I can’t do any programming but I’m willing to help in anything I can, by giving suggestions, ideas, helping with design issues, translating it to spanish or even considering making a donation I’ve downloaded and tested your first release, and just added some bugs reports and feature requests to the tasklist. Keep this development going!

I have also came up with your post on LyX-devel, (in fact, I was googling with the exact terms “a corkboard implementation for lyx” so I found your both posts here and in lyx-devel). I scanned through both now, saving to read thoroughly for some time later as I am very much “loaded” now with papers to be submitted, etc. But please keep up the idea and know that you have supporters here:)

Best wishes

First, the features it supports in both semantic (LaTeX-space) and rich-text modes of operation, erroneously marked as being unsupported in the article:

* Footnotes: Supported in both semantic and rich-text modes. Footnotes are added in-line, as with LyX, making them more contextual and easy to maintain. They can be exported as either footnotes or endnotes depending upon user preference in the document compile option. In semantic mode, footnotes will be converted to LaTeX syntax, or look-alike XHTML cross-linking.
* Figures: Handled easy via drag and drop in rich-text mode. Place the image resource unit the document flow and it will be exported as RTF image and can be imported into compatible word processors. In semantic mode, usage of MMD syntax is required, and images will be collected and organised during Compile for the user.
* Legends: See above (not sure why identical technical features are listed separately in the article)
* Graphics: See above.
* Commentation: In semantic mode, Scrivener annotations will be marked as such by classed span elements which can be handled programmatically, or inserted as marginalia in LaTeX. In rich-text mode they can be formatted as Word pop-up comments, or embedded red text.

Now for features supported in semantic mode:

* Hyperlinks: Simple Markdown syntax both inline and cross-referenced.
* Cross-links: Use of anchor style ID linking the header titles, and header labelling is supported. By default it produces automatic labels based on camelCasedContent, but manual labels can be supplied and linked to from anywhere in the document. Arbitrary point-to-point cross referencing is not possible by default, though adding support for it would be fairly trivial.
* Cross-references: See above.

Front matter:
* Table of contents can be generated via LaTeX, naturally, and there is also a built-in script for adding an anchored ToC to XHTML documents.
* Abstract & Synopsis: Built-in support.

Back matter:

* Bibliography: Full BibTeX support is built-in.
* Glossary: Has basic support built-in.
* Appendix: Has basic support built-in.

Revision control:

While not built-in, Scrivener has full a SVN compatibility mode. If used with a front-end which can handle auto-add commits (Scrivener generates a lot if files based on the user’s usage of summary and notation features), full version control and collaboration is possible.

In short, the article is well written. I was a huge fan of LyX for many years, and I still do use it as a post-processor for Scrivener (the workflow is practically seamless since semantically defined data in Scrivener turns into LaTeX marked files which LyX can then convert and utilise). I have nothing against LyX, but I feel the author of the article failed to dig deeply enough into Scrivener’s more advanced features and a capabilities.

Yes, on the surface it supports a simple model, and for most authors this is plenty enough as their editors and publishers handle all the nitty gritty. For academic usage, there is enough support in rich-text mode for many, and those that require more power and are willing to dip into MultiMarkdown and LaTeX or some other format, the power is nearly limitless.

There are futher interface niceties in Scrivener that this plug-in proposal would not address. Its full-screen mode for distractionless writing is very elegant and completely “widgetless”, and the Edit Scrivenings mode which allows the user to selectively designate edit focus (as opposed to always being confronted with the entire text as in LyX), can be great for the creative mind which suffers under clutter and distraction.

LyX is fine (a bit more complicated than I need), but horrible at managing the structure of a document like Scrivener.

If you build this plugin, I’ll use it!