|
| 1 | +.. image:: ./images/gsas2.png |
| 2 | + :scale: 25 % |
| 3 | + :alt: GSAS-II logo |
| 4 | + :align: right |
| 5 | + |
| 6 | +GSAS-II Citations |
| 7 | +==================================== |
| 8 | + |
| 9 | +.. index:: Impact |
| 10 | + |
| 11 | +The GSAS-II usage statistics must be considered in the context of the software that it replaced, GSAS/EXPGUI. It is difficult to catalog citations to GSAS, since it was |
| 12 | +only published as a Los Alamos document report starting in 1985. However, EXPGUI, a graphical user interface to GSAS, was published and captures a large fraction of the GSAS user base. While GSAS-II was started as a project in ~2008, it was far from complete until around 2013 when first published: |
| 13 | + |
| 14 | + Toby, B. H., & Von Dreele, R. B. (2013). "GSAS-II: the genesis of a modern open-source all purpose crystallography software package". *Journal of Applied Crystallography*, **46**\(2), 544-549. `doi: 10.1107/S0021889813003531 <https://doi.org/10.1107/S0021889813003531>`_ |
| 15 | + |
| 16 | +Citations to this paper and the 2001 paper by B.H. Toby are tracked by the Web of Science and are our best way to track use of GSAS/EXPGUI and GSAS-II. Google Scholar citation rates are significantly higher, but their accuracy is unknown. |
| 17 | +While GSAS-II has been recommended as a replacement for GSAS/EXPGUI since ~2015, when GSAS-II incorporated pretty much all of the GSAS/EXPGUI capabilities judged still relevant, but for quite a few years after that, both EXPGUI and GSAS-II usage continued to grow, as shown in the citation rate plot below: |
| 18 | + |
| 19 | +.. image:: ./images/EG_Cites.pdf |
| 20 | + :alt: Citations of EXPGUI and GSAS-II |
| 21 | + :align: right |
| 22 | + |
| 23 | +Only in the past few years has EXPGUI use started to fall. GSAS-II use continues to grow and at a rate considerably higher than the fall-off in EXPGUI use. |
| 24 | + |
| 25 | +GSAS-II Code Statistics |
| 26 | +==================================== |
| 27 | + |
| 28 | +.. index:: Code statistics |
| 29 | + |
| 30 | +As of October 2025, GSAS-II codebase contains: |
| 31 | + |
| 32 | + * 170,000 lines of Python code, of which 65,000 are used for the GUI. |
| 33 | + * 6,000 lines of Fortran code |
| 34 | + * 450 lines of C code |
| 35 | + |
| 36 | + Some of the compiled code (Fortran and C) provides computational kernels for that would be too slow if completed in Python. The majority of the Fortran code is a group theory analysis routine developed by Allen Larson to derive symmetry from Hermann-Mauguin space group symbols (short or long and including non-standard settings) developed for the NRCVAX program. This code was also used in GSAS, but has been updated to correct for errors. |
| 37 | + |
| 38 | +The documentation material consists of: |
| 39 | + |
| 40 | + * 14,000 words of web documentation |
| 41 | + * 50,000 words of context-specific help (~100 pages formatted) |
| 42 | + * 62 tutorials (140,000 lines of HTML and 500 words of MarkDown) |
| 43 | + * ~450 pages (as .pdf) of code documentation, generated by Sphinx from the Python source code |
| 44 | + |
| 45 | + |
| 46 | +Vendored (externally produced, but distributed with GSAS-II) utilities included are composed of: |
| 47 | + |
| 48 | + * 4,000 lines of Python code |
| 49 | + * 10,000 lines of Fortran code |
| 50 | + * 100 lines of Cython |
0 commit comments