JOSS

.github/workflows/draft-pdf.yml

@@ -0,0 +1,23 @@
+on: [push]
+
+jobs:
+  paper:
+    runs-on: ubuntu-latest
+    name: Paper Draft
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+      - name: Build draft PDF
+        uses: openjournals/openjournals-draft-action@master
+        with:
+          journal: joss
+          # This should be the path to the paper within your repo.
+          paper-path: joss/paper.md
+      - name: Upload
+        uses: actions/upload-artifact@v1
+        with:
+          name: paper
+          # This is the output path where Pandoc will write the compiled
+          # PDF. Note, this should be the same directory as the input
+          # paper.md
+          path: joss/paper.pdf

joss/paper.bib

@@ -0,0 +1,94 @@
+@techreport{GSW-m,
+  title={Getting started with {TEOS-10} and the {G}ibbs {S}eawater ({GSW}) oceanographic toolbox},
+  author={McDougall, Trevor J and Barker, Paul M},
+  publisher={{SCOR}/{IAPSO}},
+  series={WG 127},
+  pages={1--28},
+  year={2011},
+  note={ISBN 978-0-646-55621-5}
+}
+
+@article{GSW-Python,
+  title={TEOS-10/GSW-Python: v3.4.1.post0 (v3.4.1.post0)},
+  author={Eric Firing and Filipe and Andrew Barna and Ryan Abernathey},
+  year={2021},
+  doi={10.5281/zenodo.5214122}
+}
+
+@article{MILLERO200850,
+title = {The composition of Standard Seawater and the definition of the Reference-Composition Salinity Scale},
+journal = {Deep Sea Research Part I: Oceanographic Research Papers},
+volume = {55},
+number = {1},
+pages = {50-72},
+year = {2008},
+issn = {0967-0637},
+doi = {10.1016/j.dsr.2007.10.001},
+url = {https://www.sciencedirect.com/science/article/pii/S0967063707002282},
+author = {Frank J. Millero and Rainer Feistel and Daniel G. Wright and Trevor J. McDougall},
+}
+
+@article{feistel2003new,
+  title={A new extended Gibbs thermodynamic potential of seawater},
+  author={Feistel, Rainer},
+  journal={Progress in Oceanography},
+  volume={58},
+  number={1},
+  pages={43--114},
+  year={2003},
+  publisher={Elsevier},
+  doi = {10.1016/S0079-6611(03)00088-0}
+}
+
+@article{feistel2008gibbs,
+  title={A Gibbs function for seawater thermodynamics for- 6 to 80 C and salinity up to 120 g kg--1},
+  author={Feistel, Rainer},
+  journal={Deep Sea Research Part I: Oceanographic Research Papers},
+  volume={55},
+  number={12},
+  pages={1639--1671},
+  year={2008},
+  publisher={Elsevier}
+}
+
+@techreport{TEOS-10,
+  title={The international thermodynamic equation of seawater – 2010: Calculation and use of thermodynamic properties},
+  author={{IOC, SCOR, and IAPSO}},
+  edition={June 2015},
+  series={Intergovernmental Oceanographic Commission, Manuals and Guides No. 56},
+  year={2010},
+  publisher={UNESCO},
+  note={196 pp.}
+}
+
+@manual{GSW-Julia,
+  title={TEOS-10/GibbsSeaWater.jl},
+  author={Alexander Barth and et. al.},
+  year={2020},
+  url={https://github.com/TEOS-10/GibbsSeaWater.jl},
+  note={GitHub Repository.}
+}
+
+@manual{GSW-R,
+  title={TEOS-10/GSW-R},
+  author={Dan Kelley and Clark Richards},
+  year={2022},
+  url={https://github.com/TEOS-10/GSW-R},
+  note={GitHub Repository.}
+}
+
+@manual{GSW-FORTRAN,
+  title={TEOS-10/GSW-FORTRAN},
+  author={David Jackett and Paul Barker and Glenn Hyland},
+  year={2017},
+  url={https://github.com/TEOS-10/GSW-Fortran},
+  note={GitHub Repository.}
+}
+
+@manual{GSW-C,
+  title={},
+  author={Frank Delahoyde and Eric Firing and et. al.},
+  year={2022},
+  url={https://github.com/TEOS-10/GSW-C},
+  note={GitHub Repository.}
+}

joss/paper.md

@@ -0,0 +1,71 @@
+---
+title: 'Gibbs Sea Water Oceanographic Toolbox of TEOS-10 implemented in Rust'
+tags:
+  - Rust
+  - oceanography
+authors:
+  - name: Guilherme P. Castelao
+    orcid: 0000-0002-6765-0708
+    affiliation: 1 # (Multiple affiliations must be quoted)
+  - name: Luiz Irber
+    orcid: 0000-0003-4371-9659
+    affiliation: 2
+affiliations:
+ - name: Scripps Institution of Oceanography
+   index: 1
+ - name: UC Davis
+   index: 2
+date: 2 October 2022
+bibliography: paper.bib
+---
+
+# Summary
+
+The Gibbs Seawater Toolbox (GSW) is a key software for Oceanography since it
+provides consistent thermodynamic properties of seawater, conversions, and
+other utilities. GSW has been adopted since 2009 by the Intergovernmental
+Oceanographic Commission as the official description of seawater. Although it
+is available in several computer languages, most implementations, such as
+Python [@GSW-Python], Julia [@GSW-Julia], and R [@GSW-R], are wrappers around the C library [@GSW-C].
+
+Here we introduce a version of GSW implemented in pure Rust (GSW-rs),
+initially developed for inclusion in microcontroller firmware to support
+autonomous decisions and onboard Machine Learning. The same implementation
+also works on regular computers and can seamlessly replace GSW-C on apps and
+libraries by maintaining compatibility with the GSW-C Foreign Function
+Interface (FFI). Thanks to zero-cost abstraction, GSW-rs does not impose
+performance and readability trade-off, allowing it to be written for clear
+understanding and closer to the original scientific publications. Therefore,
+it is easier to verify and maintain. Another key aspect is the support for
+testing. GSW-rs is subject to unit tests as well as validation against the
+reference dataset from TEOS-10, allowing for consistent development through
+continuous integration.
+
+Modern oceanography strongly relies on autonomous platforms - such as Argo
+floats, Spray underwater gliders, and Saildrones - to provide sustained
+observations. Software robustness and performance are critical requirements
+for these platforms to operate with low energy budgets and up to several years
+in a single deployment, making Rust an optimal language for this task. At the
+same time, the expanding cloud infrastructure can give the illusion of
+infinite computing, but convenient program languages such as Python must rely
+on high-performance languages in the backend to optimize bottlenecks. A Rust
+implementation of GSW allows sustainable and efficient progress, from embedded
+to high-performance computing.
+
+# Statement of Need
+
+While GSW is already implemented in several languages, there is no uniformity
+among those. The Matlab implementation (GSW-m) [@GSW-m] is the most
+complete [see Appendix N from @TEOS-10] and up to date, but it is based
+on a commercial language, restricting its use. Several other implementations,
+including those for Julia, Python, and R, are wrappers around the
+C implementation (GSW-C), which is hence the actual foundation for the
+alternative Open Source family of solutions. Although it is powerful, C lacks
+some features and conveniences of modern languages. Here we present an
+alternative using Rust language, resulting in comparable performance to GSW-C,
+while providing an efficient framework that accelerates the development effort
+and minimizes errors. For embedded systems, GSW-rs is a requirement for a
+pure Rust firmware able to make sense of sensor measurements in real time.
+In addition, for any application, GSW-rs provides the option to conform more closely to GSW-m than does the present GSW-C implementation. Function coverage, however, does not yet match either GSW-C or GSW-m.
+
+# References