• The REPL: Issue 29 - December 2016

    Learn Graphviz and Up Your Diagramming Game

    I’ve bumped with Graphviz in the past only indirectly. After reading this article, I will definitely keep it in mind when needing to create diagrams in the future. I like the idea of specifying the relationships in a diagram in text – and manage it with source control – and then generating a graph from it.

    Microservices? Please, Don’t

    Sean Kelly explores why some of the often-repeated arguments for microservices. Certainly, some of the benefits of microservices can be achieved without needing to separate them into different web applications, which in itself can brings complications in deployment, coordination and increased network interaction.

    RbNaCL: The Ruby Cryptography Library

    This repository is a great idea: Provide a cryptographic library that makes it easy and straightforward to use high-level cryptography correctly, and avoid the many pitfalls of trying to assemble a secure system from cryptographic primitives. The APIs are designed to provide abstractions like public-key / secret-key encryption, digital signatures, etc.

    Read on →

  • Enforcing Style

    Most programming languages provide some flexibility on what they consider valid syntax. Take a boring piece of code:

    Read on →

  • Renewing a Let's Encrypt Certificate

    I previously wrote about changing my certificate authority to Let’s Encrypt. About the only downside I found about using it with my hosting service, Nearly Free Speach, is the need to manually renew every 3 months. Today, I went through the process and found it to be relatively simple.

    Read on →

  • The REPL: Issue 28 - November 2016

    Open-Sourcing Yelp’s Data Pipeline

    The Yelp Engineering team has been posting regularly about they structure data consumption between different teams. The backbone of their system is Apache Kafka, but they have created a lot of tooling around it. In this announcement, they have open sourced (Apache License 2.0) many of these tools. MySQL Streamer pipes data from MySQL to Kafka. Schematizer stores and tracks the various data schemas used throughout their pipeline. There is a lot to learn in this projects and the blog itself, which provides an overview of how they approach dealing with data.

    Offshoring roulette: lessons from outsourcing to India, China and the Philippines

    Troy Hunt writes a lengthy post about his experience offshoring development work to teams in India, China and the Philippines. He goes through the motivation for offshoring in the first place, the challanges and rewards, and the differences he encountered in different countries. His conclusion:

    if you’re looking at hourly rate as a metric for outsourcing success, you’re doing it very, very wrong!

    NIST’s new password rules – what you need to know

    The United States National Institute for Standards and Technology has come up with new guidelines for password policies. If you are wondering which password rules to follow in your product, these are a great baseline. Note that the NIST policies contradict the FBI’s. While you are at it, consider if you actually need to store a password at all. Medium, for example, emails you a link to log in. Because of “forgot your password” functionality in most sites, access to your email is essentially equal to access to the site. Medium just made it explicit and removed the need for them to store users passwords. Those passwords are probably re-used elsewhere. If you don’t store them, you can’t loose them. Right?

    Read on →

  • Benchamarking With abprof & abcompare

    This week at RubyConf I learned about two new tools.

    Read on →