Hey Marko,

Yes. The concept of Middleman is to generate a static site which consists of HTML/CSS/Javascript files. Any YAML data incorporated from your tempting files will be rendered static into the HTML files.

Lately I’ve been working a lot with RubyMotion (and loving it). In iOS development it is common to group your delegate methods under a #pragma mark. Xcode manages these automatically for you whenever you assign a delegate or datasource, but your favorite editor (TextMate) does not. So your files will become cluttered very fast whenever you are implementing multiple delegates and/or datasources. Luckily there is a way to get pragma marks for your TextMate symbol list…

stdout-kch:

In Objective-C code you’ll often see blocks of code preceded by something like this:

#pragma mark Foobar methods

That’s a directive that helps you organize code.

In Xcode and TextMate (and probably every other editor), you’ll find in the editor’s status bar a menu pop up that allows you to…

jayrobinson:

Lately I’ve been doing a ton of reading on decreasing website load time, image optimization techniques, responsive loading for different devices, and every detail I can find in between. One quick takeaway is that I will always tick the “progressive” checkbox in Photoshop’s Save for Web dialog…

If you’re awesome and practice test-driven development and you happen to know interwebs memes like Nyan Cat, you should definitely check out Nyan Cat RSpec Formatter. It takes your RSpec test results and displays them in the familiar Nyan Cat rainbow. Nyan Cat music not included.

Lately I have been working a lot with YAML data in my HAML templates via Middleman. YAML is great for providing data to display on your site. It allows for easy modification and addition trough separate .yml files instead of having to copy-paste HTML/HAML. YAML is very intuitive to write and work with from Ruby (and templates). But if you are in need of some good documentation, check out this YAML Cookbook.

Jonathan Snook shares his vision and some good tips and techniques on how to organize your CSS code.

It’s great to use innovations such as CoffeeScript, Sass/Less and templating engines like Haml and Slim but they all need to be compiled in order to be understood by browsers. Command-line tools are provided but it can get tedious when working with multiple of these tools at a time. Building a huge rakefile might be a solution but it would be nice to have a flexible and ready to use solution, right? Luckily there is!

Enter Middleman: static site development at its finest. Middleman is static site generator based on Sinatra. It consists of a single command-line tool that watches a directory for changes and runs a server serving the compiled files.

Installing Middleman

To install Middleman and initialize a project:

$ cd my_project
$ gem install middleman
$ middleman init my_project --bundler --template=html5

I used two flags: --bundler creates a Gemfile containing a fixed version of middleman itself. --template=html5 adds HTML5 Boilerplate as a starting point. You can use the added index.html and style.css if you are OK with using plain HTML and CSS. I prefer HAML and SASS. You can drop in these HAML/SASSified versions of HTML5 Boilerplate I created. Or download the latest versions from this actively maintained Compass version of HTML5 Boilerplate at GitHub.

Start the server from the command-line:

$ cd my_project
$ middleman server

.. and open http://localhost:4567 in your browser.

Now just start developing your site! Any file you drop in your projects source folder will be automatically compiled by Middleman. The terminal will also notify you of any errors in your files.

Exporting your site

When you are done developing your site you will need to build it. Building means compiling all the source files for a production environment. This can be achieved by running the middleman build command:

$ cd my_project
$ middleman build

You will find the compiled site in the build folder.

Compressing assets

Middleman will optionally minify CSS & Javascript and compress images. You can enable these optimizations by editing the config.rb file to include the following lines:

configure :build do
  activate :minify_css
  activate :minify_javascript
  activate :smusher
end

Smusher image compression requires for the middleman-smusher gem to be installed:

$ gem install middleman-smusher

If you’re using Bundler, just add gem "middleman-smusher" to your Gemfile and run bundle.

Automatic browser reloads with LiveReload

LiveReload is a Ruby gem and browser extension which automatically reloads your browser when you make changes to source files.

This is how to install it on OSX:

1. Add rb-fsevent and guard-livereload to your Gemfile
2. Run bundle
3. Install a browser extension (Safari, Google Chrome, Firefox)
4. Run middleman server --livereload
5. If you’re using Safari, right-click the page and click ‘Enable LiveReload’

Your browser will now automatically reload upon file changes!

Follow the installation instructions on the LiveReload Github page for instructions on how to install on Linux and Windows.

Conclusion

Working with Middleman is a really pleasant experience. It definitely changed the way how I work on static sites. Its flexible system allows you to swap any templating or markup language and there are a whole lot of other features to discover like YAML data, templates & layouts and a ‘blog mode’. I tried a whole lot of other solutions like Nanoc and writing my own rakefiles, but in the end I definitely recommend Middleman. For more information, check out the excellent documentation over at middlemanapp.com.

I am currently working on a responsive design which contains an embedded Vimeo video. After spending an hour trying to get the damn video to scale (while maintaining aspect ratio) I found this wonderful jQuery plugin called FitVids. It handles it all in just one line:

($ '#videoContainer').fitVids()

Check it out!

rentzsch:

Good is the enemy of Great Latin-1 is the enemy of UTF-8

You write web apps. You understand the web is global, and want to support internationalization. You want UTF-8.

UTF-8 is extremely sane. Well, as sane as an encoding can be that features backwards-compatibility with ASCII. Everything…

It came to my attention that teehan+lax, the team behind the excellent iPhone GUI PSD did an Android GUI PSD too. This PSD contains all Android interface elements so you can use them in your Photoshop mockups. Check out their downloads page too, sweet!

Last week I went on a bootcamp with my team to build a prototype for an upcoming web app. We decided on a test-driven approach to the project. I feel like this is one of the best decisions we’ve ever made. It might be slow to get into, but once you are you never want to go back to regular ‘ol programming style. Never I’ve had so much faith in the code I’ve written. It also really is a huge step forward in productivity.

I’m working quite a lot with Javascript as I’m the front-end developer on the team. In this tutorial I will now explain how I handle with tests in Javascript and I hope I can convince you of the awesomeness that is test-driven development!

The ingredients

There are a couple of ingredients to the automated testing workflow I’ve come to love:

1. Jasmine - A testing framework for Javascript
2. Jasmine-jquery - Some sweet Jasmin selectors and fixture support
3. CoffeeScript - Programmer happiness++
4. Jasmine Headless Webkit - Outputs feedback into the terminal
5. Guard - Makes it possible to run tests directly when you save them

Prerequisite

Jasmine Headless Webkit uses a QTWebKit Widget to run the tests in. Therefor we’ll need QT on our system. If you’re on OSX, the easiest way to install it is via Homebrew:

$ brew install qt

If you’re on another OS check the Jasmine Headless Webkit website and look in the Qt 4.7.X section for instructions.

Setting up the Ruby environment

If you haven’t already done so, install RVM by reading the installation documentation. Now create a new RVM configuration by running the following commands in a terminal in your project directory:

$ rvm use 1.9.2@myproject --create
$ echo 'rvm use 1.9.2@myproject' > .rvmrc

This will create and use an enclosed Ruby 1.9.2 environment within your project and will make sure it will use the correct Ruby version every time you enter your project via the command line. Try this by cd .. and cd myproject. The first time, you’ll be asked to confirm you want to use the .rvmrc file we just created.

We will now create a Gemset using Bundler. With Bundler we will create a “Gemfile” which will list the gems needed for our project. Installing it is as easy as running:

$ gem install bundler

Now create a file called Gemfile (yes, without an extension) in your project directory root and add the following lines to it:

source 'http://rubygems.org'

gem 'jasmine'
gem 'jasmine-headless-webkit'
gem 'guard-jasmine-headless-webkit'

Save the file and run the following command:

$ bundle

This will download and install the gems and their dependencies to a contained gemset within the RVM environment. It will also create a new file called “Gemfile.lock” which is kind of a binary file containing the gems.

We’re now ready to configure our environment. Let’s get started!

Setting up Jasmine

Jasmine comes with an init function which will set up a directory structure for you. Run the following command from your project root directory:

$ bundle exec jasmine init

With bundle exec, the jasmine executable will be called from the gemset.

Download the latest release of jQuery and drop it into public/javascripts/vendor (you’d have to create the ‘vendor’ folder). Next download jasmin-jquery to spec/javascripts/helpers.

We’re now going to edit the Jasmine configuration file. Open up spec/javascripts/support/yasmine.yml and replace the contents with the following:

src_files:
  - 'public/javascripts/vendor/jquery-1.6.2.min.js'
  - 'public/javascripts/*.{coffee,js}'

spec_files:
  - '**/*[sS]pec.{coffee,js}'

This ensures that jQuery is picked up at the start so we can use it in our tests. We’ll also tell Jasmine to look for .coffee files since Jasmine Headless Webkit supports CoffeeScript!

You could actually start testing right now but there’s one more thing..

Configuring Guard to automate testing

Guard lets us start a process when filesystem events occur. In our case, Jasmine will run the tests when we edit them. To enable Guard for our project, run the following command:

$ guard init jasmine-headless-webkit

This creates a “Guardfile” with all configuration we need. To start watching your project, simply run guard from your project root. This will automatically run all tests when you save a spec file.

Bonus (Growl notifications): Add the gems rb-fsevent and growl_notify to your Gemfile and run bundle to enable Growl notifications for Guard!

Let’s test our testing setup

Finally we’ve got our testing environment all set up so it’s time to test if our setup works.. The actual writing of tests is beyond the scope of this tutorial but for now I’ve provided a spec and an implementation file. I might do an in-depth testing tutorial soon.

spec/javascripts/pwd_spec.coffee

public/javascripts/Password.coffee

If you’ve run guard already, you should see that the 5 tests that came with Yasmine passed. Once you’ve created above files there should be 8 passing tests.

Further reading

We had to do quite some research to find the optimal testing workflow and I believe we succeeded in finding our way in this. I hope this setup will make the awesome test-driven development that everyone should practice somewhat easier.

If you’d like to learn more about this workflow, you’ll like these resources:

• Meet CoffeeScript Peepcode - all about writing Jasmine tests in CoffeeScript!
• jasmine-ajax - Easy AJAX testing
• Jasmine wiki

Just discovered a great use for Photoshop’s Content-Aware Fill function: hiding watermarks in stock images!

For a design I’m working on I needed a soil background. I was planning on just trying different ones but I didn’t want to buy them all. E.g. this photo on Dreamstime has this big Dreamstime logo in the center. Content-Aware Fill-ing this section hid the watermark perfectly. This way I could easily try several images in my comp before buying a larger size.

Fantastic browser extension (Safari, Chrome) to identify fonts. Works by hovering text, clicking reveals detailed info such as font-size and line-height.

Photoshop gives you the ability to create custom “New Document” presets. These come in quite handy when you’re designing for multiple (mobile) screen resolutions. From the “New Document” dialog you can create and delete presets, but for some reason there are no edit or export options. Fortunately we can add and edit these presets by editing a specific TXT file.

The default New Document presets get stored in a file called Default New Doc Sizes.txt located in Applications/Adobe Photoshop CS5/Locales/en_US/Support Files. If you’re on Windows it should be in /Program Files/Adobe/Adobe Photoshop CS5/Required/. You can edit this file to your likings, by adding new presets and deleting any unneeded presets. Below are the presets I use for Mobile (iOS and Android) design. You can add them between “Standard screen sizes” and “Generic mobile device sizes”, somewhere around line 189:

These values are taken directly from the iOS HIG and the Android Developers Guide. Note that these lines include the correct resolutions (pixels/inch) which could come in handy if you’re going to print your designs. Also, Color Profile: “Don’t Color Manage this Document”. Read Setting Up Photoshop For Web, App and iPhone Development on Smashing Magazine for more information on that subject.
The presets are currently subsets of the “Mobile & Devices” section. Unfortunately due to localization restrictions, it seems you cannot create your own sections. Hope it is of some help anyways!