Let’s consider Jane Street as an example. in their what we do page, I see their mission is advance quantitative research. Accordingly I expect OCaml developments which support that end goal. In their Machine Learning page I do not see any mention of OCaml. It was nice for their trading system but it is not aligned with their whole business cycle. OCaml has owl, which aims for scientific safe code. If Jane Street is not adopting it, then who would do?
I am calling out to the community to
Survey business models of active contributors to the OCaml ecosystem, and
Initiate a roadmap that fulfills the gaps between OCaml development and industry needs.
Tarides is the only company that well-communicates these.
Contributors to OCaml should see a bigger picture of where OCaml is heading and the business motivating those directions.
Sorry for the somewhat at-odds question, but I’m wondering why the assumption is that there is a (large?) gap between the existing OCaml ecosystem and industry needs. Supposedly, if large, well-established industries (Meta, Bloomberg, Docker, Jane Street, etc) are using OCaml is because the present state of the ecosystem already satisfies their needs (at least to a sufficient extent), right?
My (not particularly well-informed) feeling is that the main issue with industrial adoption of OCaml is not a question of technical quality of the ecosystem, but one of size: the number of people programming in OCaml is small, so it is hard to hire OCaml programmers, which makes it risky for companies to use OCaml in their products, which leads to fewer OCaml offers, which discourages programmers from learning OCaml in the first place.
And the reason for the small size of the OCaml ecosystem is again, not technical, but economical: there is much less money being put into OCaml than into other well-known languages. And on this point, one may be interested by the following talk by the creator of Elm: https://www.youtube.com/watch?v=XZ3w_jec1v8
the main issue with industrial adoption of OCaml is … one of size.
I personally see a great business value in simple Algorithms for Decision Making. Moreover, The world is sparked by AI agents augmented with reasoning capabilities but modern dev workflows are not well-integrated with symbolic or logical modeling
I got attracted to OCaml, and no other ecosystem, because of its unique design philosophy, combining formal safety with pragmatic agile. That sweet spot should be convincing for investors.
Thinking as an economist, why don’t the OCaml community address the challenges not solved in any other ecosystem, and well-utilize OCaml’s competitive unique edge, to gain funding and scale-up?
Otherwise, Why would anyone invest in a new ecosystem?
Perhaps not exactly what you have in mind, but I expect that you would be interested to check out what Imanda is doing these days.
On a completely different note, keep in mind that each contributor has their own motivations, and achieving mass popular appeal or adoption might very well not be at the top of many lists. Personally my very limited contributions are motivated by wanting to improve the best programming system I know, but I don’t assume technical excellence and wide adoption are highly correlated.
Personally my very limited contributions are motivated by wanting to improve the best programming system
Big problems and big solutions can be realized only by collective cooperative efforts. That’s why we have The Linux Foundation, as Linux is no longer a project of indivisual hobbyists.
I don’t assume technical excellence and wide adoption are highly correlated.
OCaml may not be widely adopted but its small community should be aligned.
Why? A lot of hobby projects lead to nice developments. Plenty of research projects are very interesting. Having fun is a legitimate enough reason to code. Not everything has to revolve around a capitalist purpose and be heavily optimised towards that goal.
Luckily OCaml and its community doesn’t belong to a single entity. It is one of its qualities.
Cooperation does not entail centralized capitalism. Decentralized community-driven projects need a special kind of an economic model to sustain. See Working in Public by Nadia.
I would agree with @Khady that an entire community doesn’t really need to be aligned to produce industrial software (and my opinion is that it shouldn’t). The OCaml community may be small, but it’s made up of high-quality people, which has led to the language already being used in industry — so there are already “business models” (and not necessarily from companies, either) that work.
That sweet spot should be convincing for investors.
I think most investors are not technical enough to know that, and might also be wary of dev that are very enthusiastic about a given technology, because they perceive them as devoted to that technology and not to the business.
That’s why a good engineer communicates technical specifications as a story. Investors may not know how Rust addresses immutability but they know the hardware cost of Memory Leaks.
This. Getting inspired to do a hobby project is one of the first points where people may come in contact with OCaml, and, a while later, some of these people end up pitching OCaml to their company, or go on to create their own companies. Creating welcoming spaces to explore, learn, share, get feedback, and meet the community is IMO essential.
One thing we could do is to allow filtering the OCaml Planet by type of content (e.g. company / personal / research / open source project), for people who only want to see specific parts of what is happening in the OCaml community.
It would be great if companies using OCaml become more vocal about how they use it. Then we can feature and share this kind of content more so more people can discover it more easily. This increased visibility can make hiring OCaml developers easier for these companies. We’re also always available for success story interviews, where OCaml companies and the problems they solve are featured on OCaml in Industry.
