added some thoughts

paper/paper.bib

@@ -1,21 +1,21 @@
 
-@inproceedings{bastian_gephi:_2009,
+@inproceedings{gephi,
 	title = {Gephi: {An} {Open} {Source} {Software} for {Exploring} and {Manipulating} {Networks}},
 	url = {http://www.aaai.org/ocs/index.php/ICWSM/09/paper/view/154},
 	abstract = {Jacomy},
 	author = {Bastian, Mathieu and Heymann, Sebastien and Jacomy, Mathieu},
 	year = {2009}
 }
 
-@article{wapman_webweb:_2019,
+@article{webweb,
 	title = {webweb: a tool for creating, displaying, and sharing interactive network visualizations on the web},
 	shorttitle = {webweb},
 	journal = {Journal of Open Source Software},
 	author = {Wapman, Hunter K and Larremore, Daniel B},
 	year = {2019}
 }
 
-@article{hunter_matplotlib:_2007,
+@article{matplotlib,
 	title = {Matplotlib: {A} 2D graphics environment},
 	volume = {9},
 	doi = {10.1109/MCSE.2007.55},
@@ -27,7 +27,7 @@ @article{hunter_matplotlib:_2007
 	pages = {90--95}
 }
 
-@article{bostock_d3:_2011,
+@article{d3,
 	title = {D3: {Data}-{Driven} {Documents}},
 	volume = {17},
 	number = {12},
@@ -36,7 +36,7 @@ @article{bostock_d3:_2011
 	year = {2011}
 }
 
-@book{aric_a._hagberg_exploring_2008,
+@book{networkx,
 	address = {Pasadena, CA USA},
 	title = {Exploring network structure, dynamics, and function using {NetworkX}},
 	volume = {7},
@@ -46,13 +46,12 @@ @book{aric_a._hagberg_exploring_2008
 	year = {2008}
 }
 
-@inproceedings{ellson_graphviz_2001,
+@inproceedings{graphviz,
 	title = {Graphviz {\textemdash} open source graph drawing tools},
-	abstract = {Graphviz is a heterogeneous collection of graph drawing tools containing batch layout programs (dot, neato, fdp, twopi); a platform for incremental layout (Dynagraph); customizable graph editors (dotty, Grappa); a server for including graphs in Web pages (WebDot); support for graphs as COM objects (Montage); utility programs useful in graph visualization; and libraries for attributed graphs. The software is available under an Open Source license. The article[1] provides a detailed description of the package. The Graphviz software began with a precursor of dot in 1988, followed by neato in the early 90{\textquoteright}s. The features expanded greatly over the years, driven by user request. Graphviz became Open Source in 2000, and was recently distributed on about 500,000 CDROMs as an add-on package for the SUSE Linux release, and is redistributed by Debian, Mandrake, SourceForge, and soon OpenBSD. 2 Areas of Application Thanks to the variety of components available and its open, {\textquotedblleft}toolkit {\textquotedblright} design, Graphviz supports a wide variety of applications. The foremost application is probably presentation layouts, such as including graphs in papers. As stream processors, the Graphviz tools can be used as co-processes with interactive components to provide dynamic layouts for debuggers, process monitors, program analysis software, etc. Graphviz tools have been adopted as a visualization service by the W3C Resource Description Framework},
 	booktitle = {Lecture {Notes} in {Computer} {Science}},
 	publisher = {Springer-Verlag},
 	author = {Ellson, John and Gansner, Emden and Koutsofios, Lefteris and North, Stephen and Woodhull, Gordon and Description, Short and Technologies, Lucent},
 	year = {2001},
 	pages = {483--484},
 	file = {Citeseer - Full Text PDF:/Users/bfmaier/Zotero/storage/LS2KQZ99/Ellson et al. - 2001 - Graphviz {\textemdash} open source graph drawing tools.pdf:application/pdf;Citeseer - Snapshot:/Users/bfmaier/Zotero/storage/IXILUSLA/summary.html:text/html}
-}
+}

paper/paper.md

@@ -0,0 +1,72 @@
+---
+title: Netwulf: Interactive visualization of networks in Python # maybe you have a better title
+tags:
+  - Python
+  - JavaScript
+  - networks
+  - visualization
+  - interactive
+authors:
+  - name: Ulf Aslak
+    orcid: 0000-0000-0000-0000
+    affiliation: "1, 2" # (Multiple affiliations must be quoted)
+  - name: Benjamin F. Maier
+    orcid: 0000-0001-7414-8823
+    affiliation: "3, 4"
+affiliations:
+ - name : Centre for Social Data Science, University of Copenhagen, DK-1353 København K
+   index: 1
+ - name : DTU Compute, Technical University of Denmark, DK-2800 Kgs. Lyngby
+   index: 2
+ - name: Robert Koch-Institute, Nordufer 20, D-13353 Berlin
+   index: 3
+ - name: Department of Physics, Humboldt-University of Berlin, Newtonstr. 15, D-12489 Berlin
+   index: 4
+date: 20 April 2018
+bibliography: paper.bib
+---
+
+# Summary
+
+I'm just collecting some thoughts here
+
+* visualizing networks is an important part of communicating their structure
+* there already exist several tools to explore/visualize networks and their structure and properties.
+* "networkx" [@networkx] is the most commonly used Python package for network analysis
+* it provides visualization functions which are either slow or inconvenient as they
+  rely on third-party dependencies like "graphviz" [@graphviz], and use a
+  a variety of parameters. These parameters then have to be manually scanned, 
+  reproducing a new visualization for every changed parameter.
+* Gephi [@gephi] is mostly focused on both interactive analysis and visualization.
+  However, networks must be saved and loaded as separate files. Java has to be installed.
+  an automated view of a network which was previously manipulated in Python is not possbile
+* Webweb [@webweb] provides an interactive visualization for Python network objects based
+  on the force layout of d3.js [@d3], however
+  its main purpose is exploration of network features
+  and single-time visualizations:
+  the stylized network and visualization parameters may not be saved to reproduce
+  Further, a variety of styling parameters are only accessible through its API
+  which again may provide difficulties when trying to find the optimal visualization.
+* Netwulf provides a simple interactive interface to stylize a network which was previously
+  manipulated in Python. 
+* Its philosophy is that network manipulation should be done programmatically (i.e. in Python)
+  but the efficient generation of a visually pleasing network representation should be
+  done manually
+* Its main purpose is to easily generate both a network visualization
+  and find the optimal visualization parameters to reproduce the visualization. To this end, both are 
+  saved to objects in the Python instance running Netwulf.
+* It further provides functionality to
+  redraw the visualization in Python using matplotlib [@matplotlib] to save it as a vector graphic
+  or image in arbitrary resolution.
+* more description
+
+
+# Figures
+
+![Interactive visualization of a modular network in Netwulf.](random_partition_graph.png)
+
+# Acknowledgements
+
+Both authors contributed equally to the software, documentation, and manuscript. B. F. M. is financially supported as an *Add-On Fellow for Interdisciplinary Life Science* by the Joachim Herz Stiftung.
+
+# References