The Future of Roy

101 views
Skip to first unread message

Andreas Hartmann

unread,
Dec 20, 2012, 5:13:08 PM12/20/12
to roy...@googlegroups.com
Hi everyone,

I'm currently evaluating options for making client-side development more
efficient and fun while producing code that is easy to understand and to
maintain. Roy is very promising. But before making serious investments
in establishing a Roy infrastructure at our company and encouraging my
fellow developers to spend their time familiarizing with it, I'd like to
get a better feeling whether Roy has the potential of becoming a
production-ready language in the mid-term future.

On [1], Brian describes the language as experimental. Does that imply
any fundamental conceptual limitations?

Is it possible to predict the future of Roy, compared to other
approaches to solve the JavaScript problem, e.g. Fay [2] which looks
promising to me but has the disadvantage of requiring a rather
heavyweight Haskell development environment?

There are some other subjects that I would like to address:

* Licensing: The current license seems to be the MIT license. Can we
assume that the compiler, tools etc. (excluding third-party tools of
course) will be distributed under this license (or similarly liberal
licenses, no copyleft) in the future?

* Community: Is there anybody here who is interested in establishing a
community around Roy? At the moment this forum seems to be the only
place to get help, and there doesn't seem to be much traffic. I guess
one issue is that the major work load is on Brian, who probably has the
best knowledge about the language internals, but who had happier things
to deal with lately. Congratulations by the way!

* Roadmap: Is there a public roadmap? How are proposals and
contributions handled?

* Commercial support: Is there anybody here who would be willing to
provide commercial support, including knowledge about the internals of
the compiler (bug fixing, extensions etc.)?

I'm very happy to contribute all the code I produce while improving my
Roy development infrastructure. Here's what I got so far: A fork of the
Roy project [3] which uses Optimist to parse the command line arguments
and allows to pass directories as sources, and an SBT plugin [4]. Both
are work in progress.

Any feedback would be very much appreciated. Thanks a lot in advance and
a Merry Christmas to everybody!


[1] http://roy.brianmckenna.org/
[2] http://fay-lang.org/
[3] https://github.com/devkat/roy
[4] https://github.com/devkat/roy-sbt-plugin

-- Andreas
Message has been deleted

Brian McKenna

unread,
Dec 23, 2012, 3:04:15 AM12/23/12
to roy...@googlegroups.com
On 21 December 2012 08:13, Andreas Hartmann <andreas....@gmail.com> wrote:
> Hi everyone,
>
> I'm currently evaluating options for making client-side development more
> efficient and fun while producing code that is easy to understand and to
> maintain. Roy is very promising. But before making serious investments in
> establishing a Roy infrastructure at our company and encouraging my fellow
> developers to spend their time familiarizing with it, I'd like to get a
> better feeling whether Roy has the potential of becoming a production-ready
> language in the mid-term future.

Depends on your definition of mid-term. I'm hoping to have Roy version
1.0 out early next year (hopefully by end of March).

> On [1], Brian describes the language as experimental. Does that imply any
> fundamental conceptual limitations?

I mean for it to imply that the language is still changing :)

> Is it possible to predict the future of Roy, compared to other approaches to
> solve the JavaScript problem, e.g. Fay [2] which looks promising to me but
> has the disadvantage of requiring a rather heavyweight Haskell development
> environment?

Fay is awesome.

The main problem I find is an impedance mismatch between Haskell and
JavaScript. Roy makes changes or sacrifices to lessen an impedance
mismatch.

A good example is laziness. Roy is strict so that it becomes more of a
1-to-1 mapping with JavaScript.

> There are some other subjects that I would like to address:
>
> * Licensing: The current license seems to be the MIT license. Can we assume
> that the compiler, tools etc. (excluding third-party tools of course) will
> be distributed under this license (or similarly liberal licenses, no
> copyleft) in the future?

My personal projects are always MIT licensed. I have absolutely no
intentions to change the license.

> * Community: Is there anybody here who is interested in establishing a
> community around Roy? At the moment this forum seems to be the only place to
> get help, and there doesn't seem to be much traffic. I guess one issue is
> that the major work load is on Brian, who probably has the best knowledge
> about the language internals, but who had happier things to deal with
> lately. Congratulations by the way!

Thanks :)

I've had a couple of people interested in getting into the internals
but it seems people get lost fairly quickly. My plans to address this:

* Make the code more functional to make it easier to understand
* Use more good JavaScript libraries to reduce code in the compiler
* Write some documentation/blog posts about the internals

> * Roadmap: Is there a public roadmap? How are proposals and contributions
> handled?

I've added my current roadmap to the wiki:

https://github.com/pufuwozu/roy/wiki/Roadmap

Proposals have mostly from through GitHub issues but this mailing list
is also a fine place :)

> * Commercial support: Is there anybody here who would be willing to provide
> commercial support, including knowledge about the internals of the compiler
> (bug fixing, extensions etc.)?

I can't but I'd happily train anyone who would like to do any
consulting work with Roy.

> I'm very happy to contribute all the code I produce while improving my Roy
> development infrastructure. Here's what I got so far: A fork of the Roy
> project [3] which uses Optimist to parse the command line arguments and
> allows to pass directories as sources, and an SBT plugin [4]. Both are work
> in progress.

Both are awesome!

Thanks heaps for your interest. Roy is still in progress and it is
still a bit of a pain to work with but I definitely have solid plans
to move it forward. It's taking a bit of time so all of your
improvements is much appreciated.

And I'm sorry for the slow replies, lots of stuff going on ;)
Reply all
Reply to author
Forward
0 new messages