| • Justin Saros' work on Lumen.js
|
| • Lumen.js as an alternative to monolithic Rails application development
|
| • Justification for the statement "JavaScript won't be a framework"
|
| • Lumen.js as a solution to alleviate problems with monolithic Rails app dev
|
| • Breaking down Rails responsibilities into build, application framework, and sensible defaults/conventions
|
| • Comparison between Ruby on Rails and Node.js for application development
|
| • Accidental creativity in open-source communities
|
| • Importance of sensible defaults and convention-driven design
|
| • Comparison of Ruby and Node.js ecosystems
|
| • Lineman as a tool for build tools, not a framework
|
| • Single responsibility principle in tool design
|
| • Integrating with front-end frameworks (e.g. Ember, Angular)
|
| • Avoiding code generation and promoting flexibility
|
| • Consultant perspective on choosing technologies for clients
|
| • Balancing traditional and rich JavaScript front-end development approaches
|
| • Client's system and needs are prioritized when working with them
|
| • Greenfield apps are built with consideration for user experience, budget, and business goals
|
| • Rich client applications may be necessary for certain applications, but can be more expensive to build and maintain
|
| • Simple, traditional web applications can serve companies well initially, but may become inefficient and unmaintainable over time
|
| • Agile development methodologies, such as one-week sprints, can lead to the "simple trap" where complexity is added in a monolithic, unstructured way
|
| • Breaking up a monolithic, complex application can be difficult and require significant rework
|
| • Discussion of the distribution of new projects, with 2/3 being "fat client" and 1/3 being "all back end"
|
| • Comparison of traditional Rails and JavaScript "fat client" projects, with a 50/50 split between the two
|
| • Explanation of the goal of Lineman, a JavaScript framework, to make "fat client" JavaScript web applications as easy to build as traditional server-side HTML web applications
|
| • Features and benefits of using Lineman, including rapid prototyping, consistency, and meta-plugins
|
| • Discussion of how to integrate Lineman with existing Rails applications, including proxying back to the Rails application and using a shared repository
|
| • Strategies for managing separate repositories for Lineman and Rails applications
|
| • The benefits of using PagerDuty for incident management and resolution
|
| • Comparison between Lineman and Yeoman as front-end project tools
|
| • Criticisms of Yeoman's approach to project setup and community-driven generators
|
| • Features of Lineman, including its table for comparing Lineman and Yeoman
|
| • Overview of Lineman's test runner, Testum, and its capabilities
|
| • Discussion of Lineman's testing story and its advantages over Yeoman.
|
| • Lyman's author mentions that Lyman's dependency management is limited, and it relies on a default configuration that "just works"
|
| • The author uses Lyman's helper directory to store their own test helpers, and suggests this may be beneficial to others
|
| • The author criticizes Yeoman's approach to dependency management, specifically Bauer, which is seen as a "fancy downloader" rather than a true dependency management tool
|
| • The author argues that Yeoman's approach encourages users to use Bauer as a dependency management tool, which can lead to issues with version conflicts and transitive dependencies
|
| • The author suggests that a better approach would be to write leaner, meaner applications that don't rely on a large number of JavaScript plugins
|
| • The author proposes that users should commit their vendor dependencies to ensure control over the application's dependencies
|
| • The author notes that Yeoman's approach to dependency management is "diametrically opposed" to their own, and that it's a matter of perspective and experience.
|
| • The importance of focusing on building an application quickly, rather than getting bogged down in engineering problems.
|
| • Critique of using Browserify, Require, and Bauer, as they introduce unnecessary complexity.
|
| • Discussion of Lineman.js, which is primarily written in CoffeeScript, but generates JavaScript output.
|
| • The author's preference for CoffeeScript, citing its benefits and maintainability.
|
| • Criticism of users who refuse to use Lineman.js due to its CoffeeScript basis, labeling it as a "straw man argument" and "entitlement".
|
| • Discussion of contributor feedback, with the author expressing frustration with users who open issues without offering to contribute or open a pull request.
|
| • The author's observation that the Ruby community is smaller and more homogeneous, whereas the JavaScript community is diverse and global.
|
| • Difficulty in managing and trusting open source dependencies
|
| • The "bozo button" concept for dealing with problematic open source contributors
|
| • The intersection of technical and social aspects of open source
|
| • Corporate backing of open source and its implications
|
| • The "social coding contract" and Justin's upcoming talk on the topic
|
| • The risks of relying on unmaintained or poorly maintained open source dependencies
|
| • The need for users to take greater responsibility in understanding their dependencies and their maintainers
|
| • The consequences of relying on a web of open source dependencies that may be insecure or unreliable
|
| • Open-source companies have a responsibility to give back to the community despite getting free value from open-source software
|
| • Maintaining open-source projects can lead to burnout due to the emotional toll of supporting users and dealing with issues
|
| • There is a need for a cultural shift in how users and maintainers interact, with users having a deeper understanding of open-source and contributing back
|
| • Demystifying open-source code can help users contribute and reduce burnout for maintainers
|
| • Both users and maintainers need to meet in the middle, with users having a deeper understanding of open-source and contributing back, and maintainers being more accessible and supportive.
|
| • The process of debugging and understanding code is becoming more accessible and encouraged for beginners
|
| • The increasing social nature of pull requests on platforms like GitHub is changing how people approach code and debugging
|
| • There is a need for better tools to help developers transition from using APIs and documentation to digging into code themselves
|
| • The languages Go and Rust may be easier to develop tools for due to their nature
|
| • The process of taking over an open-source project, specifically RSpec, can be complex and time-consuming
|
| • The importance of having a clear plan for maintenance and control in open-source projects to avoid control issues and encourage collaboration
|
| • Concerns about maintaining RSpec Given without sufficient resources or guidance
|
| • Comparison of RSpec Given to Jasmine Given
|
| • Inspiration from Jim Weirich's approach to open-source and community building
|
| • Tribute to Jim Weirich's impact on the Ruby community
|
| • Upcoming projects and availability for questions through Twitter or GitHub |
