Show HN: Lady Deirdre 2 – Rust Framework for Compilers and LSP Servers | Hacker News Reader

Show HN: Lady Deirdre 2 – Rust Framework for Compilers and LSP Servers

github.com

141 pointsEliah_Lakhin36 comments

Greetings!

I would like to share with you my project, Lady Deirdre.

Lady Deirdre is a framework that helps you create new programming languages in Rust. It is specifically designed to develop compilers and interpreters with support for code editor extensions (LSP servers) from day one.

The framework includes essential components to design parsers and semantic analyzers capable of incrementally reparsing dynamically evolving source code.

Lady Deirdre can be seen as a replacement for preexisting projects with similar goals, such as Tree-Sitter, Rowan, or Salsa. However, Lady Deirdre aims to offer a unified framework API that guides you through the steps of programming language development, providing even more components necessary to develop a full-featured language ecosystem. For example, components to develop a source code formatter.

I will be happy to answer any questions.

Ilya

Open articleView on HN

This is cool, I think an integrated compiler and LSP is a great standard for new languages. I'd go one further though and say that syntax highlighting should also be part of that language core. This way you have a single source of truth wrt to that language: how it should look, what it means and how it runs.

Tree-sitter is widely supported (both in editors and on web) for syntax highlighting as well as making semantic nodes available for external tools to interact with. Is there any chance you'd add a tree-sitter integration to this project? Or conversely, build out a compatible API that can be used with editors/tools that use tree-sitter's library?

The licensing is a bit confusing - for example, what happens with open-source projects that use this that are then used in commercial projects?

> build out a compatible API that can be used with editors/tools that use tree-sitter's library?

That's an interesting idea. Tree-Sitter and Lady Deirdre are quite different in their approaches to parsing. Tree-Sitter is a GLR parser, while Lady Deirdre is a recursive-descent parser. In the Lady Deirdre API, there are customizable traits that let you define new types of files with parsers ("documents" in terms of Lady Deirdre). Perhaps it would be possible to create an adapter, but I would implement it as a separate crate.

> what happens with open-source projects that use this that are then used in commercial projects?

Good question. The idea is that if you link to Lady Deirdre in the Cargo.toml dependencies, it is up to the commercial project authors. They will compile the actual executable intended for selling by downloading both your crate and my crate. However, I'm not a lawyer, and this is not legal advice. Just my thoughts.

First, I really like language frameworks like this. I think it could become very useful and popular.

That being said, I really doubt anyone will buy a commercial license. People don't sell compilers and IDEs, and the ones who do are large corporations who make everything themselves. Look at state of language tooling today: nearly every compiler is open-source, every language server is free, every IDE is free except JetBrains (a well-known company with a large reputation) and Sublime Text (being heavily replaced by VS Code), and anything not open-source has an active open-source alternative. Nobody's going to buy a license from you to sell their compiler, because nobody's planning to sell their compiler.

For this reason I'd recommend changing the license. I don't think it's overly-restrictive or dishonest, I think it's fair to expect being paid if someone makes >$200,000 off your work. But nonetheless it hurts adoption, a lot of people will see "proprietary" and not even read the license text. You're more likely to make money distributing it as MIT / Apache, letting it get popular, and setting up donations/sponsorships to fund development. But honestly, if you're looking to make money this isn't the space to do so: you could be hired by someone to work on PL, or you could sell something like a game, but you're not going to sell your own PL.

Thank you for your feedback!

I agree with your point about the licensing. I would also add that tools for the development of compiler front-ends are quite a niche market. So, honestly speaking, I don't plan to earn much from my project regardless of the license terms. This work is part of a higher-level in-progress toolset, which is closer to the end users. I have dedicated it as a separate project primarily for public preview, with some restrictions on distribution and use, as I haven't decided on the overall toolset distribution model yet. But it is possible that I will change the licensing terms of Lady Deirdre in the future to something less restrictive (maybe even MIT) to make it more popular, this is just not my current goal. I apologize for any inconvenience my current licensing terms may cause.

You could however offer a cloud based static analysis API for security companies, which could be a viable source of revenue to fund the development.

> I apologize for any inconvenience my current licensing terms may cause.

Friendly advice from a stranger, worth what it costs: I believe the greatest inconvenience of a commercial license will be to _you_, as opposed to end-users.

The market is so saturated with open source, developers expect it, and commercial licenses cause huge usage friction and doubt.

The "dual license" died as a viable business model around the time Meta (née Facebook) and Microsoft starting investing billions into free-as-in-beer OSS, 10-15 years ago.

Today, this model will only sabotage usage. People using your OSS is good. Even if your goal isn't to become popular, usage & feedback are learning, and while it's fine for popularity not to be a goal, I would encourage you not to proactively target the opposite of popularity.

If you have some other IP (the other tech you're describing) that can remain proprietary or wrapped behind a service, then people using other pieces of your stack for free is a _really good thing._ At least, they're aware of your brand and have started to trust it. At best, they're ready to use your commercial offering out of the gate.

IANAL/IANYL. Make it dual-licensed, AGPL with MIT by request. Indicate that you are presently declining MIT license requests. The chances of your agreement being adhered to are near zero, regardless of the legality. On the other hand, any mention of GPL will send lawyers running for the hills - AGPL more-so (and legally so, it makes your code basically useless to a corp). The MIT carveout is merely there so that contributors don't throw a fit if you choose a more permissive license down the line.

It's 5 thousand after you've made 200 thousand though? I think that's the only restriction? Unless he updated it before I got to look.

> It's 5 thousand after you've made 200 thousand though?

That's correct. The license agreement requires purchasing a commercial license per product if it reaches a certain revenue threshold. I believe this price should be feasible for a business, and this restriction is not unfair for regular programmers either. To clarify, the license needs to be renewed annually to continue receiving new versions from me. Additionally, I reserve the right to change the price in the future. However, license renewal is not a strict obligation. You can continue using the previous versions if you buy the license at least once.

The idea is to replace the donation model widespread in typical OSS projects with a formal obligation to purchase a license. I believe this approach more accurately expresses what the authors really want in exchange for their labor.

I think your approach is good. There are several ways to make money by independently making software:

1. Charge money for the software

2. Charge money to help people implement/use the software

3. Leverage the experience and credibility gained to get a company to hire you or sell consulting services

4. Hope that people give you money for the software out of the goodness of their heart

Opting for an MIT license pretty much ensures that the software will only make money for you indirectly via option 3, and maybe you'll get a trickle via option 4. There are numerous examples of people making popular OSS that become frustrated with companies that are using their software to make money, and they aren't getting anything for it. They have to work for other people for money despite having made software that generates lots of value. If you want to ensure that if your software creates substantial value, you get some of it, you need a license like yours.

I think there is a strong argument that if you want to get paid for your software, you simply shouldn't make it any flavor of source-available. A company may actually be more willing to buy software that is closed source than open source, because the closed-source software is simply perceived as being more valuable. Of course, as an engineer, I love that the source code is available.

Yeah the readme claims you can use this in an open source project. This is weird. An open source project that bundles this software (as a language implementation necessarily must) will _not_ remain open source. The only way a related work remains open source is if it depends on this project as a _peer_ dependency.

So as it is, I seriously doubt that anyone wanting a quicker way to implement a language will use this; why would I want to hamper adoption of my language by hobbling it with such a license?

This is awesome. Thank you!

I like the Alpha Centauri reference here lol.

I was just thinking about that haha. I'd think she's kinda the opposite of a rust lover though.

(did you know the game + expansion are available for 4.99 on Steam and it runs out of the box on linux? Just saying)

The way the headline was written it seemed like this was the second iteration of Lady and not just a 2.0 release announcement

It would have also gone a long way if you had mentioned the licensing, which I am always interested in: https://github.com/Eliah-Lakhin/lady-deirdre#copyright

As for a question: why the seemingly needless location of everything down in a "work" folder? Is there something else that you envision one day living at the top-level which you just planned for by putting everything someone would care about one further click away?

You are right, this is indeed the second iteration of the project. Since the first iteration (released about a year ago), many things have changed and been reworked. The most notable change is the introduction of the semantic-analysis framework. Lady Deirdre 1 was merely an incremental parser library.

> why the seemingly needless location of everything down in a "work" folder?

It's just easier for me to organize the filesystem this way on my local machine. Everything that is unrelated to the development is outside of the work directory. Also, the license agreement refers to the "work". Perhaps it would be clearer for users if the work directory is clearly dedicated.

> Is there something else that you envision one day living at the top-level which you just planned for by putting everything someone would care about one further click away?

Everything that I planned to publish that is related to Lady Deirdre is already in the repository. I use this project for my other programming language project that I plan to release soon, but it will be in a separate GitHub repo. Actually, Lady Deirdre initially was separated from the language project codebase as I thought it may be useful for other programming language authors.

"replacement for preexisting projects with similar goals, such as Tree-Sitter, Rowan, or Salsa."

To be a true replacement, there is at least one crucial feature missing: A LICENSE file that starts with "The MIT License (MIT)".

Nothing against "source available" but thinking to have a chance to stand in for more permissively licensed projects is very much unrealistic.

It’s amusing to me that we as a community both complain that OSS maintainers aren’t paid and then expect someone who’s making the source available to pick OSS. Let people try alternate funding models and honestly as a hacker community we should be a bit more supportive of the work required to bootstrap something.

Yeah I would like to see more ideas like this.

I really hope there's a model out there for OSS at least where folks can get paid.

I always wondered what if engineers had a $100/month budget to donate to OSS founders. It wouldn't cost an employer much and would be a good way to give back.

Yeah I would like to see more ideas like this.

I really hope there's a model out there for OSS at least where folks can get paid.

I always wondered what if engineers had a $100/month budget to donate to OSS founders. It wouldn't cost an employer much and would be a good way to give back.

> there is at least one crucial feature missing: A LICENSE file that starts with "The MIT License (MIT)".

How much, monetarily, would you be willing to contribute to implement that feature?

Nothing, and that is exactly my point. I have nothing against monetization at all.

What I am saying is, that if the elevator pitch for the project you are trying to monetize is "this is a replacement for $MIT-LICENSED-PROJECT" you will have a hard time.

The question the elevator pitch should answer first is how much this brings to the table to make me consider paying instead of just using the permissively licensed project for free.

But it is a replacement. One project don't need to be a pareto obtimum to claim to be replacement, it's ok even if there are some drawbacks.

Having to pay is clearly one drawback for some, but if you save this money by not having to develop some features or getting some support, then it may be worth it.

Thank you to the commenters who highlighted the licensing. I would ordinarily be very interested in this premise but I don't want to risk unconsciously adapting ideas from what I read that could put me in tension with the terms or the author, and I can't use it directly.

Are you worried about copyright infringement?

I'm basically trying to avoid situations where I might be legally obligated to prove a negative, that I did not take ideas or code from a proprietary yet source-available project that is topically very adjacent to some projects I work on. I write with Rust, use Rowan and tree-sitter crates, create LSP implementation, and so on.

I don't have a retained lawyer or legal budget, so it's just risk aversion on my part to treat this project as radioactive and never once look inside. Same reason a lot of folks don't look at GPL'd things.

The burden of proof is on the accuser so you never need to worry about proving a negative.

Unless there are patents involved, and we’re just talking about possible copyright infringement, then you’re free to borrow the ideas just not the expressions. Which in practice means if you’re not blatantly copying then the proof of Structure, Similarity and Organization is prohibitively expense for the plaintiff.

Reply on news.ycombinator.com