What could I build in Hylo today?

[notramo]

What is the current status of the language? Is it already suitable for simple CLI or web server side projects, or is it currently only for programming language theory enthusiasts?

[kyouko-taiga]

Right in the middle. It's definitely more than what PL theorists would consider a toy language, but it's also far from being ready for prime time.

[notramo]

I could not find libraries for it, or any existing projects written in Hylo. As a newcomer, this seems very much in experimentation stage.

Could you provide some practical examles for where its limits are currently? E.g. some projects that could be developed in the curret version of Hylo without hitting walls.

I'm not very motivated to go through the tutorial only to realize after completing it that I can not use it for any significant use cases other than widening my programming language knowledge.

I'm also curious why Hylo is not as popular as e.g. Zig despite both are new and innovative.
Some possible reasons can be that Zig aims to be a better C, and there is much need for that because C is horrible. Meanwhile Hylo aims to be a better Rust, but Rust is already quite good and many programmers enjoy developing in it (and it's already a matured language).
Another reason could be that there is no standard library for Hylo (or very difficult to find it), so it only attracts programmers who want to build libraries, and not those who want to build tools. It seems there is a similar situation regarding Pony, which has a very small standard library. The extensive stdlib could be a reason why Crystal and Zig reached success quite early. I learned Crystal during its beta. It was very easy as a newcomer to get started. I was able to start building the tools I wanted to build, immediately after learning the language basics, as the necessary components were available (in the case of Crystal, it's mostly the extensive stdlib).

There are some recommendations that I would welcome as a potential Hylo user:

• A few official packages - this is huge. It don't have to be in the stdlib, but it could be under the Hylo organization on GitHub, officially maintained. Some recommendations: HTTP server and client, JSON serialization, KDL or TOML parsing for config files, and maybe a SurrealDB adapter for data storage.
• Describe the current abilities of the language on the website, so newcomers can immediately compare it to their use case, and decide whether to invest time in learning it, or wait a bit until it matures enough to be suitable for their use cases.
• An easily searchable package registry for community packages.
• Target platform support (OS and CPU combinations) summary on the website - it's especially relevant while in such an early stage and only a few platforms are supported.
• A page on the website with the list of tools written in Hylo and companies using Hylo.

[kyouko-taiga]

I'm afraid Hylo is not ready for anything "practical" one would like to do with it. You asked if it was only for "programming language theory enthusiasts" and it is not, but that doesn't mean it is ready for serious projects either. We are still actively working on the compiler and polishing some aspects of the language design. One should expect a very unstable experience and many breaking changes.

Your suggestions are very nice and fully welcome. Unfortunately we're not yet at a point where we can produce this contents.