The hottest Testing Substack posts right now

And their main takeaways
Category
Top Technology Topics

AI and the end of integration testing

Sunday Letters • 39 implied HN points • 17 Sep 23
🕹 Technology Testing
  1. AI will change how we build software, making traditional integration testing much harder or even impossible. This is because the possibilities for what software can do will become nearly infinite.
  2. Without effective integration testing, we will rely more on monitoring and telemetry to ensure software is working well in real time. Companies will need to focus on creating systems that check themselves while they run.
  3. New challenges will arise, like how to assess things like 'helpfulness' or 'safety' in software. We may need to create new measures and standards to evaluate these more complex qualities.

The Black Box "Problem"

Prompt Engineering • 39 implied HN points • 10 Jul 23
🕹 Technology Testing
  1. Our understanding of how AI works is limited due to the black box problem.
  2. Engineers build based on understanding, while scientists dissect to understand.
  3. It's more important for AI to work as intended and be continuously tested than to fully understand how it works.

Episode 33: Better Built By Burkhard

burkhardstubert • 99 implied HN points • 01 Jan 23
🕹 Technology Testing
  1. Test-Driven Development (TDD) helps developers get quick feedback while coding, improving overall project quality. This means fewer mistakes and less time spent fixing problems later.
  2. Using TDD can reduce the complexity of code by breaking down problems into smaller parts, making it easier to manage and understand.
  3. TDD encourages a culture of continuous improvement and teamwork, allowing all developers to take responsibility for the code they write. This leads to better collaboration and a more successful project.

Can the NRC be Reborn?

Gordian Knot News • 219 implied HN points • 20 May 23
🔬 Science Testing
  1. Reforming the NRC for nuclear power to succeed requires honesty and transparency in communication.
  2. A reformed NRC must establish a compensation scheme for harm caused by radioactive material exposure.
  3. The NRC's focus should shift to emission enforcement rather than reactor design to ensure safety and accountability.

sqlmesh test

davidj.substack • 47 implied HN points • 12 Dec 24
🕹 Technology Testing
  1. Unit tests and data tests are different. Unit tests check if a function works right with set inputs, while data tests check if the data meets certain conditions.
  2. Running tests locally can save costs and speed things up. If you test your code on your own machine, you don’t have to pay for the cloud data warehouse until you’re ready.
  3. Creating external models in sqlmesh can be automated, making it easier to document source tables. You just run a command to generate the necessary files instead of doing it manually.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Up to 103 tennis players failed doping tests and escaped bans

Honest Sport • 19 implied HN points • 31 Dec 23
🎾 Sports Testing
  1. Many tennis players have failed drug tests but faced no penalties. Out of 156 players who tested positive from 2013 to 2019, around 103 didn’t receive any sanctions.
  2. The International Tennis Federation has reduced the testing for blood doping significantly since 2018. This might have allowed players to avoid detection during important tournaments.
  3. The rules around doping cases can be confusing, as some players can continue playing even after testing positive for certain drugs, leading to a lack of transparency about who is really being punished.

Deterministic Simulation Testing - Antithesis

Why Now • 1 implied HN point • 20 Jan 26
🕹 Technology Testing
  1. Deterministic simulation testing runs your entire distributed system inside virtual machines controlled by a deterministic hypervisor so each test run is reproducible. It replaces wall-clock time with instruction-count-based virtual time so timing-dependent bugs can be replayed exactly.
  2. The platform combines property-based testing, fuzzing, and fault injection to automatically explore many scenarios and surface rare race conditions. All tests run in sandboxed clones of production so you can inject network blips and failures without risking real users.
  3. Determinism is achieved with techniques like single-core execution, intercepted time calls, and deterministic I/O plus numerous micro-optimizations. The outcome is precise, replayable failures that make debugging and fixing distributed-system bugs much easier.

Explaining Anomalous GCSE Results

Cremieux Recueil • 90 implied HN points • 21 Feb 24
🚌 Education Testing
  1. Some Black African students in the UK perform better on GCSE exams than on IQ tests, leading to a puzzle that is not easily explained by sampling differences.
  2. The discrepancy between GCSE performance and general intelligence indicates that GCSEs may be biased indicators of intelligence, favoring certain groups over others.
  3. Psychometric bias might explain why Black Africans in the UK excel in GCSEs despite IQ tests suggesting otherwise, highlighting the presence of biases in assessment methods.

How to stay motivated while learning Java?

Weekend Developer • 19 implied HN points • 06 Oct 23
🕹 Technology Testing
  1. When learning Java, start by writing code as if you're in a procedural language to ease the transition
  2. Writing unit test cases is important to become a seasoned Java programmer; it helps validate your code
  3. Refactor your code by breaking logic into simpler methods, identifying and grouping methods/variables into classes, and reviewing your code for improvements

How and Where to Get Your Testosterone Tested

Hard Pivot • 19 implied HN points • 17 May 23
🏥 Health & Wellness Testing
  1. Consider using InsideTracker for first blood test, Instalab if in NYC/SF, and Jason Health for T levels
  2. Start with comprehensive blood tests for cholesterol, thyroid health, etc., before focusing on testosterone levels
  3. Different providers offer various testing options for testosterone levels, including Total T, Free T, and Bioavailable T

Mocking OpenAI - Unit testing in the age of LLMs (Part One)

Laszlo’s Newsletter • 21 implied HN points • 23 Feb 25
🕹 Technology Testing
  1. Unit tests are still important even with LLMs. They help ensure your code behaves as expected, even when using unpredictable AI tools.
  2. Mocking is needed to effectively test code that relies on LLMs. Instead of calling the actual AI, you create a 'fake' version that simulates its behavior.
  3. Using libraries like 'respx' can simplify mocking in your tests, and it's essential to handle things like retry logic carefully to keep tests fast.

I Am A Sequencing Absolutist

ASeq Newsletter • 80 implied HN points • 06 Oct 23
🏥 Health & Wellness Testing
  1. The author strongly believes in the power and importance of DNA and RNA sequencing in diagnostics.
  2. Many clinicians tend to delay diagnostic tests unless absolutely necessary, relying on common symptoms and tests first.
  3. Despite the potential of advanced genetic testing, clinicians may be hesitant to order tests if there are no clear treatment options available.

Standardized usability questionnaires: post-task measures

UX Psychology • 59 implied HN points • 23 Dec 21
🕹 Technology Testing
  1. Post-task questionnaires capture user impressions immediately after a task, offering insights into user experience.
  2. Popular post-task questionnaires include the After-Scenario Questionnaire (ASQ), Subjective Mental Effort Question (SMEQ), and Expectation ratings (ER), each providing valuable usability feedback.
  3. When conducting usability studies, using multiple post-task questionnaires, like ASQ and SEQ, is recommended for a better understanding of user perceptions.

Big News! The Rapid Screen Pilots have arrived

Joshua Gans' Newsletter • 119 implied HN points • 18 Dec 20
🏥 Health & Wellness Testing
  1. The CDL Rapid Screening Consortium, led by Creative Destruction Lab and 12 companies, is implementing rapid antigen screens to enable daily screening to identify and isolate infectious individuals and protect others.
  2. The consortium aims to make rapid screening a part of daily life in Canada, with workers and visitors being frequently screened, helping to minimize exposure and bring economies back to normality.
  3. The initiative addresses the challenge of scaling pandemic management efforts, focusing on ongoing, regular screening to ensure a safer environment and eventually bringing back normalcy to daily life.

Think Like a Detection Engineer, Pt. 2: Rule Writing

Detection at Scale • 39 implied HN points • 25 Jul 22
🕹 Technology Testing
  1. Analyzing security data effectively involves identifying and flagging bad behaviors near high-risk assets.
  2. Writing rules based on observed attacker techniques and behaviors allows for a clear path to action in response to detected threats.
  3. Testing rules through phases like unit testing, backtesting, staging, and production helps refine and ensure alert accuracy before implementation.

Defining Mock Data Using OpenAPI Overlays

The API Changelog • 6 implied HN points • 18 Jul 25
🕹 Technology Testing
  1. Mock data is important for making good API documentation and for testing. You can use random data or tools like Faker to make it look realistic.
  2. Adding too much mock data can make your API documents very large. Using overlays lets you keep the original API document clean while still providing examples.
  3. Overlays can add, change, or remove information in your API definition without affecting the main document. This way, you can customize it for different needs without causing issues for users.

Software engineering practices you should probably be following in 2024

The Tech Enabler • 3 HN points • 27 Feb 24
🕹 Technology Testing
  1. Prefer statically typed languages over dynamically typed ones as they offer better reliability and help catch errors earlier.
  2. Utilize automatic code formatting tools for consistent code style and to save time on debating formatting during code reviews.
  3. Consider using queues as data sources in event-driven systems for better scalability and leverage infrastructure-as-code tools like Terraform for repeatable and manageable infrastructure.

045: Why You Should Manage Your Riboflavin Status and How to Do It

Harnessing the Power of Nutrients • 139 implied HN points • 16 Feb 19
🏥 Health & Wellness Testing
  1. Riboflavin is crucial for various bodily functions, including energy metabolism, antioxidant system, and methylation.
  2. Assessing riboflavin status accurately is essential; riboflavin supplements may vary in effectiveness based on individual health conditions.
  3. Consuming dietary sources rich in riboflavin, like liver and milk, can help maintain optimal riboflavin status.

Complacency

Joshua Gans' Newsletter • 59 implied HN points • 05 Oct 20
🏥 Health & Wellness Testing
  1. Complacency during a crisis can lead us to let our guard down, requiring constant adherence to protocols to manage the situation.
  2. Strict protocols are necessary, especially in high-risk environments like the White House, and must be followed diligently to prevent breakdowns in the system.
  3. The ongoing COVID-19 crisis demands vigilance from all of us - we need to innovate ways to maintain strict screening protocols and not let our guard down.

Refactoring BabyAGI - Code Quality and LLMs

Laszlo’s Newsletter • 32 implied HN points • 28 Apr 23
🕹 Technology Testing
  1. Refactoring an early implementation of autonomous agents based on LLMs with clean architecture principles
  2. When analyzing a legacy codebase, focus on finding the main entry point and understanding variable usage
  3. Consider moving external dependencies into their own classes and introducing a 'Task' class to improve code structure

dbt Reimagined

Pedram's Data Based • 33 implied HN points • 09 Mar 23
🕹 Technology Testing
  1. Utilizing Domain-Specific Languages can enhance data modeling with features like type safety and validation.
  2. Implementing debuggers in SQL modeling can help understand the system's state and improve troubleshooting.
  3. Enabling unit testing in dbt with alternative adapters could simplify validation of logic.

Scaling ain't cheap or easy

Joshua Gans' Newsletter • 39 implied HN points • 27 Nov 20
🏥 Health & Wellness Testing
  1. Scaling up Covid-19 testing is crucial for returning to normalcy before widespread vaccine distribution.
  2. Implementing large-scale testing efforts requires a coordinated push similar to military operations and campus-wide testing strategies.
  3. Preparing for economy-wide frequent testing demands meticulous planning, infrastructure development, and preparedness on a national level.

Large Language Model Observability

Monitoring Monitoring • 3 HN points • 04 Apr 23
🕹 Technology Testing
  1. Startups are focusing on solving observability challenges for teams using Large Language Models (LLMs) like GPT-4.
  2. LLM-based applications involve sending prompts in English to an API, raising questions about prompt quality, speed optimization, and cost management.
  3. Emerging startups are exploring automating generative testing and incident response using AI models like GitHub's Copilot.

The New Hotness

Dan’s MEGA65 Digest • 11 implied HN points • 16 Jan 24
🕹 Technology Testing
  1. Next batch of MEGA65 computers to ship on June 1st, 2024.
  2. v0.96 release candidate available for testing with new features and bug fixes.
  3. Mention of collaboration with The 8-Bit Guy in making a video about MEGA65.

B.1.1.7

Joshua Gans' Newsletter • 19 implied HN points • 11 Jan 21
🏥 Health & Wellness Testing
  1. The B.1.1.7 variant of SARS-CoV-2 is 50% more transmissible, likely spreading rapidly globally under several names, including 'UK Variant' and 'Supercovid.'
  2. B.1.1.7 mutations mainly affect the spike protein, raising concerns about vaccine efficacy. Current vaccines may still be effective, but wide distribution could lead to the virus evolving to evade immunity.
  3. B.1.1.7 is outcompeting other variants due to increased transmissibility rather than higher virulence, emphasizing the importance of maintaining mitigation efforts, particularly in high-risk settings, and ramping up testing to contain the spread.

Worry Diary

Joshua Gans' Newsletter • 19 implied HN points • 17 Dec 20
🏥 Health & Wellness Testing
  1. The lack of information solutions during a family Covid-19 scare caused days of anxiety and uncertainty.
  2. Rapid testing and comprehensive information sharing could have greatly reduced worry and uncertainty during the crisis.
  3. Having access to quick and accurate testing results, along with better information systems, could have provided clarity and peace of mind in the situation.

New Fast Tests and Distress from Regulators

Joshua Gans' Newsletter • 19 implied HN points • 28 Aug 20
🏥 Health & Wellness Testing
  1. Widespread frequent testing is crucial to quickly deal with Covid-19 and reduce economic pain. Challenges include the need for cheap, rapid tests and regulators changing their approach for pandemic mitigation.
  2. The new Abbott test is a $5 fast test authorized by the FDA, but is limited to use on symptomatic individuals. The test requires administration by healthcare providers, impacting its potential impact.
  3. Regulations can be changed to expand the use of tests like Abbott's which may be more valuable for identifying infectious individuals. The decision to limit the test to symptomatic use may hinder efforts to stop the virus spread.

How "new type" helps avoid production outages

Blog System/5 • 4 HN points • 09 Mar 24
🕹 Technology Testing
  1. The 'new type' idiom in programming involves wrapping native types in domain-specific types to prevent errors when passing values to external APIs.
  2. Avoid using primitive types like integers or strings directly in your code base by creating new types like `Username` and `Size` for better code structure.
  3. Adopting the 'new type' pattern can help prevent production outages by catching issues like function argument swaps early on and enforcing correct usage through type safety.

027: Why You Should Manage Your Selenium Status and How to Do It

Harnessing the Power of Nutrients • 39 implied HN points • 18 Feb 17
🏥 Health & Wellness Testing
  1. Selenium is crucial for combating oxidative stress and thyroid function but can vary in food, making it important to manage our selenium levels.
  2. Key markers like plasma selenium and glutathione peroxidase activity help assess selenium status, with specific recommendations for supplementation.
  3. Incorporating selenium-rich foods like organ meats and seafood can be a reliable way to improve selenium levels, with insight on supplements and dosages.

Stubs in Maineffectjs

Arvind’s Substack • 1 HN point • 17 May 23
🕹 Technology Testing
  1. In Maineffectjs, stub dependencies faster to test specific functionalities.
  2. The new 'stub' function in Maineffectjs automatically creates mocks for you, reducing complexity.
  3. Maineffectjs allows for flexible stub implementations, such as Jest or Sinon, to enhance testing capabilities.

Feature flags are ruining your codebase

Leading Developers • 3 HN points • 05 Mar 24
🕹 Technology Testing
  1. Feature flags can make codebases more complex and harder to maintain, especially when used as an excuse to avoid making hard decisions like completely removing a feature.
  2. Having too many feature flags can lead to wasted time on dead code, increased testing burden, and making testing a substitute for fixing issues.
  3. Different types of feature flags, like release toggles, experiment toggles, and permission toggles, require specific management approaches to prevent the codebase from becoming unmanageable.

Tests for Vitamin K Status: A Closer Look

Harnessing the Power of Nutrients • 59 implied HN points • 28 Apr 13
🏥 Health & Wellness Testing
  1. Blood levels of vitamin K mainly reflect recent intake rather than long-term nutritional status. Testing the carboxylation status of vitamin K-dependent proteins is a more accurate way to assess vitamin K status.
  2. Clotting disorders are a significant concern, so tests for vitamin K-dependent clotting factors are well-established. These tests prioritize clotting function over other aspects of vitamin K's role.
  3. The most common marker for vitamin K status in research studies is the carboxylation status of osteocalcin, reflecting vitamin K status in bone. A promising future marker is desphospho-uncarboxylated matrix Gla protein for assessing vascular health.

[Guest post] Edition 24: Pentesting LLM apps 101

Boring AppSec • 3 HN points • 13 Oct 23
🕹 Technology Testing
  1. Pentesters should care about security implications of integrating LLMs in applications.
  2. Identifying LLM usage in applications can involve looking for client-side SDKs, server-side APIs, and popular adoption signs.
  3. Assessing LLM-integrated applications requires manual testing, tooling like Garak and LLM Fuzzer, and aiding developers in defending against vulnerabilities.

Dependency Injection Demystified

Jacob’s Tech Tavern • 3 HN points • 06 Jun 23
🕹 Technology Testing
  1. Unit testing helps in writing maintainable code by separating concerns and breaking code into manageable chunks.
  2. Modern language features like async/await and functional reactive programming provide great coding ergonomics but require careful testing to avoid flakiness.
  3. Dependency Injection separates the tasks of gathering ingredients and cooking, making code more testable and maintainable.