Living in San Francisco, engineering at Heroku, and happily writing software in Go, Ruby and JavaScript
Logging is one of the easiest ways for an application to communicate information to humans or other machines. How many requests has the application received or how quickly has it served responses? How slow are the slowest actions and how often do they occur? These can be easily communicated through logs.
Choosing an API versioning strategy can be difficult. There are many approaches and
each come with their own pros and cons. Two of the most common practices are to version
in the URL or with headers. For Heroku's public API we decided to version with the
Accept header.
We have many amazing open source projects today, but none of them were just amazing overnight and none of them would be considered amazing without the people that have and do contribute to them. Projects themselves are good ideas, but its the people behind them that make them amazing.
This is a follow up post to Rake Tasks 101. In the 101 post we created Rake tasks, setup dependencies and made our tasks reusable by passing in parameters. In Rake Tasks 102 we'll be building on those practices, interfacing with a Rails environment and leveraging the cron to automate our Rake task.
I've been working with Rake quite a bit on my current project so I thought I'd share some beginner tips.