tomrenner-com

The author argues that the term "full-stack developer" is misleading and unhelpful, especially for early-career professionals. They suggest "generalist" is a more accurate term that focuses on adaptable skills rather than implying expertise across all technology layers. The phrase shifts focus to technical gaps rather than highlighting broader development abilities.

The author explains their motivations for creating a personal website, citing narcissism, the desire to code more, and wanting to share opinions about software development. They built the site themselves using Ruby, Sinatra, jQuery, HTML, and CSS rather than using a site builder. The site will feature reviews, comments on software-related topics, and serve as a learning tool through writing.

The article argues that customer support should be viewed as valuable user testing rather than interruptions. Developers should recognize that users have expertise in their problem domains, while developers have software expertise. User feedback often reveals UX problems that need addressing.

The article discusses avoiding unproductive work by focusing on value over complexity. It recommends prioritizing tasks by their value to the team rather than personal interest or difficulty, and taking shortcuts by addressing only major requirements first.

Agile development emphasizes early feedback over upfront design but faces scope creep risks on large projects. Finding the right balance between minimal upfront design and enough structure to prevent major changes is challenging. The goal is to write code once, spend minimal time, and maximize future reusability while maintaining flexibility.

The author purchased a £190 Asus Chromebook C201 for lightweight, portable coding. After enabling developer mode and using Chromebrew to install development tools like Vim and Python, they found it works well for productive coding on the go despite some limitations with ChromeOS updates.

The article discusses how relying solely on human memory for knowledge is risky due to forgetfulness and the "bus factor" problem. It emphasizes creating systems to document information, allowing humans to focus on creative problem-solving while computers handle remembering. Effective systems should clearly show priorities, capture all incoming tasks quickly, and be always available.

The author reflects on how obsessing over time efficiency caused stress without delivering promised productivity gains. After five years of professional work, they've adopted a healthier approach that prioritizes work-life balance over constant optimization. They argue that most skill development happens at work, and reducing stress about minor time usage leads to better focus and learning.

The "Temple of Fail" is a weekly team retrospective practice where members share mistakes they made, discuss how to avoid them in the future, and foster a learning environment. The session is conducted in a light-hearted, no-blame manner to encourage open sharing and minimize repeated errors across the team.

The eXtreme Tuesday Club discussed Basecamp's Shape Up product development framework. Participants identified strengths like team ownership and reduced communication overhead, but raised concerns about fixed six-week cycles and potential team division. Several attendees expressed interest in adopting specific techniques like "fat pen" design and "betting" terminology.

The article describes using Gephi software to visualize Twitter network graphs, though the method no longer works since Twitter API changes. It provides a beginner's step-by-step guide for creating network visualizations from Twitter data, including downloading Gephi, installing the TwitterStreamer plugin, and configuring layout and appearance settings.

The article suggests creating a "Nice :-)" folder in your work inbox to save emails containing praise, thanks, or positivity. This simple practice creates a personal happy place that can provide joy during tough workdays and help counteract negative thoughts by preserving evidence of success.

The article presents a talk given at Codebar Festival 2021 about how staying in one place professionally doesn't necessarily mean stagnation. The speaker's slides are available on Slideshare for those who want to see the full presentation.

The article examines practical implementation of DORA metrics for DevOps performance monitoring. It details the raw data needed for each metric from existing tools like version control, bug tracking, and monitoring systems. The author emphasizes automated data extraction and clear team definitions for key terms.

Facebook's new Metaverse venture is discussed by Kit Wilson and software engineer Tom Renner, who explore what this digital reality might look like. The conversation examines the implications of this emerging technology.

The article discusses the software industry's focus on learning new technologies versus building institutional knowledge. It argues that while developers constantly learn new tools, the industry lacks historical analysis of past technologies to inform current practices. The author calls for more historical perspective in tech culture rather than just chasing new trends.

The Log4J vulnerability highlighted how dependencies can introduce significant security risks. Developers often import packages to save writing minimal code, adding thousands of lines of untested external code. The author proposes minimizing dependencies and requiring full justifications for any new package additions.

The article introduces the concept of "saying the quiet part out loud" as a team practice of explicitly stating the reasoning behind actions and decisions. This approach creates opportunities for alignment, disagreement, and clarification within teams, fostering open communication and trust.

The article examines "inevitabilism" - the belief that certain developments are unavoidable - as a framing technique used in technology debates. Tech leaders present AI as an inevitable future, shifting conversations from whether we want it to how we'll adapt. The author argues we have choices about our technological future.

The article discusses three pieces about technology's negative trends: the acceptance of mediocre quality in tech products, structural inequality in the tech industry, and individuals responsible for internet degradation. It highlights concerns about declining standards, diversity barriers, and accountability in tech.

The article argues that successful software development processes should prioritize building trust among team members. It examines how practices like code review, testing, team-building, and agile methodologies all serve to establish technical, architectural, interpersonal, and organizational trust. The appropriate approach depends on team context rather than following rigid methodologies.

The article discusses digital gardening as a content publishing approach, critiques the degradation of the web into corporate platforms, and examines Spotify's publication of AI-generated songs from deceased artists without permission. It explores personal publishing philosophies and ethical responsibilities in digital platforms.

The article critiques how large language models are anthropomorphized as human-like assistants, which excuses their unreliability and encourages emotional bonds with users. This framing conflicts with the tools' actual performance, as they become less accurate with more prompting despite conversational interfaces.

The article covers three topics: challenges with cycle time metrics in software development, applying learning theory to AI tool design, and security vulnerabilities in GitHub Actions and the Nix ecosystem. It discusses measurement pitfalls, better AI tool approaches, and supply chain attack risks.

The article argues that large language models represent a 400-year confidence trick, tracing from mechanical calculators to modern AI. It claims LLM vendors build trust through machine reliability, exploit emotions via fear and flattery, and create urgency about job obsolescence. The author contends that despite massive investment, most AI implementations fail to deliver promised returns.