thechrisoshow

Brainstorming doesn't work

2012-02-08T00:00:00Z

The Jan 30 edition of the New Yorker has a fantastic article by Jonah Lehrer about brainstorming. Read it here: GroupThink by Jonah Lehrer (subscription required).

The conclusion of the article is quite surprising. Studies show that groups using brainstorming techniques churn out fewer ideas than if the individual participants had written down a list of ideas on their own. Also, the ideas that were generated were poorer quality than the individual efforts.

The ineffectiveness of brainstorming stems from the very thing that we thought made it important – not critiquing other’s ideas.

From the article:

“Debate and criticism do not inhibit ideas but, rather, stimulate them relative to every other condition,”

Lehrer explains that the key to a productive group thinking session is to encourage debate and dissent. When someone says something that you don’t agree with it gets the brain working, and forces you to question your assumptions. Rather than spit out any random thing, you actually have to think!

So, let’s put this into practice! But how do you have a reasoned debate session that doesn’t erupt into a heated argument?

1. Appoint an adjudicator

Make sure someone is in control. Their job is to keep everyone on topic and to ensure that everyone gets a fair say.

2. Stay on target

Start with phrasing the problem like a question and have it written down on a white board for all to see. If the adjudicator is spotting someone going off target they can then swing them back by pointing at the question.

3. Critique the idea not the person

Instead of saying ‘You are wrong’ say ‘I disagree with you’. The key is to focus on what people are saying.

Avoid inflammatory phrases. As soon as you say “That’s a stupid idea” it sounds like you’re saying “You are stupid.” This naturally causes people to get defensive and guarded. If the idea is stupid, you need to explain why it’s stupid.

12 Angry Men - the best way to come to an agreement?

4. Listen

One person should speak at a time while the group listens. Put your pride aside and listen to what people have to say. Don’t interrupt when someone’s saying something that sounds dumb, instead make a quick note on a piece of paper and then continue hearing them out. Let them have their say, and when they’re done let ‘em have it.

5. Have a time limit

Debates can get tiresome – especially if you’re not sure how much longer you have to go! Make these meetings last no longer than 25 minutes. Any shorter and you don’t have enough time for a proper debate, any longer and people lose interest and their brains get fried.

Auto-refresh a webpage with javascript html

2011-06-15T00:00:00Z

I’m currently working on an app that has a dashboard-style page that we needs to be automatically refreshed. After digging around, I realised that the code to do this was ridiculously trivial:

UPDATE

Frederic in the comments mentioned a way nicer way of doing this, without needing Javascript. Simple add a refresh meta tag in the header of your html.

e.g.

The browser automatically knows how to handle it, and you don’t need any messy javascript!

Find more information about meta refresh here: http://en.wikipedia.org/wiki/Meta_refresh

How to host a static website on S3

2011-06-05T00:00:00Z

Updated: January 19, 2012

UPDATE: I was horribly wrong about how to do the domain forwarding for naked domains – however there’s a simple answer. The good folks at wwizer have got a free naked domain rewriting service. In order to use it simple change your A (Host) to point to the ip address 174.129.25.170 and you’ll be sorted.

In February this year Amazon announced that you can now host static websites on S3. This is a great cost effective way of hosting small, simple websites (like an iPhone app site for example).

Here’s how you can do it:

Buy the domain name for the site you want to host. It’s best to do this first, because the S3 bucket name has to be the same as the domain.
Get an S3 account (it’s cheaper than you think, I only get billed 3c a month).
Log into the aws management console and create a bucket with the same name as your domain (complete with www). So for my website www.bigbuttonsapp.com I called the bucket www.bigbuttonsapp.com
So that people have permission to view your website you’ll need to upload a bucket policy. Select on the newly created bucket, and click Actions > Properties. On the permissions tab click Add bucket policy and copy the following in (swapping www.bigbuttonsapp.com for your own domain):

{
  "Version":"2008-10-17",
  "Statement":[{
    "Sid":"PublicReadGetObject",
        "Effect":"Allow",
      "Principal": {
            "AWS": "*"
         },
      "Action":["s3:GetObject"],
      "Resource":["arn:aws:s3:::www.bigbuttonsapp.com/*"
      ]
    }
  ]
}

While you’ve got the bucket panel open, click the Website tab and check Enabled. Also put the filename of your index file in the Index Document field.

You can also optional include the name of your Error Document. This is the will be the document that is shown for any 4XX class of errors.

Click the Upload button and upload your website into that bucket. (In my case it was 3 files, index.html, error.html and styles.css)
To test it worked, click on the Endpoint link on the Websites tab of the bucket properties. Hopefully you should see your website up and running! (It might take a few minutes to kick in).

Domain names

Almost there, now just have to get the domain name running correctly. For this you’ll need to set up a CNAME redirect for www.
I know it’s horrible, but I use Godaddy, and here’s how I got this to work.

Log into the Godaddy domain management console and edit your zone file.
Amazon S3 static website doesn’t support naked domain rewriting, but thankfully wwizer provide it as a free service. In order to use it simply change the A (Host) to the ip address 174.129.25.170
Edit the www subdomain to point to the endpoint that S3 gave you.
Save the Zone File
Click save, and now you should be able to browse to your site, all up and running!

BONUS UPLOAD SCRIPT

Here’s a script I created that let’s you upload your files to S3 with ease. It doesn’t handle subdirectories, and it uploads all the files all the time (which isn’t very cost efficient) – but for small websites it’s pretty useful.

To use it:

Install the S3 gem with gem install aws-s3
Copy the script into your website directory
Change the access key, secret and bucket name to your amazon credentials.
Run it with: ruby upload.rb

Introducing Big Buttons

2011-04-30T00:00:00Z

Here’s a video demo of an iPhone app I’ve been working on called Big Buttons. The idea behind it is that it allows you to control your mac with your phone.

I’ve made a bunch of buttons that do a wide range of things, but users can also make their own buttons using AppleScript.

I have a modest budget for a designer to help me make Big Buttons beautiful, so email me at thechrisoshow@gmail.com if you’re interested in helping out. Also email me if you want to join the beta.

Speed up gem installs by not building the docs

2011-01-24T00:00:00Z

You know what it’s like – you install a gem and are confronted by minutes of this:

Installing ri documentation for bananas-0.4.9...
Building YARD (yri) index for bananas-0.4.9...
Installing RDoc documentation for bananas-0.4.9..

I never look at the docs on my local machine, I either read the source, or look up the docs online. So why bother installing all those docs?

Edit your /.gemrc file and add the following lines:

install: --no-rdoc --no-ri --yri
update: --no-rdoc --no-ri --yri

Boom – no more updating your docs – installing gems just became RIDICULOUSLY FAST.

Are your cukes not timing out?

2010-11-23T00:00:00Z

Whenever we’d run the full cucumber suite we found that were not timing out if they couldn’t find a particular element on the page.

For us, the problem was to do with Timecop.

You see, we’re using Timecop in our features like this:

  # in features/cucumber_feature.feature
  Given the date is "21 June 2010"
    
  # in features/step_definitions/time_steps.rb
  Given /^the (date|time) is (.+)$/ do |_,string|
    Timecop.freeze(Chronic.parse(string))
  end

This is great, because it allows us to fix the time to whenever we want. And because we use Chronic we can use natural language.

However, because this messes with time, the time outs weren’t getting called properly! The easy fix was to put this in our a file in the features/support directory:

  # in features/support/hooks.rb
  After do 
    Timecop.return 
  end

This gets called after every scenario and sets the time back to normal, thus removing the timeout bug.

How to make embeddable snippets with Rails

2010-10-19T00:00:00Z

Embedding a gist is a doddle – you create the gist, and copy the snippet, and BAM you end up with something like this:

I wanted to add this ability to Spoiler Tweet so that people can embed spoilers on their blogs (and also in the hopes that one day I’ll get access to #newtwitter ’s right hand side pane.)

Here’s an example of a Spoiler Tweet embed:

How’d I do that? Well, with Rails it’s incredibly simple. In Spoiler Tweet I have a TweetsController for displaying Tweets.

# app/controllers/tweets_controller.rb
class TweetsController < ApplicationController
  def show
    @tweet = Tweet.find(params[:id])
  end
end

All I had to do to get an embeddable snippet to appear was to add a show.js.erb template file:


document.write('')
document.write('<%= escape_javascript(render(:partial => "embedded_tweet")) %>');

What this does is insert a javascript snippet into the page that does two things:

include a custom stylesheet (for styling your embed). Make sure that this css file is compact and specifically targets your markup.
inject a partial onto the page

Note the escape_javascript method – this ensures that your partial parsed in a way that js can understand.

Now to embed the file, just simply have your users put something like this in their template files:

Pretty simple!

Get real with your specs, punk

2010-04-27T00:00:00Z

In my rspec spec definitions I was finding myself constantly repeating the word “should”. Like:

it "should add two numbers together" 
it "should taste like bananas"

And it’s not just me that does this, the standard rspec rails generators have “should” in them too:

it "should create a new instance given valid attributes" do
  Banana.create!(@valid_attributes)
end

This irritated me, why can’t rspec be like shoulda and include the word “should” in the actual spec definition like this:

should "taste like bananas"

I asked my Twitter followers what I should do, and Joe Martinez responded:

Should is just noise! Eureka!

Jon Larkowski from Hashrocket gave a talk at The Scottish Ruby Conference about some of the more idiosynchratic aspects of RSpec, and sure enough in his examples there weren't any sign of “should” in the examples. (Read it here)

So, now my specs look like this:

it "smells like bananas" 
it "adds two numbers together"

This just feels better – more decisive somehow. As my colleague Andrew Donaldson said:

“Gives a rough manly edge to your specs, Clint Eastwood would never say 'should'”.

So, in order to remove the unnecessary shoulds from my specs I came up with this RSpec formatter:

This produces a list of your specs that include the word “should”.

So be like Clint, and be decisive with your specs, Punk

An example of the art of instant feedback

2010-04-15T00:00:00Z

This morning on Twitter:

I love stuff like this. Kudos to the freerange team.

Gravatar default images - a better way

2010-01-26T00:00:00Z

At Harmonypark we decided to put gravatars into one of our applications (which is so easy it’s not funny) – and when the time came to select the default images for the gravatar I piped up: “I know how to fix this!” and pointed to a blog post I wrote about it long ago: (You can read it here.) My colleague Ebony came up with a great solution:
“Why not use another gravatar as the default image.”

Which is an utterly brilliant idea – that way the default image gravatar will scale correctly, and we wouldn’t have to mess around with having different image sizes for the different scaled default images. Here’s how we did it:

1) Go to gravatar.com and add a default gravatar for some email address that you won’t use.

2) Install the gravatar plugin

script/install http://github.com/woods/gravatar-plugin/

3) Extend the gravatar plugin to allow for a default email address by putting this code in lib/gravatar_extension.rb (remember to include this file in a initializer).

module GravatarHelper  
  module PublicMethods  
    def gravatar_for_with_default(object, options={})
      options.merge!(:default =>  
           gravatar_url("your-default-email@example.com", options))
      gravatar_for_without_default(object, options)  
    end  
    
    alias_method_chain :gravatar_for, :default  
  end  
end

4) Now, everytime a user is missing a gravatar, the default image will be your default gravatar, this will handle all the awesome scaling stuff that gravatar does too