Internet.

sudo apt-get install wpasupplicant

Now, as I’ve already mentioned all network connections are handled by the NetworkManager, so if we want to do our own thing we need to get rid of it. Note that once we’ve done this your machine won’t have any network access, so make sure you’ve got another machine close by in case of need. You can do this in two ways – either disable NetworkManager (something of an arcane process), or more simply (and the way I did it) remove the relevant packages:

sudo apt-get remove network-manager network-manager-gnome

OK, so now you have a machine with no networking capability at all. Never fear, though: your machine will soon be out of solitary confinement. Using your favourite text editor in sudo mode, edit the file /etc/network/interfaces and set it up as follows:

auto lo
iface lo inet loopback

auto wlan0
iface wlan0 inet dhcp

wpa-ap-scan 1
pre-up sudo /sbin/wpa_supplicant -iwlan0 -c/etc/wpa_supplicant.conf -Dwext -B
pre-up sleep 5
post-down sudo killall -q wpa_supplicant

So, what has this done? Well, the first couple of lines set up the loopback network device, which is used for all sorts of things in Linux. Don’t ever get rid of this unless you like a lot of pain in your life.

The second two lines tell the system that we want the wlan0 interface (that’s the wireless one) to start get going automatically, and acquire and IP address through DHCP. The final chunk tells the system how to go about setting up the connection:

wpa-ap-scan 1

This setting means that the ESSID of your wireless network is publicly visible. If your router hides the ESSID, set this to 2.

pre-up sudo /sbin/wpa_supplicant -iwlan0 -c/etc/wpa_supplicant.conf -Dwext -B
pre-up sleep 5

This specifies the command to be used to set up the wireless connection. -i is the name of the interface, -c specifies the location of the configuration file (more on this in a moment), and -B runs wpa_supplicant in the background (very important!). The -D option sets the type of driver to be used. The wext driver is fine in most cases, even if you’ve set up your network card’s drivers through ndiswrapper.

The final stage is to configure wpa_supplicant so it knows about your wireless network. In the interfaces file, we specified that the configuration file is located at /etc/wpa_supplicant.conf, so edit that file. The contents of the configuration file should be as follows:

ctrl_interface=/var/run/wpa_supplicant

network={
        ssid="<your SSID>"
        psk="<your wireless network's password>"
        proto=RSN WPA
        key_mgmt=WPA-PSK
        pairwise=CCMP TKIP
}

The ctrl_interface line tells wpa_supplicant where to dump any information it needs while it’s running. The setting here is the default, and I see no reason to change it. Below that is simply the details of your wireless network. Note that the ssid and psk value must be in quotes. The other values seem to work well for most wireless network situations.

That should be all you need to do. To see if it’s worked, reboot the machine. If you’ve set up an SSH server, you should be able to log in remotely straight away. Or, take a look at the list of devices connected to your wireless router, and you should see that it’s connected itself without your intervention. Now you can stick the machine in its dusty corner and let it do its thing.

The wireless network card in my machine unfortunately doesn’t support Wake-On-LAN. If it did, I could stick my server in the attic and never look at it again. Maybe I’ll try that one day.

In recent years, the minute’s silence at sports grounds has been replaced by the minute’s applause. Whether it’s thought that using the universal sound of appreciation is more appropriate to sporting and entertainment figures than the more sombre requirements of people killed in battle, or whether it’s to cover for the fact that a reasonable proportion of football fans are unable to focus on anything for a full minute (or at least stand still for that length of time) is unclear, but it certainly seems more fitting in its context.

I’ve been thinking about all this since listening to a program on commemorating the war dead on Radio 4 this morning, and wondering how I’d like my passing to be commemorated. I am reminded of the example of Jeremy Hardy, who once said:

“I would quite like to die in a way that causes maximum inconvenience to others – on an escalator, or while having my blood pressure taken. Or at somebody else’s funeral. Attention-seeking to the last.”

As a comedian, it’s understandable that Jeremy wants to get as much attention as possible, and it will certainly cause a few people to remember his passing. For myself, I prefer to cause as little consternation amongst large groups of people as I can, so lasting remembrance is unlikely to come my way through this route.

I believe I fit into that category of people who will never amount to much on the world stage, and will pass on with no corresponding outpouring of public emotion. However, should a large group of people feel that I’m worthy of being commemorated in a public act, I will at least want people to remember that. So neither a minute’s silence for me, nor a minute’s applause. I want a minute’s hopping on the spot. It would look ridiculous and unseemly, and most would fail in the attempt, but the thought amuses me and it should (if done properly) amuse everyone who’s trying to hop too. You may not wish to hop, and that’s your right, and I won’t be upset (I’ll be rotting). But you should be amused, and if you’re not I don’t want you commemorating me at all, you miserable sod.