Issue #450

22 September 2023


Issue #450
pointer.io


Tuesday 22nd September’s issue is presented by FusionAuth

Don’t Build Your Own Auth. Try FusionAuth Today.

FusionAuth integrates with any tech stack and is deployable anywhere – cloud, on-premise, or even the server running under your desk. You're less than five minutes from login/registration, social logins, SSO, MFA, passwordless, user management, passkeys, and much more.


Join the thousands of developers who trust FusionAuth for their identity needs.

Elevate Your Performance Review Conversations With These 12 Expert Tips


tl;dr: (1) Hone your narrative with a personal press release. (2) Managers try this template to structure your next review conversation. (3) Zoom out to get a fuller picture. (4) Spend more time than you think you need to with your high performers. (5) Differentiate between interpersonal and performance feedback. (6) Ask this question to give feedback that lands - “Does this feedback resonate with you? Why or why not?”


Leadership Management

Don’t Cross The Beams

— Kent Beck


tl;dr: Kent discusses the nuances of code refactoring, emphasizing the difference between "horizontal" and "vertical" refactorings. "Thinking of refactorings as horizontal and vertical is a heuristic for turning this situation around – eliminating risk quickly and exploiting proven opportunities efficiently." Kent also introduces the concept of "Succession," which deals with breaking down the refactoring process into manageable, efficient steps.

CareerAdvice

Common Authentication Implementation Risks And How To Mitigate Them

— James Hickey


tl;dr: Data breaches are more common than ever. Ensuring a secure authentication system is critical to your trust with customers. Whether you build or buy your auth solution, this article offers insights into secure practices that can help keep you and your customers safe.


Promoted by FusionAuth

Management BestPractices

LLMs Demand Observability-Driven Development

— Charity Majors


tl;dr: “Many software engineers are encountering LLMs for the very first time, while many ML engineers are being exposed directly to production systems for the very first time. Both types of engineers are finding themselves plunged into a disorienting new world—one where a particular flavor of production problem they may have encountered occasionally in their careers is now front and center. Namely, that LLMs are black boxes that produce nondeterministic outputs and cannot be debugged or tested using traditional software engineering techniques. Hooking these black boxes up to production introduces reliability and predictability problems that can be terrifying.“ Charity believes that the integration of LLMs will necessitate a shift in development practices, particularly towards Observability-Driven Development, to handle the nondeterministic nature of these models.

 

Management LLM AI

“Computers are good at following instructions, but not at reading your mind.”


— Donald Knuth

How Microsoft Does Quality Assurance (QA)

— Gergely Orosz


tl;dr: Microsoft's approach to Quality Assurance (QA), focusing on the Software Development Engineer in Test (SDET) role. The SDET role was designed for engineers who focused on writing automated tests and maintaining testing systems. "An SDET is a developer who works in a test team and not a development team." This role was retired around 2014, as Microsoft moved towards a more integrated approach where all software engineers became responsible for testing their code.


Management

Selective Column Reduction For DataLake Storage Cost Efficiency


tl;dr: “As Uber continues to grow, the sheer volume of data we handle and the associated demands for data access have multiplied. This rapid expansion in data size has given rise to escalating costs in terms of storage and compute resources. Consequently, we have encountered various challenges such as increased hardware requirements, heightened resource consumption, and potential performance issues like out-of-memory errors and prolonged garbage collection pauses.”


Performance

SQL Join Flavors

— Anton Zhiyanov


tl;dr: Anthon provides a look at SQL joins, covering types like "Qualified Join," "Natural Join," "Cross Join," "Partitioned Join," and "Lateral Join." He explains that a "Qualified join" is an umbrella term for the most common types of joins: inner, left, right, and full. "A qualified join connects records from two datasets into one, according to the matching criteria you specify," he states. The article also warns against the use of "Natural Joins," stating they are "almost always a bad idea."


SQL

When Zig Outshines Rust - Memory Efficient Enum Arrays

— Adrian Alic


tl;dr: Adrian discusses the issue of memory fragmentation in Rust, particularly when using Enums with different-sized variants explaining that one of the biggest motivators for efficient Enum arrays has been compilers, as "Big ASTs can incur a hefty performance penalty during compilation." The article suggests using a "struct-of-arrays" (SoA) approach or creating one vector per variant, termed as "array of variant arrays" (AoVA), to mitigate fragmentation.


Zig Rust

Notable Links


AgentVerse: Flexible framework for multi-agent environments.


Depcheck: tool for analyzing the dependencies.


Instagraph: Convert text input or URL into knowledge graph.


Kamal: Deploy web apps anywhere.


OpenTofu: Manage your cloud infrastructure.


Click the below and shoot me an email!


1 = Didn't enjoy it all // 5 = Really enjoyed it


12345