Substance Composer Alpha-1 is ready

[Michael Aufreiter]

Hi everybody!

In the past months we were working on an editor, based on the Lens user experience. It's a work in progress, but shows how articles can be written in a web-based environment. We're planning to turn it into a full-fledged scientific editor in the near future. We're curious about your feedback! If you have ideas how this could evolve please let us know at in...@substance.io.

http://substance.io/composer-alpha1/

Cheers,

Michael

[Ian Mulvany]

So I'm thinking that when it get's to the point where we can export, a way to bootstrap people using this might be to support drag and drop of XML (as happens now with lens.substance.io), and then export to a zip archive that has an NLM XML file in there. 

That might be a specific use case, could that be a configuration option?

[Michael Aufreiter]

It's easily possible to import from NLM on the fly and start editing based on that input (with some restrictions, since we don't have support for inline images, tables, formulas etc.). It's also easy to generate valid NLM from the Substance/Lens document model. However the structure of that generated NLM will differ from the original input file and only contain the main contents (not the publisher-specific article meta data).

If you decide to import NLM once, and from then on only manipulate the content via the Composer you should be fine. There needs to be a single source of input, otherwise you're shooting yourself in the foot.

Good idea:

NLM -> [ JSON-based editing ] -> NLM'  -> NLM post processing (adding meta information etc., can be done again if NLM' changed)

Bad idea:

NLM -> [JSON-based editing]  -> NLM' -> [manual NLM editing] -> [JSON-based editing]

Well it's also okay to continue with manual editing of the NLM at some point. But then.. never try to edit with the Composer again, since you'll loose information.

-- Michael

[Ian Mulvany]

good points. In my mind we might end up with something like:

Word - [automate] --> EarlyNLMXML --> LensJSON -- [peer
review/comments manual] --> LensJSON -- [automate] --> TargetExportXML
-- [automate] --> PMCNLM -- [Final Corrections??] --> Publish

The tricky bit is with the final corrections. Obviously the substance
platform could go to "publish" instantly, but as a publisher I need to
get PMCNLM for getting the doc into pub med. If we could get the
corrections happening in Lens, and make sure that the final published
output was the same, or close enough, then there is a lot of
opportunity here.
- Ian

---
Head of Technology - eLife
Submit now - http://submit.elifesciences.org/
twitter: @IanMulvany

> --
> You received this message because you are subscribed to the Google Groups
> "elife lens" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to elife-lens+...@googlegroups.com.
> To post to this group, send email to elife...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

[Ivan Grubisic]

I was thinking of writing a python converter that would take the Lens JSON and create a more standardized NLM XML. That would be useful for oa-sandbox too for those that have their tools built around XML files already.

Might make that transition a little easier too.

[Gus Clemens]

Ivan, a tool like that would be quite useful for us (Landes Bioscience). We're moving various parts of our production pipeline to using JSON objects as storage, so being able to easily export NLM XML from a Lens JSON object would be a useful tool.