March 12, 2010

Staying motivated on your personal side projects is really quite hard! Moose and I have moved closer to the city, so now I have a bit more time and space and have been hacking on Infinite 8-bit Platformer again. Here is a graph of commits-per-month on the project:

Infinite Platformer commits per month to March 2010

I prefer to think of them as "experience points" rather than commits, because making games should be the meta game. :)

I also made a page on the website where you can see the commit log.

I set up a twitter stream for the commits too: http://twitter.com/infinite8bit, and also an identi.ca account if you're that way inclined: http://identi.ca/infinite8bit

You can subscribe to the RSS feed of those commit messages: http://twitter.com/statuses/user_timeline/121975057.rss

click for detail

On my current contract with RjDj I have to work as close to London time as possible, so this means that I have some mornings somewhat free, and I've been spending the hours between 8am and 11am drinking coffee and working on my game.

As you can see, even at the beginning of this month there have been almost as many commits as there were last month, hooray! There are 37 line-items left in the TODO file at the moment. My rule of thumb is about two line-items per working day. If I can work at roughly 3 half-days per week I should have the beta release ready for testers in about three months. That's assuming that the networking library I previously wrote, PodSixNet, is up to the task of linking everyone together into some kind of platformy MMO goodness. If I do a release without the networking stuff it means I can get something in front of people in probably one month. That's a very motivating thought! The important thing is that I don't rush myself though. Slow software development makes better quality software (although implemented software is better than vapourware).

I have found that the best way to stay motivated is never to feel guilty about not working on something. If I don't feel like it I just forget about it and do something else. Eventually if the project is worth it I'll get excited about it and naturally come back to work on it again. Judging by the graph, that seems to be exactly what happens. Some months I leave it entirely (zero commits), but over the long term progress continues because somewhere deep inside I know this project is worth it. One of life's great lessons is that at some point we must always let go of the things we love, and I think this really applies to staying motivated about projects too!

Ok, time to write some more code. :)

Boring stuff

Here is the very small bash shell script which produces the csv file used to generate the graph (unfortunately it skips empty months so you have to add those manually):

#!/bin/sh
bzr log --short --forward | sed -n -e 's/.*\([0-9]\{4\}-[0-9]\{2\}\).*/\1/p' | uniq -c | sed -n -e 's/\ *\(.*\)\ \(.*\)/\2\,\1/p'

Here is a very small bash shell script that I wrote for turning emails into tweets:

!/bin/sh
header="START"
# concatenate all lines (ignore email header)
while read line; do
    if [ "$header" == "" ]; then
        message=`echo $message $line`
    else; header=$line
    fi
done
# post to twitter
wget -O - --user=xxxx --password=xxxx --post-data="status=$message" https://twitter.com/statuses/update.xml

That last line is a handy one-liner for posting to twitter!

To post to identi.ca, change the url to "identi.ca/api/statuses/update.xml".

Need software development advice? Book a call with me.