Botan: The first milestone

Posted 2024-03-11 12:15:00 ‐ 12 min read

After 8 months of work, we've reached an important milestone. Now we're here to tell you about what is coming next.

Botan: The First Milestone

What are you using cryptography for? Would you be interested in trying something new? Perhaps an alternate backend for your cryptography needs?

I am Leo Dillinger, a member of Haskell Cryptography Group, coordinating with the Haskell Foundation to develop free and open-source software for you and the Haskell ecosystem.

We've been working on botan, a cryptographic kitchen sink, and after 8 months of work, we have reached an important milestone: we've successfully published the first released version 0.0.1 of botan-bindings and botan-low, and botan has made it to package candidate status. Now, we are hard at work on getting the first release version of botan ready too, and could use your feedback!

Botan is an comprehensive, open-source, BSD-licenced, C++ cryptography library with a stable C API. It offers a broad variety of functionality and algorithms, including post-quantum cryptography, is developed and maintained by an active community, and has been audited in the past.

By binding to Botan, we have solved a significant problem of providing much of the necessary 'cryptographic kitchen sink' via a suitably performant, suitably licensed, open-source library. Furthermore, we do this without imposing a large maintenance burden on the Haskell community, as we are not required to maintain the Botan cryptography library itself, only the bindings to it.

A call for users

Do you use one of the following libraries?

  • cryptohash
  • crypto-pubkey-types
  • crypto-cipher-types
  • cryptonite
  • crypton
  • password

That's where you come in! We'd like you to consider giving botan a try!

botan-low is viable out-of-box, and botan will be having it's own 0.0.1 release soon enough. We are interested in seeing how they perform in the real world, and we can only do that with users - we can't listen to your feedback if there isn't any!

Or perhaps you may be a user of a library with one of these buried deep in the dependencies?

  • amazonka-core
  • hpack
  • scotty
  • servant
  • stack
  • tls
  • x509
  • warp
  • websockets

These are popular libraries, and their list of transitive dependents is quite large. They sit at the root of a lot of production haskell code, and anything involving networking, APIs, and the internet is likely to depend on at least one of them.

Would it surprise you that they are all directly or indirectly dependent on crypton, which contains unaudited C implementations that must be maintained by the Haskell community as a fork after cryptonite was abandoned by its original author.

This is a problem, and we're trying to change that with bindings to botan.

Having a solid, reliable, and well-maintained cryptography libraries is a huge benefit to the Haskell ecosystem in general, and is essential for most commercial and industrial use cases.

What are you using cryptography for? Let us know in the comments or with upvotes.

A new phase

Today, we're here to tell you about what we're working on next.

As we all know, perfect is the enemy of good; no software is perfect the first time, and we release things when they work, and then continue to improve them. We hope that this is simply the first version and first step on a journey of many improvements small and large.

With this milestone, the project enters a new phase in the software development lifecycle - maintenance and development. During initial development, we were nimble, and could make choices arbitrarily - but now that we have something that works, with an initial release and users, we have to keep it working all the while we continue further development. We now have other people invested in this, and can't make choices willy-nilly - we owe it to our users and stakeholders to listen to them.

The second milestone

That is what this next milestone is about - listening to feedback, improving the user experience, and seeing where the pain points are. Here's what we've heard, and here's what we're planning for the next three months.

Improved installation support

One of the biggest pieces of feedback that we've received is the need for improved support for the installation of the botan3 C++ library. This was a recurring item, and we've heard you loud and clear.

We'd like to spend a good chunk of time improving the installation process, with a specific focus on Windows1 and Linux support, working on the CI + unit tests for improved reliability and reporting overall.

1

Definitely more highly requested than anticipated

We're also looking into using build-type: Configure to automatically fetch and/or build the latest compatible version of Botan C++ via Cabal if it is not already installed, in order to streamline the process on all operating systems - we'd like for usage to be as easy as adding botan to your dependencies.

Development of a drop-in interface replacement for crypton

This is obviously on our mind, given our call for users, and was mentioned several times in feedback. crypton is a dependency in many important libraries in the Haskell ecosystem, and we would like to build an interface that is as near a drop-in replacement for crypton as possible.

There will be some differences, as botan doesn't necessarily support everything2 in the same way as crypton does, but we'll give it our best effort to make migration as simple as possible.

2

There are a few things that crypton supports that botan doesn't, but also vice versa - botan supports things like modern post-quantum algorithms and crypton doesn't.

Development of a high-level libsodium-like interface

We'd like to expose a high-level libsodium-like interface of selected best-in-class algorithms in order to make usage dead simple. We don't want you managing primitives yourselves - we want you calling a simple function purely or in an appropriate monad / transformer.

This might be a bit of a stretch goal, in favor of focusing on replacing crypton.

Continued development of the cryptographic typeclasses

The development of cryptographic typeclasses is ongoing, in an effort to improve per-algorithm type safety and ergonomics through more consistent handling of cryptographic primitives such as keys, nonces, and ciphertexts.

This is also a necessity for the crypton drop-in replacement.

Continued refinement of botan

There are many continued refinements and improvements to the botan library that we would like to apply.

Improved error handling is a big one, as we'd like to use HasCallstack appropriately throughout the library.

We'd also like to work on the MonadRandomIO / NonDeterministic monad in order to integrate with random and take advantage of the Uniform class to make sampling / generation of random data easy.

We'd like to also take advantage of the secure memory erasure by providing functions for creating temporary objects that are automatically zeroed when they go out of scope. This is a big one, we don't want sensitive data hanging around in memory.

Finally, we'd also like to continue improving, unifying, and standardizing the botan-low bindings for consistency overall, as the techniques for binding to functions were refined over time, and have not been evenly applied.


These are some of the things that we'd like to accomplish over the coming months.

You can find more details in the updated second funding proposal. This proposal is a continuation of the efforts of the first proposal, and is motivated by the same long-term goals.

A flag planted on the horizon

As part of the Haskell Cryptography Group, we're about more than just maintenance of existing libraries - we want to develop a full suite of modern cryptography libraries.

Cryptography is a different niche, but we see the success of Haskell web server libraries as an example of a healthy, community-driven ecosystem. There are packages ranging from the low-level like wai supporting multiple backends from the swiftly simple scotty to the deeply complex servant, and a supporting ecosystem - a rich set of libraries that provide a flexible enough set of solutions at whatever level of abstraction you need, that take full advantage of Haskell's powerful type system.

We see the success of the haskell web server ecosystem as an example of a healthy, developed niche, and as an example of Haskell making something safer and easier to use - something that we aspire to do with cryptography.

That level of development takes work, and time, and we've just hit our first milestone - bindings to a modern, stable, audited open-source cryptography library. Now begins the long work of establishing an ecosystem on top of it.

Here's what we're looking into for the long-term future:

Improving APIs with higher-order functions

We'd like to build higher-order functions to take care of complicated server/client multi-step algorithms, such as SRP6. Bundling up the necessary sequences of actions into a higher-order function that takes a couple of IO functions as arguments is a safer, more ergonomic, and more reliable way of performing these actions than handing the user a series of steps that they must call in the right order.

Integration into libraries as an alternative to crypton

We'd like to integrate botan with other libraries as an alternative to or replacement forcrypton, either through the drop-in interface, or by migrating entirely - and we'd like to make that easy by providing the appropriate tools and flags.

Of particular interest is the Haskell web ecosystem, which currently relies heavily on crypton.

Split off cryptographic classes as a separate package to be backend agnostic

We see botan as one of many potential backends, and backend-agnostic cryptographic typeclasses are like wai - a common interface. There is a difference between algorithms (sets of operations) and typeclasses (specific use cases). Algorithms are 'how we do it', typeclasses are 'what we want to use it for'. If we've defined our typeclasses correctly (as by use case), they should be backend-agnostic, regardless of the particulars of the implementation - otherwise the implementation could not fulfill its duties.

For the moment, these classes are part of botan, and we'd eventually like to split these cryptographic classes off into their own cryptography library.

Implementation of more advanced algorithms

There's a whole host of interesting and useful cryptography algorithms that have been developed in the last decade - including post-quantum cryptography, and we'd like to be able to provide them as tools for you to use.

Once we have a stable foundation, such as botan, we want to move on to implementation of more advanced algorithms, such as Merkle trees and Signal's Double ratchet / Apple's PQ3

We don't want you to be messing around with cryptographic primitives - we want you importing libraries like double-ratchet3 and sparse-merkle3 and distributed-json3 instead of having to implement it yourself.

3

These libraries do not yet exist, but we'd like them to.

Building an application framework that takes care of cryptography & security

Ultimately, we'd like to build an application framework that abstracts away cryptography & security, not unlike how the Haskell web ecosystem successfully manages away much of the complexity of HTTP servers. We'd like to develop a comparable system, but for cryptography - one that comes with modern post-quantum key exchange and encryption and secure transport built-in. Wouldn't that be something - an application framework with out-of-box working post-quantum transport scheme?

Easy-to-create applications with built-in security could be a killer application for Haskell, in an era where data safety is becoming a primary concern.

How can you help?

We've already taken the first step of binding to a modern, stable cryptography library. Now it is time to take the next. We'd like to ensure the longevity of this project as we tackle the next set of challenges.

We need users, and we need sponsors!

You can help us by commenting, voting, or pledging support - your activity here lets us know that you are interested in what we've done and what we're going to be doing. but it also has a tangible effect of demonstrating the ongoing success of this project!

Help us keep going! Follow the devlog for more frequently updated details!

Special thanks

Huge thanks to Mercury for funding the first proposal and helping us reach this milestone - this library would not be in this state without their support. If you are an engineer looking for a savvy place to work, I hear they are hiring!

Signed

Leo Dillinger,
Haskell Cryptography Group

Hecaté,
Haskell Cryptography Group

Jose Calderon,
Executive Director, Haskell Foundation