Introducing Seed, an interactive software environment built on ASDF
Faré
fahree at gmail.com
Wed Nov 1 22:19:45 UTC 2017
Looks cool and interesting, and the code looks clean, too (though
maybe you should read and take to heart my "asdf best practices"
document currently at
https://github.com/fare/asdf/blob/master/doc/best_practices.md ).
I'm not sure how ASDF fits in it and that it deserves such prominent
part in your annoucement (unless you swap out that part when
announcing on other venues, which you should if you didn't yet). Where
are your ASDF extensions defined?
However I don't have time to look in detail right now. Still, what
technology did you use to produce those slides? They are neat.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org
Government is a disease masquerading as its own cure.
— Robert LeFevre (1911-1986)
On Wed, Nov 1, 2017 at 5:55 PM, Andrew Sengul <ml at imagegardenphoto.com> wrote:
> Hi everyone,
>
> I've created a new software environment based on Common Lisp and it'd be
> great to get your feedback. Called Seed, this system can be used to
> create Lisp programs and interact with them in many ways. It's a bit
> like a web-based IDE, but unlike most IDEs, it uses a visual programming
> language as its foundation.
>
> Seed displays Lisp code in the form of an interactive tree grid. It's
> possible to create customized interface elements within the tree grid;
> these elements are converted into Lisp code for compilation. A simple
> example is a color picker for use choosing colors inside a HTML
> template. Drop-downs and toggle switches are also easy to insert, and
> these elements can be composited to create more sophisticated
> interfaces. And unlike Emacs and other editors that are designed around
> text files, Seed is designed around editing ASDF systems.
>
> When using Seed, the top-level interactive elements are called "systems"
> - contrast to "buffers" in Emacs. Each Seed system is an ASDF system
> with particular extensions. Specifically, Seed systems all have a .seed
> file in their source directory that defines their behavior within the
> scope of the Seed environment.
>
> Check out this video to see Seed in action and a brief tutorial:
>
> https://vimeo.com/237947324
>
> This is a slide deck that goes into further detail:
>
> https://www.dropbox.com/s/ssjd05bmrfnzg5e/SeedPresentationTechnical.pdf?dl=0
>
> I don't know if Seed is the first general-purpose visual programming
> environment, but I'm sure its paradigm is among the most flexible. It is
> implemented as a Web application, with each atom and form within a Lisp
> program represented by a module manifested using the React framework for
> JavaScript. Because Seed's interface is implemented in React, the entire
> ecosystem of React components can be brought to bear when building
> custom software interfaces.
>
> Ready to try Seed out yourself? It's free software and the repo is here:
>
> https://github.com/phantomics/seed
>
> The installation instructions start here:
>
> https://github.com/phantomics/seed#the-first-step-installing-and-running-seed
>
> Read the usage tutorial here:
>
> https://github.com/phantomics/seed/wiki/Introductory-Tutorial
>
> So far, only SBCL has been tested. This is alpha software and you will
> encounter bugs aplenty.
>
> I envision Seed as something that could one day encompass most of the
> desktop productivity functions that people now use office apps for. The
> data-driven nature of the interface will make it easy to mash up and
> modify these tools to design custom workflows.With Lisp's homoiconicity
> it may even be possible to develop machine learning algorithms to tailor
> Seed systems to the jobs they're intended to do. Emacs has always been
> seen as the flagship of Lisp-based environments, but because of its
> byzantine interface and its tight coupling to the text buffer as the
> main interactive unit, I can't see a future where Emacs is usable by
> regular people for desktop productivity. I think Seed may be able to get
> there.
>
> Let me know what you think of Seed, particularly as an ASDF use case. Best,
>
> Andrew Sengul
>
>
More information about the asdf-devel
mailing list