A short while ago I wrote about visualizing transport by using 131500's TDX data, converted to GTFS, and served by GeoServer. Because I've started playing around with Google's Fusion table, I thought it would be interesting to see what all the transit stops in Sydney look like in FT. So, voila!

Recently I tried editing my Cisco's ACL at home on the train. It went something like this:

• I logged in
• I started updating the ACL
• I hit a blackspot in my 3g coverage
• My command stops at "router(config)#access-"
• I get a call from my partner saying "what did you do to our internet!"

Although it is simple enough to just ask her to "flip the switch on the black box", I still don't like doing it. Plus, if she's not home, I'm stuck. This accident immediately reminded me of one of a trait of the 'reload' command: it can be scheduled.

In the case of updating a device remotely, it is as easy as:

router# reload in 2
router# conf t
router(config)# [type in desired commands]
router(config)# exit
router# reload cancel

If the commands are entered in fine, then cancel the reload. If there is a problem, then the router will reboot and resort to the startup config.

If you want to use Graphserver to do some analysis with GTFS, you will need to convert GTFS into the database. This is how I did it.

Get an appropriate AMI from Amazon's EC2

I used the following AMI. If you have enough memory, you don't need to do this.

ami-7000f019

Lookup and read the GTFSDB INSTALL.txt document

Prepare system

sudo apt-get install mercurial
hg clone https://gtfsdb.googlecode.com/hg/ gtfsdb
sudo apt-get install python-setuptools
sudo easy_install psycopg2
sudo apt-get install build-essential

Download GTFS database

ubuntu@domU-12-31-39-00-5D-B8:/mnt/gtfsdb$ pwd
/mnt/gtfsdb
sudo python setup.py install
sudo wget http://cdn.kelvinism.com/google_transit.zip
sudo apt-get install python-psycopg2

Prepare configuration file

#default.cfg
[options]
create = True
database = postgresql://nsw:131500@10.128.49.175:5432/nsw
filename = /mnt/google_transit.zip
geospatial = True
#schema = None

Perform import

screen
python gtfsdb/scripts/load.py