November 2016 Summaries

At Repl.it, a platform aiming to make programming more accessible, they are introducing a new feature that allows users to debug their Python 3 code directly from the browser without any setup time. This new debug pane provides common operations typically found in debuggers, such as stepping through code and evaluating expressions. The feature is currently in beta and has several improvements planned, including support for breakpoints, a scope pane, watch expressions, and more. Users can try it out with Repl.it's recursive factorial program and provide feedback or suggestions to help shape the future development of this feature.

Repl.it` is a simple yet powerful in-browser development environment that initially ran only languages that could be compiled to JavaScript and run in the browser, but later expanded to include more features such as filesystem access. The service was built using Go, Docker, and multiple other languages responsible for evaluating code in their respective environments. To scale the service, the author used EC2 instances, which are virtual private servers, and added horizontal scaling by launching multiple machines with the same configuration. They also used AWS's Elastic Load Balancer to distribute traffic across instances and addressed issues such as single point of failure and round-robin routing. The author eventually automated the deployment process using Codedeploy, but encountered issues with visibility over deployments, lack of comprehensive deployment strategies, and bugs in the system. After addressing these issues, they moved on to using Ansible for automation, which allowed them to launch new clusters quickly and easily. The author concluded that scaling a service like Repl.it required a combination of technical expertise, patience, and continuous learning.

The introduction of lint support for Python3 marks a significant advancement, allowing developers to identify errors and potential mistakes in their code in real-time as they type, rather than running the code and manually checking for errors. This feature is powered by pylint, which not only detects syntax errors but also highlights potential programmer mistakes such as unused variables and redefined built-ins, thus streamlining the coding process and enhancing error detection efficiency.