Introduction

In the relentless pursuit of delivering value to customers, with unparalleled speed and reliability, the software development world has yet to widely embrace a revolutionary principle – the “Constant State of Ship”. This state, where software artefacts and products are perpetually poised for release into production environments within just 15 minutes’ notice, has emerged as a driving force behind best practices that enable true continuous deployment. Remarkably, this groundbreaking concept formed the foundation of the pioneering “Javelin” software development approach, a visionary approach conceived by FlowChainSensei (Bob Marshall) at Familiar circa 1996 and onwards, foreshadowing the industry’s even-now-yet-to-be-realised embrace of these practices.

The Power of “Constant State of Ship”

The “Constant State of Ship” serves us as an unyielding forcing function, inviting teams to adopt and adhere to a comprehensive set of best practices that catalyse the seamless flow of software into production. Let us explore how this principle reinforces each of thirteen fundamentals of Continuous Delivery (hat tip to Dave Farley):

The 13 Fundamentals Enabled

1. A Repeatable, Reliable ProcessWith the ever-present possibility of an imminent release, teams may choose to establish a well-defined, automated pipeline for building, testing, and deploying their software. This process needs to be repeatable and reliable, minimising the risk of human error and ensuring consistency across releases.

   The “Constant State of Ship” mindset suggests that teams have a streamlined, automated release pipeline that can be triggered at any moment. Manual steps and ad-hoc and emergency exception procedures become liabilities, as they introduce variability and increase the chances of mistakes during deployment.

   To achieve this repeatability and reliability, teams are supported to invest in build automation tools, automated testing frameworks, and deployment automation pipelines. Every step of the release pipeline can be codified, documented, and thoroughly tested to ensure predictable outcomes each time.

   Moreover, the “Constant State of Ship” principle fosters an environment of continuous learning and improvement. Any failures or issues encountered during a release are promptly analysed, and the release process is refined to prevent future occurrences. This cycle of continuous feedback and optimisation ensures that the release pipeline remains reliable and efficient, even as the codebase and systems evolve over time.

   By operating in a “Constant State of Ship” mode, teams are invited to treat the release pipeline as a critical component of their software development lifecycle, investing the necessary resources and effort to make it repeatable, reliable, and capable of delivering changes to production environments at a moment’s notice.

2. Automate All the ThingsIn a “Constant State of Ship” paradigm, manual interventions become significant bottlenecks and risks, hindering the required velocity and reliability. Automation becomes imperative, spanning every aspect of the delivery pipeline, from code compilation to infrastructure provisioning. The threat of an imminent release leaves no room for error-prone manual processes that could delay or derail a deployment. Teams must automate build processes, test execution, environment provisioning, deployment steps, and release orchestration to ensure consistency and minimise the risk of human error.
3. Maintain a Releasable StateThe core tenet of “Constant State of Ship” requires that the codebase and associated artifacts remain in a perpetually releasable state. This principle invites teams to address issues promptly, maintain a high level of code quality, and vigilantly consider the accumulation of technical debt. Any defects, bugs, or instabilities in the codebase could potentially disrupt an imminent release, leading to costly delays or failures. Teams must adopt practices like continuous integration, automated testing, and ensemble programming to ensure that the codebase remains in a stable, deployable state at all times.
4. Focus on Robust (Real) Quality Assurance

   In the “Constant State of Ship” paradigm, where the possibility of demand for an immediate release is ever-present, quality assurance cannot be treated as an afterthought. “Constant State of Ship” invites the integration of quality practices throughout the entire development lifecycle, ensuring that quality is baked into the software from inception to deployment.

   While testing plays a role, it is merely one facet of a comprehensive quality assurance strategy. Teams may choose to adopt a holistic approach that emphasises quality as a continuous, pervasive practice woven into every aspect of the development approach.

   This begins with cultivating a culture of quality-driven development, where every team member participates in collective ownership and responsibility for the quality of their work. Practices such as clarity of (quantified a la Gilb) requirements, ensemble programming, peer code reviews, adherence to coding standards, and continuous static code analysis can help identify and mitigate potential issues early in the development cycle.

   Furthermore, “Constant State of Ship” invites teams to embrace principles of iterative and incremental development. By breaking down complex features into smaller, manageable, well-bounded increments, teams can more effectively manage quality risks and ensure that each increment and subsystem meets the required quality criteria before progressing to the next.

   Continuous integration and deployment pipelines play a pivotal role in this quality assurance strategy, enabling teams to continuously validate and verify the software’s functionality, performance, and stability with each incremental change. These pipelines automate the execution of various quality checks, including unit tests, integration tests, and performance tests, providing real-time feedback and enabling teams to address issues promptly.

   However, quality assurance extends beyond mere testing alone. Teams have the opportunity to adopt a holistic approach that encompasses design practices, architectural decisions, and operational readiness. By considering quality implications at every stage of the software development lifecycle, teams can proactively identify and mitigate potential risks, ensuring that the software remains in a releasable state at all times.

   “Constant State of Ship” elevates quality assurance to a core discipline that permeates every aspect of the software development effort. By fostering a culture of quality-driven development and adopting continuous quality practices, teams can attend to the needs of all the Folks That Matter, with confidence, knowing that their software meets the highest standards of reliability, stability, and performance.

5. Implement Robust Deployment PipelinesAchieving a “Constant State of Ship” necessitates the implementation of robust deployment pipelines. These pipelines automate the entire process of building, testing, and deploying software changes, ensuring consistency and minimizing the risk of errors. With the ever-present possibility of an imminent release, teams cannot afford manual, error-prone deployment processes. Automated deployment pipelines provide a standardised, repeatable path to production, reducing the likelihood of failed or inconsistent deployments.
6. Monitor the PipelineRegular smoke testing of the deployment pipeline is crucial in a “Constant State of Ship” mode. This practice helps catch issues early, before they can impact production environments, ensuring the pipeline’s reliability and preventing costly downtime. The possibility of an imminent release amplifies the importance of having a thoroughly validated deployment pipeline. Smoke tests act as a safety net, verifying the integrity of the pipeline and identifying any potential issues that could disrupt a deployment.
7. Integrate ConstantlyThe “Constant State of Ship” mindset encourages teams to integrate their changes frequently, often multiple times per day. This practice surfaces issues early, reduces merge conflicts, and ensures that the codebase remains in a releasable state, ready for deployment at any given moment. Infrequent integration can lead to divergent codebases, making it harder to identify and resolve conflicts, which could potentially disrupt an imminent release. By integrating frequently, teams can maintain a stable, unified codebase that is always primed for deployment.
8. Evolve the ArchitectureMaintaining a “Constant State of Ship” over time invites the continuous evolution of the system’s architecture (see also: Reverse Conway). Are teams prepared to refactor and adapt their architectures to accommodate new requirements, technologies, and scaling needs, without compromising the ability to release rapidly and reliably? As products grow and evolve, architectural decisions made early on may become hindrances to continuous deployment. The “Constant State of Ship” principle invites teams to proactively evaluate and evolve their architectures, ensuring that they remain flexible, scalable, and conducive to rapid releases.
9. Leverage Data EnvironmentsWith the constant possibility of an imminent release, the ability to provision and manage data environments becomes critical. Teams may choose to adopt practices like database versioning, data seeding, and data masking to ensure consistent and reliable testing and deployment across environments, minimising the risk of data-related issues in production. The “Constant State of Ship” mindset invites a robust data management strategy that enables seamless and repeatable deployments, regardless of the data complexities involved.
10. Mirror Production EnvironmentsTo minimise the risk of issues arising from environmental differences, teams operating in a “Constant State of Ship” mode may choose to ensure that their development, testing, and staging environments closely mirror production environments in terms of configuration, data, and infrastructure. This practice helps identify and address potential issues before they impact the live production system. The possibility of an imminent release heightens the importance of having production-like environments, as any discrepancies could lead to unexpected behavior or failures during deployment.
11. Codify InfrastructureManually provisioning and configuring infrastructure for each release becomes a significant bottleneck when operating in a “Constant State of Ship” mode. Adopting Infrastructure as Code (IaC) practices, where infrastructure is defined and managed through code, enables teams to provision and tear down environments rapidly and consistently, minimising delays and reducing the risk of configuration drift. The “Constant State of Ship” principle invites a high degree of automation and repeatability in infrastructure management, making IaC a beneficial practice for ensuring rapid, reliable deployments.
12. Foster Collaborative OwnershipAchieving a “Constant State of Ship” invites a high degree of collaboration and shared ownership among team members. Siloed responsibilities and knowledge become obstacles to rapid delivery. Teams may choose to adopt practices that promote collective code ownership, cross-functional collaboration, and shared understanding of the codebase and delivery processes. The “Constant State of Ship” mindset invites a culture of collective responsibility, where all team members are empowered to contribute to and understand the entire delivery process, enabling seamless and efficient releases.
13. Continuous ImprovementOperating in a “Constant State of Ship” mode exposes inefficiencies and bottlenecks in the delivery pipeline and processes with uncompromising clarity. Teams may choose to embrace a culture of continuous improvement, regularly reviewing their practices, identifying areas for optimisation, and implementing changes to enhance their ability to deliver value rapidly and reliably. The constant presence of imminent releases acts as a driving force for continuous improvement, encouraging teams to continuously refine their processes, tools, and practices to achieve higher levels of velocity and quality. FlowChain was designed to systematise this very purpose.

The Visionary “Javelin” Approach

The “Javelin” approach (initally named “Jerid”) pioneered by me and my teams at Familiar from 1996 onward, was truly ahead of its time, recognising the transformative power of the “Constant State of Ship” mindset. By enshrining this principle as a cornerstone from its inception, “Javelin” has paved the way for the modern continuous deployment practices that have since become poised to gain industry standard status. This pioneering approach, along with FlowChain and e.g. Prod•gnosis, Flow•gnosis, Product Aikido, etc. exemplifies the spirit of continuous improvement intrinsic to the “Constant State of Ship” principle, ensuring its enduring relevance and impact.

Deep Cultural Implications

Reshaping the Culture and Mindset

Adopting the “Constant State of Ship” principle suggests a profound transformation that extends way beyond technical practices and processes – it hints at a seismic shift in the culture and mindset of software development teams and their parent organisations. This metamorphosis permeates every aspect of the organisation, reshaping shared assumptions, beliefs, and ways of working. However, navigating such a profound cultural shift can be a daunting challenge, often met with resistance and inertia.

This is where the discipline of organisational psychotherapy plays a pivotal role. By applying principles from psychotherapy, sociology, and group dynamics, organisational psychotherapy facilitates teams’ cultural and mindset shifts required to embrace the “Constant State of Ship” paradigm smoothly and effectively.

A Culture of Ownership and Accountability through Empowerment

The “Constant State of Ship” mindset fosters a culture of collective ownership and accountability. Organisational psychotherapy techniques, such as participative decision-making and fellowship, empower team members to take responsibility for the quality, stability, and deployability of the codebase and overall product. This sense of empowerment cultivates a culture of shared ownership, where individuals proactively address issues, collaborate across boundaries, and collectively strive for continuous improvement.

Embracing Transparency and Trust

Maintaining a “Constant State of Ship” requires a high degree of transparency and trust among team members. Organisational psychotherapy practices, such as surfacing shared assumptions and beliefs, encourage open communication and facilitate the identification of problems and risks early. By fostering an atmosphere where team members feel comfortable expressing concerns, sharing mistakes, and seeking help, a culture of transparency and trust emerges, enabling teams to collectively address challenges and ensure the software remains in a releasable state.

Prioritising Continuous Learning

The “Constant State of Ship” principle instills a mindset of continuous learning and improvement. With each release, teams gain valuable insights into their processes, tools, and practices. Embracing new shared assumptions becomes essential, as teams must continuously refine and adapt their approaches based on feedback and lessons learned. This culture of continuous learning fosters an environment of experimentation, where failures are embraced as opportunities for growth, and success is measured by the ability to deliver value rapidly and reliably.

Aligning Towards a Common Goal

Ultimately, the “Constant State of Ship” principle unifies teams around a common goal: meeting the needs of all the Folks That Matter with unparalleled speed and reliability. This shared mission transcends individual roles, responsibilities, and technical disciplines. It creates a sense of collective purpose, where every team member’s contribution, regardless of their specific function, is valued and recognised as essential to achieving this overarching objective.

By leveraging organisational psychotherapy techniques, organisations can accelerate and streamline the cultural and mindset shifts required to embrace the “Constant State of Ship” paradigm. This discipline not only makes the transition quicker and easier but also more cost-effective, as it addresses the root causes of resistance and inertia, facilitating a smoother and more sustainable transformation.

By reshaping the culture and mindset of software development teams, the “Constant State of Ship” principle cultivates an environment conducive to continuous deployment success. It fosters a sense of collective ownership, transparency, continuous learning, and shared purpose – traits that are indispensable in today’s rapidly evolving software landscape.

Embracing the Future

When the ability to swiftly adapt and innovate is paramount, the “Constant State of Ship” principle emerges as a beacon, guiding software development teams towards a future of quiet competence and competitiveness. By embracing this mindset, as exemplified by the visionary “Javelin” approach, teams can unlock the power to attend to folks’ needs with unprecedented speed, reliability, and quality – solidifying their organisation’s position as industry leaders in the software development arena.

For God’s sake, there is truly no longer any excuse for typos, misspellings, and grammatical errors in your posts, articles, and other writings.

Artificial intelligence (AI) and chatbots are transforming how we communicate. When integrated thoughtfully, this technology can optimise and enhance written communication between people. In this post, I’ll discuss some ways AI and chatbots can improve messaging, email, documentation, and other word-based interaction between humans.

Automated Proofreading and Editing

AI-powered writing tools already help by providing grammar and spelling checks. But newer capabilities can now also flag unclear phrasing, verbose language, overused words, and overly complex sentences. This aids writers in simplifying and refining their messaging before sending to a recipient. Readability statistics further help authors match their tone for the intended audience.

Summarisation and Translation Features

For long-form writing like reports or manuals, AI can generate a concise summary highlighting key facts, main takeaways, or action items. This allows collaborators or stakeholders to quickly grasp the essence before diving into the details. Meanwhile, instant translation functionality enables clear communication across language barriers.

Interactive Books

AI is also enhancing books through interactive elements powered by chatbots. Platforms like Ainklings.com allow authors to insert quizzes, discussion questions, exercises and other engaging features directly into the book text (or via sidecars). Readers can further highlight passages and interact with supplementary content related to the main narrative, enriching the reading experience.

Content Recommendations and Insights

Smart suggestions can enable more meaningful interactions through personalised recommendations. By analysing past correspondence as context, AI can prompt authors to include certain missing information, helpful examples, or reminders based on what the recipient would find useful. Language pattern analysis can also reveal insights for improving future discussions.

Automated Meeting Summaries and Notes

While AI currently struggles to match the creativity of human writing, it excels at capturing the salient points from meetings and presentations. Automated summaries of video sessions or collaborative spaces can save meeting participants time while ensuring everyone understands the key decisions or action items.

With thoughtful application, AI and chatbot tools can enhance understanding and engagement between people through better writing assistance, translation, summarisation, and recommendations. As these capabilities continue advancing, keeping the human audience at the center will be key to success.

What if the future of society depends on challenging AI and machine intelligence? This is a question that has been on the minds of many people, especially those who are concerned about the impact of technology on our lives. It is a question that is becoming increasingly relevant as we move towards a future where AI and machine intelligence are playing an increasingly important role in our lives.

If we fail to challenge AI and machine intelligence, then we run the risk of becoming completely reliant on it. This could lead to a future where machines are in control, and humans are merely drones. However, if we do challenge it, then we can create a future where humans and machines work together.

The first step in challenging AI and machine intelligence is to understand its limitations. While machines can perform many tasks faster and more accurately than humans, they are still woefully limited. They cannot think for themselves. This means that AI is far away from replacing human creativity and innovation.

The second step is to ensure that AI and machine intelligence are designed to benefit humans, not replace them. This means that machines should be designed to enhance human capabilities – Jobs’ “Bicycle for the mind” analogy.

The third step is to ensure that humans are in control of the machines. This means that humans should be the ones making the decisions about how machines are used, not the other way around. It also means that we need to develop ethical frameworks for the use of AI and machine intelligence, to ensure that they are used in a way that is fair and just.

The fourth step is to ensure that AI and machine intelligence are transparent. This means that we need to understand how machines are making decisions, and we need to be able to review those decisions. This will help us to identify any biases or errors in the decision-making process, and ensure that machines are making decisions that are in the best interests of humans.

The fifth step is to ensure that AI and machine intelligence are adaptable. This means that machines need to be able to learn and evolve over time, to ensure that they are always up-to-date with the latest knowledge and technology. This will help us to stay ahead of the curve, and ensure that machines are always working in our best interests.

In conclusion, the future of society depends on how we challenge AI and machine intelligence. If we fail to challenge it, then we run the risk of becoming slaves to our machines. However, if we do challenge it, then we can create a future where humans and machines work together to achieve our goals. By understanding the limitations of machines, designing them to benefit humans, ensuring that humans are in control, making them transparent, and ensuring that they are adaptable, we can create a future where AI and machine intelligence are our allies, not our enemies.

Or not. I prefer not. 

As John Seddon states in his most recent book, it’s far more useful to fully understand customers’ needs, through e.g. simple physical means, like pin-boards, T-cards and spreadsheets, before considering any automation.

And even then, automation has at least two fundamental flaws:

Inability to Cater to Variation in Demand

Automation and automated systems, presently and for the foreseeable future, cannot encompass variety in demand. As we’ve come to relate to the Little Britain meme “ Computer says no”. Customer demand inherently has variation. Thus, automation leads to a poorer customer experience, as many customer needs are handled poorly, or not at all. I cite the British Gas website and customer experience as a particularly egregious example.

Employment 

Let’s also look at the bigger picture of social cohesion, of which people having jobs is a part. Jobs give people meaning, status, and something to do. As well as greasing the wheels of commerce – employed people have disposable income which contributes to companies’ revenues.

The idea of Basic Income is all very fine (I’m a fan) but that concept has some major wrinkles to iron out before it becomes a shoe-in.

In the meantime, how about we try to create businesses – and other organisations – that provide meaningful employment to more people, rather than fewer? Will that negatively impact profit margins? I doubt. And there’s always Deming’s First Theorem in any case.

More and more often, the Software Industry is being called upon to live up to its fine moral pronouncements. Automation is an item in the negative column on that balance sheet.

– Bob