User Personas of the OCaml Platform

This section presents an overview of different user personas within the OCaml ecosystem, highlighting their unique goals, motivations, and challenges.

The aim of defining these personas is to establish a framework that can guide the roadmap for the OCaml Platform, ensuring that it addresses the unique requirements of each archetype. By understanding and catering to the specific characteristics of each persona, we can contribute to the success of the OCaml ecosystem while empowering users to achieve their individual goals.

(U1) Application Developer

Builds OCaml applications to solve real-world problems

Application developers using OCaml are focused on building solutions to real-world problems. They care deeply about application stability, reliability, performance, and scalability. Some applications examples include building web applications for the browser with Js_of_ocaml, unikernels with MirageOS, and image classifiers using Owl.

1. Application Stability and Reliability: Application Developers care about their end-users' experience. Releasing stable and reliable applications to their users is essential to provide a positive user experience and minimize failures or downtime. Stability also allows developers to reduce maintenance costs and focus on improvements instead of bug fixes or technical debt. Following today's best practices involves reproducible builds to avoid impacts from moving dependencies, maintaining tight control over dependencies to stay aware of potential security risks, and using tools to implement tests, enforce best practices, or run continuous integrations.
2. Performance and Scalability: Most applications have performance constraints they need to respect, for scalability or to provide a good user experience. Scalability is also a common concern for application developers who need to anticipate a growing user base or data sets. To improve applications performance, developers rely on instrumentation tools, profilers or benchmarks. Another common way to optimise applications' performance is to support parallel processing of performance-intensive parts of their code.
3. Distribution and Deployment: A simple distribution or deployment process is crucial for Application Developers to continuously deliver their applications to their users. Things that can help developers to simplify their distribution workflows include cross-platform builds; CI/CD pipelines; automatic generation of installers; integration with distributions platforms across operating systems (Apple Store, Windows Store, Docker Hub, etc.); tooling to generate assets for the target Platform (e.g. JavaScript, iOS, Android, etc.); and integration with third-party systems that offer a streamlined deployment story like Docker or Nix.
4. Security and Compliance: Application developers prioritize security to protect sensitive data and maintain user trust. They may have to comply with industry security standards, such as maintaining a Software Bill of Material, and follow regulations. Companies often organize security audits to proactively identify and address potential issues. Individual developers typically follow less rigorous processes, relying on security features from their package manager or vulnerability scanners. To further control dependencies and enhance security, developers may fork and vendor libraries, allowing them to respond quickly to security issues and customize dependencies to fit their specific needs.
5. Comprehensive Documentation and Resources: Access to documentation and learning resources is essential for developers of all experience levels. Encouraging engagement with community-driven resources, such as forums and knowledge bases also fosters a supportive ecosystem where developers can both consume and contribute knowledge.

(U2) Library Author

Builds re-usable OCaml components, not directly executable

Library authors write code consumed by other developers while using other people's libraries to build their own. They focus on ensuring that their libraries are well-designed, well-documented, and easy to use. Library authors often have to manage dependencies, versioning, publishing, and testing. They may create libraries for a wide range of purposes, such as web development, data processing, or machine learning, and contribute them to the opam repository for the OCaml community to use.

1. Package Publication: Publishing packages on package repositories is the primary mean through which Library Authors can share their work. They want this process to be as intuitive and frictionless as possible so they can focus on building their library, writing documentation and engaging the community.
2. Code Reusability and Modularity: Library Authors care about composability they design APIs that are easy to use and work well with other libraries. They focus on creating simple APIs and follow common practices to ensure their libraries feel familiar to developers. They also pay attention to compatibility, carefully choosing APIs from other dependencies and weighing the pros and cons of using different packages.
3. Interoperability and Portability: In some cases, libraries support specific target Platforms (operating systems, architecture, etc.), but in most cases, Library Authors want to make their libraries available to as many users as possible. To do this, they rely on tooling to test their libraries on different Platforms. Tools to support Library Authors include CI systems and cross-platform build systems.
4. Code Quality and Maintainability: When they have released a library, Library Authors accept a maintenance cost due to the evolving language and ecosystem. Every breaking change in the language or in the library dependencies will incur friction. They want this workload to remain as minimal as possible as, ideally to remain unimpacted by any changes to the ecosystem, only accepting maintenance costs for updating their libraries when needed. Conversely, they want to know the impact a release will have on their users: applications or other libraries that depend on theirs.
5. Documentation and Examples: An important aspect of sharing their work involves making sure users can use their libraries easily. To that end, Library Authors care about providing their users with good documentation and other resources to reduce the learning curve, like code examples. This documentation and resources need to stay up-to-date with library changes and improvements so they don't become a burden for the author or irrelevant to the users.
6. Community Engagement and Support: Library Authors make the effort of sharing their work with the rest of the community. They care about the health and growth of the ecosystem. They also value connection with the community by announcing their packages on forums and social media, gathering feedback from their users, or even promoting contributions to their libraries.
7. Support Application Developers: Ultimately, Library Authors want to help Application Developers (U1) succeed. They care about aspects like performance, security, and maintainability to ensure their libraries provide a solid foundation for building applications. By focusing on these areas, they empower developers to create robust and efficient real-world solutions using their libraries.

(U3) Distribution Manager

Works on maintaining a Linux distro like Debian or RedHat

Distribution managers work on maintaining Unix distributions, such as BSD, Debian or Red Hat, and ensuring that OCaml applications, libraries, and tools are compatible with their distribution policies. They care about package management, co-installability, and tight integration with external (non-OCaml) dependencies. Distribution managers often have to decide which OCaml components to include in their distributions and ensure they are well-supported and secure.

1. Package Selection and Inclusion: Distribution managers are responsible for choosing the right OCaml libraries and tools for their distribution. They need to consider factors like long-term support, stability, compatibility, and feature availability. They also need to ensure a consistent set of co-installable packages and manage any dependency conflicts that arise. By minimizing the risk of package incompatibilities and breaking changes, distribution managers create a reliable and user-friendly environment for developers. As such, seamless integration with external (non-OCaml) packages is a priority for distribution managers.
2. Security and Long-term Support: Distribution managers need to stay informed about potential security issues in the packages they include in their distributions as they are responsible for releasing hotfixes and security updates in a timely manner. They also provide long-term support for the chosen OCaml libraries and tools, like Debian's 5-year LTS support, so they need to be confident the inclusion of a library will not create an additional maintenance burden for the distribution.
3. Package Management and Tooling: Instead of relying on language-specific tooling, Distribution Managers use their distribution's existing package management infrastructure. To streamline the release process of their distributions, Distribution Managers care about having a simple way to port language-specific packages to their distribution package managers easily.

(U4) Newcomer

Learn the syntax and language features

Newcomers are individuals who are new to OCaml or interested in learning the language for personal or educational purposes. They are often drawn to OCaml because of its strong type system, functional programming paradigm, and rich ecosystem of tools and libraries. Newcomers often care most about easily setting up a development environment, having accessible learning resources, and receiving guidance from the OCaml community.

Newcomers come in all shapes and colours; some might have experimented with other languages, and for others, this might be their first introduction to CS. How one learns OCaml depends on one's affinities and previous experience. What they often care most about easily setting up a development environment, having accessible learning resources, and receiving guidance from the OCaml community. They might start by experimenting with simple OCaml programs, like a command-line calculator, or building basic web applications using a lightweight framework.

1. Setup an OCaml Environment: Setting up their environment is the first step Newcomers undertake in their journey to learn OCaml. They need to have comprehensive documentation that guides them through a straightforward installation workflow. They also need to easily understand what went wrong if they have an error at any point during their onboarding. It's also important that the installation time remains minimal: by minimizing the time it takes to start working with OCaml, Newcomers can focus on learning and experimenting with the language. Not all of them will need to install OCaml on their systems: using online IDEs, Playground, or Docker containers can simplify the setup for users who want to get started with writing OCaml as quickly as possible.
2. Learn OCaml: Some Newcomers have experienced developers in other programming languages, and some are learning OCaml as their first language. Given the huge range of learning needs, Newcomers need to have access to an exhaustive library of documentation and resources to learn OCaml, including Guides, Tutorials, Language Manual, Package documentation. All of these need to cater for the need to inexperienced developers, as well as provide comprehensive information for experienced ones. Newcomers also want their learning experience to be enjoyable, by using playgrounds, interactive exercises, or beginner-friendly content.
3. Discover the Ecosystem: Once Newcomers have set up their environment and start learning the language, they will need to interact with the ecosystem, either by using OCaml packages, asking questions on forums, reading documentation, etc. It's crucial to provide newcomers with a good understanding of how the ecosystem is organised: how to find answers to questions, what are the best practices, what are the pros and cons of two alternative packages, etc. Newcomers want to have examples they can reuse to perform specific tasks, such as creating a web server, or a command line interface.
4. Community Support and Mentorship: A welcoming and supportive community is vital for Newcomers learning OCaml. Learning a new language and discovering an ecosystem can feel daunting and Newcomers may reach out to the community for support. More experienced developers in the community can help create a positive learning environment by answering questions on forums, chats, GitHub, and other Platforms.

(U5) Teacher

Teaching the OCaml syntax and language features or using OCaml to teach other CS principles.

Teachers are individuals who are responsible for teaching the OCaml programming language to students. This can include teaching the syntax and language features of OCaml or using OCaml as a tool to teach other computer science principles. Teachers typically have to support multiple operating systems (such as Mac, Unix, and Windows) to provide a smooth onboarding experience for their students. Teachers care about creating engaging educational content, ensuring a smooth onboarding experience for their students, and integrating OCaml with educational tools like Jupyter Notebooks.

1. Offer Frictionless Setup Experience for Students: Teachers need to support a variety of environments across multiple operating systems that their students use. They want to simplify and unify the process of setting up a development environment for their students as much as possible. To do this, they want to be able to point students to a single platform that provides a simplified installation experience, such as a VSCode extension, an online editor, etc. They also work on step-by-step guides and tutorials for their students, both for configuring their environment, but also installing and using libraries.
2. Engage Students: Teachers try to accommodate different learning styles and student backgrounds. They create engaging materials and use tools like Jupyter Notebooks to support interactive learning experiences. They listen to feedback from their students, experiment with different teaching approaches and continuously improve their learning materials based on this.
3. Alignment with Curriculum and Learning Objectives: In addition to teaching OCaml, Teachers also need to align with relevant curricula (such as computer science, algorithms, system programming, etc.). Teachers design lessons and projects that demonstrate the practical application of OCaml on these topics.

(U6) Data Scientist

Uses OCaml for short-term projects, often in scientific modeling or data analysis

Researchers use OCaml for scientific modeling, data analysis, and other short-term projects. They may be academics, researchers, or professionals working in various industries, including government and corporate sectors.

Unlike Application Developers, they are not necessarily focused on long-term software stability and deployment. They may develop code for one-off analyses or rapidly-evolving models, where flexibility, iteration speed, and expressiveness are paramount. However, like Application Developers, they care about having an efficient working environment and access to resources to solve their unique problems.

1. Rapid Prototyping and Flexibility: Researchers often work on projects that require rapid prototyping and constant iteration. They value a programming environment that allows for quick changes and easy testing.
2. Rich Scientific Libraries and Tools: Access to libraries that support scientific computation and data analysis is vital for Researchers. Similarly, the ability to visualize data and analysis results is often crucial in their work. Researchers typically work with large datasets, so they need efficient tools for data manipulation, cleaning, and processing. They also often require support for various data formats and databases.
3. Interoperability with Other Languages and Tools: Since they often work within a larger ecosystem of data science tools, interoperability with other languages (like Python or R) and tools (like Jupyter notebooks or TensorFlow) can be important. This might involve calling OCaml from these languages, or vice versa, as well as integrating with other data science tools.
4. Reproducibility: Ensuring that others can reproduce their work is often crucial, particularly in an academic or research context. This involves keeping track of dependencies, versions, and environments, as well as documenting their work so that others can understand and replicate it.