Disclaimer: As a blogger I leverage the creative license to call solutions elegant, although in reality they may be far from it. Descriptive veracity is up to each reader.
Tag: ruby
Imagebrowser Rails app ported from 2.3.4 to 3.1beta
Rails 3.1beta deployed to Heroku from your iPhone
The tools:
- Local shell, LogMeIn or an ssh shell to a terminal system on your phone/tablet
- a system with git and heroku (I haven’t tried installing these on my phone)
- Internet
- DropBox and DropText for easy mobile editing. Alternatively you could use github as a web editing/auto-commiting platform
The Four Horsemen of Rails 3.1beta, Coffee-Script, jQuery, SCSS and Assets
A brief description of CoffeeScript, jQuery, SCSS and Assets
CoffeeScript is a translation and JavaScript with additional features to ease development, code readability, and functionality. It has all the best features of JavaScript as it IS JavaScript, but adds on essential development perks like terse syntax, local variable scoping, and utilities.
Why’s Poignant Guide to Ruby
Rescuing Ruby Exceptions in Sinatra
This morning granted me 25 minutes to debug an Oauth callback in AutoSub, my latest list subscription juggling tool. While an older spawn of Victus Media ImageBrowser does a fine job of doing the Oauth handshake in Rails, there were a couple of subtle differences in the implementation.
Reviving Sinatra on Mac OS X with a touch of Oauth
Building an App for Mac OS X Part 2
This is a continuation of my exploration of developing a Mac OS X application, particularly a MacRuby app which has been designed well with deep integration into Cocoa objects implemented in objective C. The series is now two posts strong 
(more…)
Building an App for Mac OS X Part 1
This is the first post in a series on understanding the basics of developing for the Mac platform. As I proceed with building the app, I’ll share helpful documentation and discoveries along the way.
Exploring speculative execution and evaluation
Consider a simple search problem with multiple paths. The main processing path can be computed in series as required functionality is identified. Completion occurs whenever an adequate solution is found. Alternatively, potential paths can be searched in batches or all simultaneously ahead of time**. This diminishes the wait for required functionality at the cost of additional processing, some of which may never be necessary. We trade computation power for statistically lower execution time.