Home

The 12 Factor App manifesto is a methodology for building software-as-a-service (SaaS) apps that are scalable, maintainable, and deployable on modern cloud platforms. Developed by engineers at Heroku, it's a set of best practices designed to enable applications to be built with portability and resiliency when deployed to the web.

Introduction to the 12 Factor App

In the early days of web development, applications were often built in a monolithic style, tightly coupled to their execution environment. This approach led to numerous issues, especially when apps needed to scale or move to different environments. The 12 Factor App methodology was created to address these challenges, emphasizing a declarative format for setup automation, clean contract with the operating system, and minimizing divergence between development and production.

The Twelve Factors

Codebase: One codebase tracked in version control, many deploys. This principle advocates for a single codebase for each service which can be deployed in any environment.
Dependencies: Explicitly declare and isolate dependencies. Applications should explicitly declare all dependencies, not relying on the implicit existence of system-wide packages.
Config: Store configuration in the environment. This factor pushes for the separation of config from code, as configurations vary substantially across deploys while code does not.
Backing Services: Treat backing services as attached resources. This means that a deploy of the app should be able to swap out a local MySQL database for a third-party service without changing the app’s code.
Build, Release, Run: Strictly separate build and run stages. This factor emphasizes the need for strictly separating the build stage (where the code is converted into an executable bundle), the release stage (where the executable is combined with the config), and the run stage (where the app is actually run).
Processes: Execute the app as one or more stateless processes. This principle states that the app should be executed in the stateless fashion, and any data that needs to persist should be stored in a stateful backing service, like a database.
Port Binding: Export services via port binding. Apps should be completely self-contained and should not rely on runtime injection of a webserver into the execution environment to create a web-facing service.
Concurrency: Scale out via the process model. This factor implies that apps should be able to scale horizontally by adding more concurrent processes.
Disposability: Maximize robustness with fast startup and graceful shutdown. This principle advocates for short startup times and graceful shutdowns to maximize robustness.
Dev/Prod Parity: Keep development, staging, and production as similar as possible. This factor aims to reduce the gaps between development and production to ensure continuous deployment for maximum agility.
Logs: Treat logs as event streams. Applications should not concern themselves with routing or storage of their output stream. Instead, each running process writes its event stream, unbuffered, to stdout.
Admin Processes: Run admin/management tasks as one-off processes. This principle states that management tasks should be run in an environment identical to the regular long-running processes of the app.

Conclusion

The 12 Factor App methodology provides a framework for building software that demonstrates key characteristics necessary for modern cloud platforms. It addresses issues of scalability, reliability, and portability, making it easier for development teams to build and manage applications effectively. As the world of web development evolves, the principles of the 12 Factor App continue to be a vital reference for building robust, scalable cloud-native applications.

January 9, 2024
in podcast
3 min read

Mastering Cloud-Native Applications - A Comprehensive Guide to the 12 Factor App Manifesto

Welcome to Continuous Improvement, the podcast where we explore cutting-edge methodologies and best practices that enhance how we build software. I’m your host, Victor Leung, and in today’s episode, we’re unpacking a crucial methodology for anyone working in the realm of Software-as-a-Service or SaaS—The 12 Factor App. Developed by the engineers at Heroku, this set of guidelines has transformed how applications are built and deployed on modern cloud platforms.

To start, let's talk about why the 12 Factor App methodology was created. In the early days, web applications were often built as monoliths—large, single units that were tightly coupled with their environments. This created a myriad of problems, especially when these applications needed to scale or be moved to different environments.

The 12 Factor App methodology was designed to overcome these challenges. It emphasizes a declarative format for automation setup, a clean contract with the operating system, and a drive to minimize the gap between development and production. These practices ensure that applications are portable, scalable, and maintainable. Let’s break down these factors.

Factor 1: Codebase. There should be exactly one codebase for a service, with the codebase being used for many deployments. This means having a single repository that can be deployed anywhere, from the developer's local environment to the final production servers.

Factor 2: Dependencies. Your application should explicitly declare and isolate dependencies. This avoids the pitfalls of implicit dependencies that can lead to conflicts between differing environments.

Factor 3: Config. Configuration settings should be stored in the environment rather than in the code. This separation of config from code helps keep the application environment agnostic, making it easy to adjust settings without changing the codebase.

Factor 4: Backing Services. Treat all backing services as attached resources which can be attached or detached to deployments without significant changes to the code.

Factor 5: Build, Release, Run. Strictly separate the build and run stages. By doing this, applications are more stable and predictable since no changes are allowed that might affect the running application during the release phase.

Factor 6: Processes. Execute the app as one or more stateless processes. This is crucial for scalability and distribution since no user session or context is stored locally.

Factor 7: Port Binding. The application should be self-contained and not rely on runtime injection of a web server.

Factor 8: Concurrency. Scale out via the process model. Essentially, this involves starting more copies of the application to handle more load.

Factor 9: Disposability. Maximize robustness with fast startup and graceful shutdown. This improves the application’s reliability in the face of hardware/software failures.

Factor 10: Dev/Prod Parity. Keep development, staging, and production as similar as possible to avoid bugs that arise from environmental differences.

Factor 11: Logs. Treat logs as event streams. This allows for more scalable and manageable logging.

Factor 12: Admin Processes. Run administrative/management tasks as one-off processes. This ensures that these tasks are performed without affecting the running application’s environment.

The 12 Factor App is more than just a methodology; it’s a philosophy for software development and deployment. It’s about building software that is robust, manageable, and above all, adaptable to the changes and scales as needed without a complete overhaul.

Thanks for joining me on Continuous Improvement. Whether you’re building your next big SaaS or scaling an existing one, keeping these 12 factors in mind can be the difference between success and failure. I'm Victor Leung, and I’ll be back soon with more insights and strategies to help you refine your craft and improve your projects. Until next time, keep coding, keep improving, and stay cloud-savvy.

January 9, 2024
in zh
1 min read

掌握雲原生應用程式 - 12因子應用程式宣言的全面指南

12因子應用程式宣言是一種構建軟件即服務（SaaS）應用程式的方法，這些應用程式具有可擴展性，可維護性，並且可在現代雲平台上部署。這套方法是由Heroku的工程師開發的，旨在使應用程式具有部署到web時的可攜性和彈性。

介紹12因子應用程式

在網頁開發的早期，應用程式經常以單體風格構建，與其執行環境緊密結合。這種方式導致了許多問題，尤其是當應用程式需要擴展或遷移到不同環境時。12因子應用程式的方法論就是為了解決這些挑戰，強調以聲明的方式進行設置自動化，與操作系統保持良好的合同，並使開發與生產之間的差異最小化。

十二因子

程式碼庫：一個在版本控制中追蹤的程式碼庫，許多部署。這條原則主張每個服務都應有一個程式碼庫，該程式碼庫可以在任何環境中部署。
依賴性：明確宣告並隔離依賴性。應用程式應明確宣告所有依賴性，不依賴於系統範疇內包的隱含存在。
配置：在環境中儲存配置。這個因子推動將配置與程式碼分離，因為配置在部署間變化很大，而程式碼則不是。
支援服務：將支援服務視為附加資源。這意味著應用程式的部署應能夠換出本地 MySQL 資料庫為第三方服務，而不用修改應用程式的程式碼。
建立，發佈，運行：嚴格分離建立和運行階段。這個因子強調需要嚴格分離建立階段（將程式碼轉換成可執行包的階段），發佈階段（將可執行文件與配置結合的階段）和運行階段（實際運行應用程式的階段）。
進程：作為一個或多個無狀態進程來執行應用程式。這個原則認為，應用程式應該以無狀態的方式執行，並且任何需要保留的資料都應該儲存在有狀態的支援服務，如資料庫。
端口綁定：通過端口綁定導出服務。應用程式應完全自包含，不應依賴於服務器在運行環境中的注入來創建面向web的服務。
併發：通過進程模型進行擴展。該因子意味著應用程式應能通過增加更多並發進程來實現水平擴展。
廢棄：通過快速啟動和優雅關閉來最大化魯棒性。這個原則主張縮短啟動時間並優雅地關機，以最大程度地提高魯棒性。
開發/生產同態：保持開發，暫存和生產盡可能相似。這個因子的目標是減少開發與生產之間的差距，確保連續部署以達到最大的敏捷性。
日誌：將日誌視為事件流。應用程式不應對其輸出流的路由或儲存感到憂慮。相反，每個正在運行的進程都將其事件流無緩存地寫入stdout。
管理流程：作為一次性流程運行管理/管理任務。這個原則認為，應該在與應用程式的常規長期運行進程相同的環境中運行管理任務。

結論

12因子應用程式的方法提供了一個構建軟體的框架，該軟體顯示了現代雲平台所必需的關鍵特性。它解決了可擴展性，可靠性和可攜性的問題，使開發團隊更容易有效地構建和管理應用程式。隨著網頁開發的世界不斷演進，12因子應用程式的原則繼續是構建強大、可擴展的雲原生應用程式的重要參考。

January 8, 2024
2 min read

Nudges - A Gentle Push Towards Better Choices

In the world of behavioral economics, the term "nudge" has become a powerful concept in influencing the decision-making processes of individuals. Coined by Richard Thaler and Cass Sunstein in their influential book "Nudge: Improving Decisions About Health, Wealth, and Happiness," a nudge is defined as any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives.

Types of Nudges

Default Nudges: These are based on setting a default choice which is more likely to be chosen because of inertia. For example, automatically enrolling employees in a retirement savings plan but giving them the option to opt out.
Social Norm Nudges: These nudges use the power of social influence. By showing that a particular behavior is the norm, individuals are more likely to conform. An instance of this is hotels indicating that most guests reuse their towels, encouraging others to do the same.
Simplification Nudges: These are based on the principle that if a choice is easier to understand, it's more likely to be chosen. For instance, simplifying the paperwork for financial aid can increase college enrollment.
Salience Nudges: These nudges make certain options more prominent or visible. An example is placing healthier foods at eye level in a cafeteria to promote better eating habits.
Feedback Nudges: Providing feedback about behavior can influence future decisions. For example, a monthly report on electricity usage compared to neighbors nudges people to reduce their consumption.

Examples of Nudges in Action

Organ Donation: In countries where organ donation is an 'opt-out' system (default nudge), there's a higher rate of organ donors compared to countries where you have to 'opt-in'.
Healthy Eating: In school cafeterias, placing fruits and vegetables at the beginning of the serving line (salience nudge) has been shown to increase their consumption among students.
Financial Decisions: Apps that round up purchases to the nearest dollar and save the difference (simplification nudge) make it easier for people to save money without feeling a significant impact on their finances.
Environmental Conservation: Showing a building’s energy consumption in real-time compared to other similar buildings (feedback nudge) can motivate reductions in energy use.
Public Health: During the COVID-19 pandemic, social norm nudges were used by displaying signs about mask-wearing being a common practice, thereby encouraging more people to wear masks.

Conclusion

Nudges represent a subtle yet powerful way to influence human behavior positively. By understanding how different types of nudges work, policymakers, businesses, and even individuals can create an environment where making the best choice becomes the easiest option. It's a testament to the power of gentle persuasion over forceful compulsion.

January 8, 2024
in podcast
3 min read

Nudges - A Gentle Push Towards Better Choices

Welcome back to Continuous Improvement, where we explore ideas that enhance our lives and reshape our behavior. I'm your host, Victor Leung, and today, we're diving into a fascinating concept from behavioral economics known as the "nudge." Popularized by Richard Thaler and Cass Sunstein in their book "Nudge: Improving Decisions About Health, Wealth, and Happiness," nudging has transformed how we think about influencing decision-making in subtle yet effective ways.

A nudge is essentially a feature of the environment that alters people’s behavior in a predictable way, without forbidding any options or significantly changing their economic incentives. It's about making it easier for people to make certain decisions that can benefit them and society as a whole.

Let’s break down some types of nudges that have proven particularly effective:

First up, Default Nudges. These are choices made easy simply by what's pre-selected for us. Think about a workplace that automatically enrolls its employees into a retirement savings plan, but gives them the option to opt out. This type of nudge takes advantage of our tendency to stick with the default setting, boosting positive outcomes like increased savings for retirement.

Next, we have Social Norm Nudges. These are all about the power of the crowd. When hotels tell you that most guests reuse their towels to save water, they're nudging you to do the same by highlighting what others are already doing.

Then there's Simplification Nudges. By making processes simpler, these nudges help people make better choices more easily. A great example is streamlining the paperwork for financial aid to help more students enroll in college.

Salience Nudges make the preferred choices more visible. For instance, placing healthier foods at eye level in a cafeteria can nudge people towards better eating habits without removing the less healthy options.

And Feedback Nudges involve giving people information about their behavior to encourage smarter decisions in the future. A monthly report showing your electricity usage compared to your neighbors can motivate you to cut back on energy consumption.

Let's look at nudges in action. From organ donation to healthy eating in schools, financial decisions, and even during the COVID-19 pandemic with mask-wearing—nudges have been applied in various areas to encourage beneficial behaviors subtly.

In organ donation, countries with an 'opt-out' system see higher donor rates than those requiring an explicit 'opt-in'. Simply by changing the default, these countries have significantly increased the availability of organs for lifesaving transplants.

In schools, placing fruits and vegetables at the start of the line nudges students towards making healthier choices. Financial apps that round up your purchases to the nearest dollar and save the difference are simplifying the act of saving money, making it feel less like a sacrifice and more like a seamless part of everyday spending.

Nudges show us that the best choice doesn't have to be the hard choice. By designing our environments to promote better decisions effortlessly, we can all benefit from the subtle powers of influence that nudges provide.

That's all for today's episode of Continuous Improvement. I hope you've gained insights into how simple changes in our choice architecture can lead to significant improvements in our behavior and decision-making. I'm Victor Leung, and I look forward to nudging you towards more fascinating topics in our next episode. Until then, keep thinking, keep improving, and let's all make the easy choices the good choices.

January 8, 2024
in zh
1 min read

引導 - 朝著更好的選擇輕輕推進

在行為經濟學的世界中，"引導"一詞已成為影響個體決策過程的一種有力概念。這一詞由Richard Thaler和Cass Sunstein在他們的影響力極大的書籍"引導:Nudge: Improving Decisions About Health, Wealth, and Happiness"The book"提出，引導被定義為改變人們行為的任何選擇架構的方面，這種改變是可預見的，而不讓人們禁止任何選項或大幅改變他們的經濟激勵。

各類型的引導

預設引導: 這些基於設置一個預設選擇的引導，由於慣性，這個選擇更有可能被選擇。例如，將員工自動納入退休儲蓄計劃，但給予他們退出的選擇。
社會規範引導: 這些引導利用社會影響的力量。通過顯示某種行為是常規，個人更有可能遵從。一個例子是，酒店指出大多數客人重複使用他們的毛巾，鼓勵其他人也這樣做。
簡化引導: 這些基於一個原則，即如果一個選擇更易於理解，則更有可能被選擇。例如，簡化資助金的申請表格可以增加大學報名率。
顯著引導: 這些引導使某些選項更為突出或可見。一個例子是在飯堂將更健康的食物放在視線高度以鼓勵更好的飲食習慣。
反饋引導: 提供有關行為的反饋可以影響未來的決策。例如，與鄰居相比，每月對電力使用量的報告引導人們減少他們的消耗。

引導在行動中的範例

器官捐贈: 在將器官捐贈設定為'退出系統' (預設引導)的國家，與需要'選擇加入'的國家相比，有更高的器官捐贈者比例。
健康飲食: 在學校飯堂，將水果和蔬菜放在供應線的開頭 (顯著引導)被證實可以增加學生的消耗。
財務決策: 將消費額進位到最接近的整數並儲存差額 (簡化引導)的應用軟件，讓人們在不感到財務壓力的情況下更容易儲蓄。
環境保護: 將一個建築物的能源消耗與其他類似建築物的實時比較顯示出來 (反饋引導)可以激勵減少能源使用。
公眾衛生: 在COVID-19大流行期間，社會規範引導被用來展示佩戴口罩是一種常見的做法的標示，從而鼓勵更多的人佩戴口罩。

結論

引導代表了一種微妙而又強大的影響人類行為的方式。通過理解各種引導的工作方式，政策制定者、商業和個人可以創建一個環境，使得做出最好的選擇成為最容易的選項。這證明了溫和的說服力超越了強制的強迫性。

January 7, 2024
3 min read

Unraveling the Mind - A Deep Dive into Cognitive Biases and How They Shape Our Decisions

Cognitive biases are fascinating and often misunderstood aspects of human psychology. In this blog post, we'll explore what cognitive biases are, why they occur, and some common examples that impact our daily lives.

Understanding Cognitive Biases

Cognitive biases are systematic patterns of deviation from norm or rationality in judgment. They arise from the way our brain processes information. These biases are often a result of our brain's attempt to simplify information processing. They can be the result of heuristics or mental shortcuts that we use to make decisions quickly.

Why Do Cognitive Biases Occur?

Cognitive biases are thought to be a form of mental shortcut, often based upon social norms and generalizations. They occur because our brains are trying to be more efficient. In a world where we are bombarded with information, our brains try to streamline the processing by using past experiences and perceptions to make quick decisions.

Common Cognitive Biases

Confirmation Bias: This is the tendency to search for, interpret, favor, and recall information in a way that confirms one's preexisting beliefs or hypotheses. For instance, if you believe that left-handed people are more creative, you're more likely to notice and remember examples that support this belief and ignore examples that don't.
Anchoring Bias: This bias occurs when individuals rely too heavily on an initial piece of information (the "anchor") when making decisions. For example, if the first car you see at a dealership is priced at $30,000, you might anchor to that price, making any car priced less than that seem like a good deal, regardless of its actual value.
Availability Heuristic: This is a mental shortcut that relies on immediate examples that come to a person's mind when evaluating a specific topic, concept, method, or decision. For example, fearing plane crashes more than car accidents because plane crashes are more dramatic and receive more media coverage, even though car accidents are statistically more common.
Bandwagon Effect: The tendency to do (or believe) things because many other people do (or believe) the same. This is evident in various aspects of social life, like fashion trends, sports fandom, or political opinions.
Dunning-Kruger Effect: This is a cognitive bias in which people with low ability at a task overestimate their ability. It’s seen in various areas of life where people believe they are more knowledgeable or competent in something than they really are.

Overcoming Cognitive Biases

Overcoming cognitive biases is not easy, but awareness is the first step. Being mindful of these biases and actively questioning our own thoughts and decisions can help reduce their impact. It's also helpful to seek out different perspectives and information that challenges our preconceptions.

In conclusion, cognitive biases are deeply ingrained in our thought processes and can significantly affect our decision-making and beliefs. Understanding them is crucial for both personal development and for making more informed and rational decisions. Remember, the first step towards change is awareness, and the journey towards unbiased thinking, while challenging, is incredibly rewarding.

January 7, 2024
in podcast
2 min read

Unraveling the Mind - A Deep Dive into Cognitive Biases and How They Shape Our Decisions

Welcome to Continuous Improvement, where we explore ways to enhance our thinking and decision-making in both personal and professional life. I'm your host, Victor Leung, and today we’re delving into a topic that touches all aspects of our lives: cognitive biases. These are the little tricks our minds play on us, influencing our decisions and judgments in ways we often don't even realize.

Cognitive biases are systematic patterns of deviation from norm or rational judgment, and they arise from the way our brains process information. Basically, our brains are trying to be efficient by using shortcuts to quickly make sense of the world around us. But sometimes, these shortcuts can lead us astray.

So why do these biases occur? It's all about efficiency. In a world overflowing with information, our brains use past experiences and generalizations to make quick decisions. While this can save us time, it can also lead to errors in judgment.

Let’s talk about some common cognitive biases that might be affecting your daily decisions without you even knowing it.

First, we have Confirmation Bias. This is when you favor information that confirms your existing beliefs. For example, if you think that early risers are more productive, you're more likely to notice articles and success stories that validate this belief, while disregarding evidence to the contrary.

Then there’s Anchoring Bias. Say you’re buying a car and the first price you see is $30,000. That number sets your expectations, and any price below that seems like a steal—even if it's not the best you could do.

Next is the Availability Heuristic. This bias tricks you into overestimating the importance of information that's readily available. For instance, if you’ve just watched a news report on an airplane crash, you might suddenly feel like flying is too dangerous, despite it being one of the safest modes of transportation.

There's also the Bandwagon Effect—the reason fashions and fads spread so quickly. If everyone else is doing it, we feel a strong pull to join in, whether it’s starting the newest diet or supporting a popular opinion.

And let’s not forget the Dunning-Kruger Effect, where people with limited knowledge in an area overestimate their own expertise. It's why someone who's just read a few articles might claim to be an expert in a subject.

Overcoming these biases isn’t easy, but becoming aware of them is the first step. By questioning our own decisions and seeking out diverse perspectives, we can begin to counteract the effects of these mental shortcuts.

To wrap up, understanding cognitive biases is key to personal growth and making informed decisions. It's a challenging journey to think more clearly and rationally, but it’s also incredibly rewarding.

Thanks for tuning into Continuous Improvement. I'm Victor Leung, reminding you to think critically, question often, and improve continuously. Join me next time for more insights into how we can all live and work smarter. Until then, stay sharp and stay aware.

January 7, 2024
in zh
1 min read

揭開心智之謎 - 深入探討認知偏見以及它們如何塑造我們的決策

認知偏見是人類心理學中令人著迷且常常被誤解的一個面向。在此篇博客文章中，我們將探討認知偏見是什麼、它們為什麼會發生，以及一些常見的影響我們日常生活的例子。

理解認知偏見

認知偏見是從常規或理性判斷中偏離的系統性模式。它們源於我們的大腦如何處理信息。這些偏見往往是大腦試圖簡化信息處理的結果。他們可能是啟示或我們用來快速做出決策的心理捷徑的結果。

認知偏見為何會發生？

認知偏見被認為是一種精神捷徑，通常基於社會規範和概括。他們發生是因為我們的大腦正在努力提高效率。在一個充斥著資訊的世界裡，我們的大腦試圖通過使用過去的經驗和感知來快速做出決策。

常見的認知偏見

確認偏見：這是尋找、解釋、偏愛和回憶信息以確認一個人先前存在的信念或假設的傾向。例如，如果你相信左撇子更有創意，那麼你更可能注意到並記住支持這種信念的例子，並忽略不支援的例子。
錨定偏見：這種偏見是當人們在做決定時過於依賴最初的一個信息（"錨"）。例如，如果你在汽車經銷商的第一部車的價格為30,000美元，你可能會把那個價格當作錨，讓任何價格低於此的車看起來都像是一個好交易，不管它的實際價值如何。
可得性啟示：這是一種依賴於人們在評價特定主題、概念、方法或決策時能立即想到的例子的心理捷徑。例如，因為飛機失事更具戲劇性且獲得更多媒體報導，即使車禍在統計上更常見，人們對飛機失事的恐懼已超過對車禍的恐懼。
從眾效應：很多人做或相信的事情，就有傾向去做（或相信）相同的事情。這種現象在社會生活的各個方面都很明顯，比如時尚潮流、體育迷或政治觀點。
鄧寧-克魯格效應：這是一種認知偏見，即在一項任務中能力較低的人高估了自己的能力。在生活的各個領域我們都能看到這種情況，人們常常認為自己在某件事情上比實際上更知識淵博或更有能力。

克服認知偏見

克服認知偏見並不容易，但是意識到它們是第一步。有意識地觀察這些偏見並主動質疑我們自己的想法和決定可以幫助減少它們的影響。尋找不同的視角和資訊也有助於挑戰我們的先入為主的觀念。

總之，認知偏見深深地根植在我們的思維過程中，並且可以顯著地影響我們的決策和信念。理解它們對於個人發展以及做出更明智和理性的決策至關重要。記住，改變的第一步是意識，而朝著無偏見思考的旅程，雖然具有挑戰性，但非常值得。