Address review 2-3 & 2-20: dot plot & interactive HTML

Merges #171

Response to reviewer 2 comment 20
Closes #149

Respond to Reviewer 2 comment 3 (interactive HTML)
Closes #132

content/02.main-text.md

@@ -152,9 +152,11 @@ Manubot uses [Pandoc](https://pandoc.org/) to convert manuscripts from Markdown
 Pandoc supports conversion between additional formats — such as LaTeX, AsciiDoc, EPUB, and JATS — offering Manubot users broad interoperability.
 [Journal Article Tag Suite](https://jats.nlm.nih.gov/) (JATS) is a standard XML format for scholarly articles that is used by publishers, archives, and text miners [@url:http://www.niso.org/standards/z39-96-2015/; @doi:10.6087/kcse.2014.1.99; @doi:10.1080/00987913.2012.10765464].
 Pandoc's JATS support provides an avenue to integrate Manubot with the larger JATS ecosystem.
-For now, the primary Manubot output is HTML intended to be viewed in a web browser.
-Authors generally use the PDF output for preprint or initial journal submissions.
-For journal submissions that require an editable source document, Manubot's DOCX output is usually sufficient.
+In the future, journals may begin accepting submissions in JATS.
+But for now, Manubot's DOCX output is usually sufficient for journal submissions that require an editable source document.
+Otherwise, authors generally use the PDF output (e.g. for preprint or initial journal submissions).
+The primary Manubot output is HTML intended to be viewed in a web browser.
+Accordingly, manuscripts natively support JavaScript and can thus include any web-based interactive visualization, such as those produced using [Vega-Lite](https://vega.github.io/vega-lite/), [Bokeh](https://bokeh.pydata.org), or [Plotly](https://plot.ly/) [@doi:10.1038/d41586-018-01322-9; @doi:10.1109/TVCG.2016.2599030].
 
 Manubot performs continuous publication: every update to a manuscript's source is automatically reflected in the online outputs.
 The approach uses continuous integration (CI) [@doi:10.1038/550143a; @doi:10.1038/nbt.3780; @doi:10.1101/344804], specifically via [Travis CI](https://travis-ci.org/), to monitor changes.
@@ -271,7 +273,7 @@ Manubot provides an ideal platform for perpetual reviews [@arxiv:1502.01329; @ta
 Concepts for the future of scholarly publishing extend beyond collaborative writing [@doi:10.22541/au.149693987.70506124; @tag:techblog-brown].
 Bookdown [@doi:10.1201/9781315204963] and Pandoc Scholar [@doi:10.7717/peerj-cs.112] both extend traditional Markdown to better support publishing.
 Examples of continuous integration to automate manuscript generation include [gh-publisher](https://github.com/ewanmellor/gh-publisher) and Continuous Publishing [@url:http://blog.martinfenner.org/2014/03/10/continuous-publishing/], which was used to produce the book Opening Science [@doi:10.1007/978-3-319-00026-8].
-Binder [@doi:10.25080/Majora-4af1f417-011], Distill journal articles [@doi:10.23915/distill.00010], Idyll [@tag:idyll], and Stencila [@tag:stencila] support manuscripts with interactive graphics and close integration with the underlying code.
+Binder [@doi:10.25080/Majora-4af1f417-011], Distill journal articles [@doi:10.23915/distill.00010], Idyll [@tag:idyll], and Stencila [@tag:stencila; @tag:elife-repro-article] support manuscripts with interactive graphics and close integration with the underlying code.
 As an open source project, Manubot can be extended to adopt best practices from these other emerging platforms.
 
 Several other open science efforts are GitHub-based like our collaborative writing process.

content/citation-tags.tsv

@@ -6,4 +6,5 @@ avasthi-preprints	doi:10.7554/eLife.38532
 steem-post	url:https://goo.gl/jGBrxE
 idyll	url:https://idyll.pub/post/announcing-idyll-pub-0a3eff0661df3446a915700d/
 stencila	url:https://elifesciences.org/labs/c496b8bb/stencila-an-office-suite-for-reproducible-research
+elife-repro-article	url:https://elifesciences.org/labs/ad58f08d/introducing-elife-s-first-computationally-reproducible-article
 livecoms	url:http://www.livecomsjournal.org/article/2031-why-we-need-the-living-journal-of-computational-molecular-science

content/response-to-reviewers.md

@@ -37,6 +37,18 @@ To that end, we've added three sentences in the Discussion section addressing th
 → I would insist a bit more on this part because it is a real strength if you consider a future integration with interactive figures (e.g. bokeh) and/or Jupyter notebooks.
 Being able to produce both a nice PDF and a HTML dynamic version would be a nice feature.
 
+The HTML-first workflow, where the PDF is created from the HTML, does permit incorporation of interactive web visualizations, as we now note in the manuscript (edits in [GH171](https://github.com/greenelab/meta-review/pull/171)).
+One downside is that it can [be difficult](https://github.com/manubot/rootstock/issues/192 "Increasing the density of PDF exports to reduce number of pages") to create heavily typeset and whitespace-optimized PDFs using CSS.
+However, Manubot envisions a future where the HTML view is the primary means of accessing the article.
+See [GH132](https://github.com/greenelab/meta-review/issues/132) for additional discussion.
+
+Regarding interactive notebooks, some projects (e.g. [_eLife_'s reproducible article](https://elifesciences.org/labs/ad58f08d/introducing-elife-s-first-computationally-reproducible-article)) are exploring incorporating customizable source code execution directly into a manuscript's HTML view.
+One challenge of this approach is that most analyses require substantial code, scattered across many layers of dependencies and inputs.
+Therefore, it is not clear that there is a manageable way to interact with all of this code directly within the HTML view of the manuscript.
+As an alternative approach, we have added Binder support to the Deep Review contribution analysis ([click to launch Binder](https://mybinder.org/v2/gh/greenelab/meta-review/binder?filepath=analyses/deep-review-contrib/02.contrib-viz.ipynb), implemented in [GH161](https://github.com/greenelab/meta-review/pull/161)).
+Binder allows readers to open the notebook used to generate Figure 2 in their browser.
+We found this approach allows users to modify source code at a scale that is more likely to be practical for complex data reanalyses.
+
 > In the discussion, it would be nice to have a vision on the future of manubot.
 What new features will be implemented?
 Do you plan to apply for a grant to further develop manubot? etc.
@@ -108,6 +120,12 @@ See for example https://swcarpentry.github.io/git-novice/01-basics/index.html
 > Figure .2 → This could be probably adapted as a raster plot where a dot is painted for each contribution (at the time of contribution).
 Line would be the same height and the legend would summarize the contribution.
 
+Based on this suggestion, we created an [alternative visualization](https://github.com/greenelab/meta-review/blob/6e7458dcb205a3d76daf25ee08236c24dc4de66e/analyses/deep-review-contrib/deep-review-contribution-dot.png) of Deep Review contributions that uses dots of varying sizes to show word counts by contributor on a monthly basis.
+While we thought the dot plot was a valuable supplement, there was not strong support for switching from the ridgeline plot to the dot plot for the primary manuscript visualization.
+However, we did update the figure caption to alert readers to the existence of alternative plots and that they can launch the source code notebook in Binder to create additional visualizations.
+
+See [GH149](https://github.com/greenelab/meta-review/issues/149) and [GH164](https://github.com/greenelab/meta-review/pull/164) for the discussion and analyses.
+
 > `GitHub and Travis CI. Fortunately, lock-in to these services is minimal, and several substitutes already exist.` → Can you name a few and how difficult it would be to migrate?
 
 ## Reviewer 3