Goal of Dune¶
The Dune project strives to provide the best possible build tool for the entire OCaml community, including individual developers contributing to open source projects in their free time, larger companies (such as Jane Street), and communities, like MirageOS and Irmin. Additionally, we aim to provide the same features for other neighbouring communities, such as Coq and possibly Reason/Bucklescript, in the future.
We haven’t reached this goal yet, as Dune still requires development in some areas to be such a tool, but we’re steadily working towards that goal. On a practical level, a few boxes must be checked, and a considerable number of details needs to be sorted out. At a high-level, we think a tool that works for everyone in the OCaml community should at least:
- have excellent backward compatibility properties
- have a robust and scalable core
- remain a no-brainer dependency
- remain accessible
- have very good support for the OCaml language
- be extensible
At this point, we’ve done a good job at 1, 3, 4, and 5. We’re currently working towards 2 and are doing the preparatory work for 6. Once all these boxes have been checked, we’ll consider the Dune project complete.
Below, we develop each point and give some insights into our current and future focuses.
Have Excellent Backward-Compatibility Properties¶
In an open source community, two types of groups exist: those with enough resources to continuously bring their projects up-to-date and those who work on them in their free time. The latter obviously can’t provide the same level of continuous support and updates as the former.
From the Dune point of view, we consider every released project with
files a precious piece that will potentially never change, so we discourage
changing Dune in a way where it could no longer understand a released project.
Of course, we can’t give a 100% guarantee that Dune will always behave exactly the same. That would be unrealistic and would prevent the project from moving forward. In order to provide good backward-compatibility properties while still keeping the project fresh and dynamic, we need to properly delimit, document, and version the set of behaviours on which users rely. For this to be manageable, the surface Dune API must remain small.
A distinguishing feature of Dune allows the user to declare which version of
dune tool they wrote the project against, and
dune will morph
itself to behave the same as this version of the
dune binary, even if it’s
a newer version. As a result, a recent
dune binary version can understand a
wide range of Dune projects written against many different versions of Dune,
and while we strictly follow semantic versioning, new major versions of Dune
effectively introduce very few breaking changes. Most projects don’t need upper
bounds on Dune.
This guarantee is of course limited to documented behaviours.
Have a Robust and Scalable Core¶
Tech companies tend be fond of big mono repositories, so for compatibility, Dune must consume large repositories without blinking. It not only needs to build fast, but more importantly, it must not impede fast feedback during development, no matter the size of the repository.
Note that we’ll only test Dune on repositories as large as people participating in Dune’s development require. Currently, the largest user is Jane Street. If someone wanted to use Dune on much larger repositories than the ones used at Jane Street, and this required a significant amount of effort on Dune, this wouldn’t be considered unless we get some help to do so and we can keep the other promises.
In particular, while making Dune scalable, we must also ensure Dune doesn’t turn into a monster, because no one wants to force their users to install a monster to build their project. This brings us to the next point of Dune being a no-brainer dependency:
Remain a No-Brainer Dependency¶
Dune is a hard dependency of any Dune project. Anyone using Dune to develop their project will have to ask their user to install Dune. For this reason, it is very important to keep Dune as lean as possible.
We need to be careful when we start relying on an external piece of software or when we introduce new concepts. We must not introduce duplication or useless stuff. The overall projects has to remain lean.
It’s also important to keep Dune as easy to install as possible. Currently, the only requirement to build Dune is a working OCaml compiler. Nothing else is required, not even a shell, and we should keep it this way.
Since Dune aims to be the best possible tool for the whole OCaml community, it’s important to keep Dune accessible. Getting started and learning Dune should be straightforward.
For that purpose, when designing the language (the command line interface or the documentation), we must take on the new-user perspective, one who just discovered Dune and its features, because Dune should be suitable for everyone! It also needs to provide advanced and more complex features for expert users. However, the documentation should always flow from the simpler concepts and common tasks to the more complex ones, even if the simpler features can be explained as instances of the more general ones.
Have Excellent Support for the OCaml Language¶
There are many, many build systems out there. Dune stands out because it primarily targets the OCaml community, so Dune must come with excellent support for the OCaml language and OCaml projects in general.
If it didn’t, Dune would just be yet another generic build system.
Perhaps in the future some of the general build system will take over, and Dune might just become a plugin in this system. It could even disappear into the language, if the compiler gains significant high-level features. But for now, Dune is a standalone build system that primarily serves the OCaml community’s needs, and to the extent that is reasonably possible, the needs of other functional language communities.
No matter the quality of the OCaml language’s support, it will never be enough to cover every single project need. For this reason, Dune must provide some form of openness for projects with need that don’t completely fit in the Dune model.
In the long run, extensibility tends to obstruct innovation, and we should always strive to ensure that we cover all the general needs of the main Dune language; however, we’ll always need an escape hatch for Dune to remain a practical choice.
It’s pretty clear that extensibility must be done via OCaml code, and currently it’s a bit difficult to use OCaml as a proper extension language, though some work is being done to help on that front.