Building at the Speed of Funny
Sep 13, 2011 23:20

I've mostly built fairly large and complicated web apps, either for myself or for clients, but recently I built a web site / app just for kicks, since I had a clever idea for something funny. It turned out to be a remarkably quick and inexpensive process, and it says a lot about the current state of web apps and hosting.

I've written on my personal blog about where the idea of came from. The short version is that it's a respectful spoof of I had been telling people that it would be amusing to do a funny riff on the name - but one night in August I decided to just go ahead and do it myself:

I hit and registered and I threw together a one page flat HTML website, used Pixelmator to mash up a picture of downtown Halifax with another picture of the Space Shuttle Atlantis taking off, stuck it on my server, and pointed the new domains at it. Even though it was now about 2am, I was now feeling kind of wired, so I set up a Twitter account (@stshipstarthere) and a GMail account ( I also set up Google Analytics tracking for the site, so I could see who was visiting and from where.

This is how far the internet has come in 2011: from a sleepy idea to a fully set up website with dedicated email, social media, and analytics tracking - in about three hours. The most time-consuming part was trying to properly mask out the smoke clouds in Pixelmator (nothing against Pixelmator - my image editing chops are stale).

I left the site as a single page on my own server for a few weeks and shared some laughs over it with friends. Over the Labour Day weekend, I decided to ramp it up to something more interesting: I turned it into a full database-backed Rails app. This still only took a few hours, since I kept things as simple as possible: I used the standard Rails scaffolding for the back-end, Paperclip and Amazon S3 to store the images, and some JavaScript adapted from another recent project for the front page slide show. The code is stored in my private repository in GitHub.

I was having so much fun with this idea that I figured it might get some traffic, and I didn't want the extra load to interfere with the business apps I'm hosting on my server - so I decided to move it to I've been using Heroku for several of my recent Rails projects and I like it a whole lot. It's also free for small projects, (and this definitely still counts as small).

(It had occurred to me that I could have just used a Tumblr instead, or even just a Flickr set or some other photo sharing/commenting site - but building web apps is what I do, and it was a refreshing change to build something simple from components instead of something complex with a lot of customizations.)

On the Tuesday morning after Labour Day, I set up a Facebook page for starshipsstarthere, added "Tweet" and "Like" and "Google+" links to bottom of each page on the site, and started getting the word out. After one week, the site has had 2853 visits and 8272 page views, as well as 117 Facebook "Like"s, 54 Twitter followers, and a Klout score of 40. It has made into local media and is receiving several new pictures a day from as far away as Korea.

Total cost? $30 for the domain registrations on (no elephants were killed and no women were objectified in the process) - and about 50 cents per month for S3 to store and serve all of the pictures. The private GitHub repository costs about $1 / month. All of the other things, from hosting to tracking and social media, are free.

Total time? The basic registration and initial flat site took about 2 hours. The full interactive site with the slideshow and back-end admin interface and social media hooks took about 4 hours of work before it went live. I've been making adjustments and adding features (better navigation, an approval workflow) so by now I've probably put a total of 10 hours into the project. This doesn't count the hours I've spent making fun mashup pictures of spaceships and Nova Scotia scenery - but that's way too fun to count as 'work'.

So the lesson is: in 2011, for simple projects, web apps are a solved problem, and if you know what you're doing and are using the right tools it can be really quick and inexpensive. Now if you get a good idea, you can actually build something before the idea gets cold.

Screencast Tips
Jul 14, 2011 18:54