Iris Optics ThinFilm — third-party software and reference data Reviewed 2026-09-26 These notices apply to the named components and datasets. They do not assign a blanket license to Jimmy's application, company branding, or user designs. Local material files supplied by a visitor remain that visitor's responsibility and are processed in the browser. They are not uploaded or published by this app. REFRACTIVEINDEX.INFO The public material collection includes the historical TFTB conversions of refractiveindex.info datasets and the reviewed analytic model collection. Each material retains its original reference, conditions, wavelength domain, and source identity where available. Unsupported models are not advertised as usable measurements. The preserved full legacy material corpus is not published. Mikhail N. Polyanskiy's refractiveindex.info database is made available under CC0 1.0 Universal. The full dedication is in licenses/RII_LICENSE.txt. Source: https://github.com/polyanskiy/refractiveindex.info-database Reference: Polyanskiy, M. N. Refractiveindex.info database of optical constants. Scientific Data 11, 94 (2024). https://doi.org/10.1038/s41597-023-02898-2 Original publication references remain associated with individual materials. Analytic optical models preserve their scientific citations and NIST/TFTB source notices. TFTB was assembled by Ulf Griesmann, NIST, and identifies most of its software as public domain. The public numerical implementation uses explicitly parsed coefficients and reviewed equations; it does not execute the historical MATLAB files or distribute third-party MATLAB optimizers. Analytic zero extinction is an approximation where the source specifies a lossless model; it is not a measured extinction coefficient. CIE COLORIMETRIC DATA Creator/publisher: International Commission on Illumination (CIE), Vienna, AT. Official dataset metadata retained in reference-provenance/ specifies Creative Commons Attribution-ShareAlike 4.0 International: https://creativecommons.org/licenses/by-sa/4.0/ The runtime's tabular adaptations are distributed under that same data license. This data license does not license the application as a whole. No endorsement by the CIE is asserted. Sources: - CIE 1931 colour-matching functions, 2 degree observer (2019): https://doi.org/10.25039/CIE.DS.xvudnb9b - CIE 1964 colour-matching functions, 10 degree observer (2019): https://doi.org/10.25039/CIE.DS.sqksu2n5 - CIE standard illuminant A, 1 nm (2018): https://doi.org/10.25039/CIE.DS.8jsxjrsn - CIE standard illuminant D50 (2022): https://doi.org/10.25039/CIE.DS.etgmuqt5 - CIE standard illuminant D65 (2019): https://doi.org/10.25039/CIE.DS.hjfjmt59 - CIE indoor daylight illuminant ID50 (2009): https://doi.org/10.25039/CIE.DS.r4gcnrzc - CIE indoor daylight illuminant ID65 (2009): https://doi.org/10.25039/CIE.DS.bd53qdqk - Relative spectral power distributions of typical fluorescent lamps (2018): https://doi.org/10.25039/CIE.DS.ukaymjdn Adaptations retained from the historical thin-film implementation: - Tables use 5 nm samples and whitespace-separated text. The observer tables cover 380–780 nm and use six decimal places. The runtime interpolates within this declared visible range; the original 1 nm observer CSVs have wider bounds. - Illuminants A, D65 and fluorescent F1–F4 match the official sampled values. D50 differs by at most 0.001 relative-power unit in the retained historical table (at 555 nm: 101.159 versus the current reference's 101.158). Observer differences are at most approximately 0.0000005. - The historical ID50/ID65 tables retain an unnormalized window-transmission scale. Multiplication by 1.1225913224532718 sets their 560 nm value to 100, matching official CIE tables within their 0.01 rounding step. This constant cancels in the implemented relative XYZ/luminance normalization. - CIE 1964 z-bar entries shown as blank/NaN in the official reference beyond its listed nonzero range correspond to zeros in the historical runtime table. The comparison records all such substitutions explicitly. reference-provenance/cie-adaptations.json records dataset identifiers, official checksums, legacy source hashes and exact numerical comparison limits. The official CSV MD5 and SHA-256 values were verified against their CIE metadata. The current application interface uses D65 and the CIE 1931 observer for its color display; other retained tables remain available to the numerical core. BROWSER SCIENTIFIC RUNTIME Pyodide 314.0.7, Mozilla Public License 2.0. Full license: licenses/PYODIDE-LICENSE.txt Source and corresponding release: https://github.com/pyodide/pyodide Distribution: https://cdn.jsdelivr.net/pyodide/v314.0.7/full/ The downloaded runtime assets are bundled unchanged and served from this site. Application-specific bridge and numerical code are separate modules. CPython 3.14.2, Python Software Foundation License Version 2 and retained historical terms. Full license: licenses/PYTHON-LICENSE.txt Corresponding source: https://github.com/python/cpython/tree/v3.14.2 This application uses the Pyodide WebAssembly build without modifying CPython. NumPy 2.4.6, SciPy 1.18.0 and pandas 3.0.2: BSD licenses and individual third-party notices. python-dateutil 2.9.0.post0: Apache/BSD terms. pytz 2026.1.post1 and six 1.17.0: MIT licenses. Full license texts and bundled subcomponent notices extracted from these exact distributed wheels appear in licenses/NUMERICAL_DEPENDENCIES_NOTICES.txt and remain inside the wheels. PLOTTING Plotly.js 3.6.0, copyright Plotly Technologies Inc., MIT license. The source bundle retains its original copyright banner. Full license: licenses/PLOTLY-LICENSE.txt Corresponding source: https://github.com/plotly/plotly.js/tree/v3.6.0 The references and passing numerical checks identify data/model provenance; they do not establish calibration or optical-constant accuracy for a particular deposition process, material sample, coating run or instrument.