Openpistemap and Mt Tremblant

I've done a bit of work on the tracks I collected at Mt Tremblant and started working on mapping it for OpenStreetMap.

I found JOSM a little painful at first, partly because I did not really know what I was supposed to be doing.

Switching to the latest version proved a good move, as the Ubuntu Intrepid version had a bug making the preferences dialog too small to change the things.

Anyway, with some help from the IRC channels, you can view the results here. Still rather a lot of pistes to add though.

UPDATE: the map link was showing my changes earlier, but they've now disappeared. I'm sure the rendering stuff will catch up again soon.

GPS and skiing

With the winter weather I'm unable to feed my GPS and cycling obsession. The good news is I've found a new obsession: OpenPisteMap. This is part of the OpenStreetMap project.

The idea is to map out ski resorts. I've been using JohnJohn to collect tracks of my recent ski trips. The results have been a bit mixed. One problem I've run into is that many ski hills are north facing, which means the GPS satellites tend to be below the horizon, giving a poor fix at times.

However, I am collecting some reasonable tracks. Below is some data from a trip to Mt Tremblant earlier this week. By plotting altitude, speed and direction (with matplotlib) you can get a reasonable feel for what is going on: I go downhill faster than uphill, you can see some lifts are faster than others etc etc.

Next job is to load the data into JOSM and see if I can create some sort of map.

Obama

A bit late, but here is a pic of me toasting Obama last week.

The T-shirt is one I had made with an image I'd created myself and it has a hidden message.

To see the message, here is the raw image.

Obama's inauguration brought back memories of Tony Blair being elected in the UK, which was a happy day for me for other reasons. People were optimistic about change, but it didn't quite turn out like that.



The image was created thanks to the Python Imaging Library and a few lines of code (see below).

I tried using the GIMP, but I'm afraid I don't use it often enough to be able to do anything non-trivial faster than I can write PIL code.

import optparse
import Image

import ImageFont
import ImageDraw

def get_weight(xoff, yoff, width, height, image):

    total = 0
    count = 0
    for x in range(width):
        for y in range(height):
            try:
                total += image.getpixel((xoff+x, yoff+y))
            except:
                continue
            count += 1

    if count == 0: count = 1
    return int(round(total/count))

def get_rgb(xoff, yoff, width, height, image):

    rr = gg = bb = 0
    count = 0
    for x in range(width):
        for y in range(height):
            try:
                r, g, b = image.getpixel((xoff+x, yoff+y))

                rr += r
                gg += g
                bb += b
            except:
                continue
            count += 1

    if count == 0: count = 1

    return int(round(rr/count)), int(round(gg/count)), int(round(bb/count))


def make_image(options):

    obama = Image.open(options.obama)

    options.width, options.height = obama.size
    scale = options.scale
    options.width *= scale
    options.height *= scale

    image = Image.new('RGB', (options.width, options.height))

    print obama.getpixel((10,10))

    text = options.text

    font = ImageFont.truetype("/home/me/.fonts/arial.ttf", options.font)

    draw = ImageDraw.Draw(image)
    width, height = font.getsize(options.text)

    xpos = ypos = height

    offset = 0
    total_offset = 0
    red, green, blue = options.red, options.green, options.blue
    
    while ypos < options.height:
        while xpos < options.width:

            for char in options.text[offset:]:
                wchar, hchar = font.getsize(char)
                if opts.rgb:
                    rgb = get_rgb(xpos/scale, ypos/scale, 1 + (wchar/scale), 1 + (hchar/scale), obama)
                    draw.text((xpos, ypos), char, font=font, fill=rgb)
                else:
                    weight = get_weight(xpos/scale, ypos/scale, 1 + (wchar/scale), 1 + (hchar/scale), obama)
                    draw.text((xpos, ypos), char, font=font, fill=(int(weight*red), int(weight*green), int(weight*blue)))
                xpos += wchar

            offset = 0
        ypos += height
        xpos = height
        total_offset += options.offset
        offset = total_offset % len(options.text)

    image.save(open('blair.png', 'w'))

def get_parser():
    """ Get an option parser.
    """
    parser = optparse.OptionParser()
    
    parser.add_option('--width', type='int', default=426)
    parser.add_option('--height', type='int', default=640)
    parser.add_option('--scale', type='int', default=2)
    parser.add_option('--offset', type='int', default=15)
    parser.add_option('--font', type='int', default=9)
    parser.add_option('--obama', default='obamagrey.png')

    parser.add_option('--red', type='float', default=1.0)
    parser.add_option('--green', type='float', default=1.0)
    parser.add_option('--blue', type='float', default=1.0)
    
    parser.add_option('--rgb', action='store_true')
    
    parser.add_option('--text', default="Please don't do a Tony Blair ")

    return parser

if __name__ == '__main__':

    parser = get_parser()

    opts, args = parser.parse_args()

    make_image(opts)

Socks and awe

Following last week's shoe throwing incident in Iraq someone with too much time has created this sock and awe game, presumably for people with too much time on their hands.

Meanwhile, here is a related pic of the US embassy in Ottawa:

 

Prorogue

Word of the week is prorogue. The PM here in Canada asked for Parliament to be suspended, to avoid losing a confidence vote.

I'm feeling a strange sense of deja vu here. I first came to Canada in June 1993, within 3 months there had been three prime ministers: Mulroney, Kim Campbell and Jean Chretien.

I just moved back here and and since then there has been an election and now this week's excitement.

Meanwhile, OJ is back in the news, just like he was when I arrived here in 1993.

I'm beginning to wonder if I'm stuck in a time warp.

Tim Henman finds his level

If you've been wondering what Tim Henman is doing nowadays, the good news is he has found his level at last.

Credit crunch solved

I just bought some electronic equipment and took a look at the bill:



It seems that even though my credit limit is zero, I have $1001 shy of $99 trillion of credi available.

So if my local electronics store has access to this sort of financing, is there really a credit crunch?