TidyDensity
A distribution catalogue that grows by one family at a time, and rarely breaks anything.
A side-by-side editorial comparison of collinear and skylight — release velocity, themes, recent moves, and the top alternatives to consider.
collinear has broken its API twice to stop making the user pick thresholds.
collinear removes multicollinearity from predictor sets through pairwise correlation and VIF filtering, with a preference order deciding which variable survives each conflict. Two major versions in thirteen months each rewrote the interface: 2.0.0 extended every function to any combination of categorical and numeric responses and predictors, and 3.0.0 moved to multiple responses, restructured the output into classed objects, and made both filtering thresholds adaptive by default. Version 3.0.1 is the first release since that is purely repair.
A frozen astronomical model quietly became the inner loop of its sibling's optimizer.
skylight returns sun and moon illuminance, azimuth and altitude for a given date, time and location, implemented as a near-verbatim transcription of a 1987 US Naval Observatory circular. The model formulation has not changed since the initial 2022 release and the author states so explicitly. Everything shipped since has been packaging, citation and speed: v1.3 moved the main routine from R to C++, and v1.4 removed a parameter check that was flooding the console with messages.
collinear removes multicollinearity from predictor sets through pairwise correlation and VIF filtering, with a preference order deciding which variable survives each conflict. Two major versions in thirteen months each rewrote the interface: 2.0.0 extended every function to any combination of categorical and numeric responses and predictors, and 3.0.0 moved to multiple responses, restructured the output into classed objects, and made both filtering thresholds adaptive by default. Version 3.0.1 is the first release since that is purely repair.
The through-line is removing decisions the user was never well placed to make. Preference-order functions were renamed twice — first onto a metric-and-model scheme in 2.0.0, then onto a response-type scheme in 3.0.0 — and f_auto() picks one when none is given; target encoding went from automatic to opt-in; max_cor and max_vif now default to NULL and trigger a data-driven threshold derived from the 75th percentile of pairwise correlations through a sigmoid and a fitted correlation-to-VIF mapping. Each change is defensible and each one broke callers, which is the cost of this approach.
3.0.1 moved the example datasets out into a separate spatialData package and fixed four crashes rather than adding anything, so the next release is most likely more consolidation on the 3.0 surface than a fourth interface.
skylight returns sun and moon illuminance, azimuth and altitude for a given date, time and location, implemented as a near-verbatim transcription of a 1987 US Naval Observatory circular. The model formulation has not changed since the initial 2022 release and the author states so explicitly. Everything shipped since has been packaging, citation and speed: v1.3 moved the main routine from R to C++, and v1.4 removed a parameter check that was flooding the console with messages.
This is a reference implementation of a published algorithm rather than a product accumulating features, and it is being maintained that way. The movement that does occur is driven from downstream: the C++ port was written for the inverse-modelling loop in the sibling skytrackr package, which calls skylight repeatedly during optimization. That reframes skylight from a standalone calculator into the compute kernel another package's fitting routine depends on.
With the model formulation deliberately fixed and the C++ path already in place, the next release is most likely another small maintenance fix. The entries give no indication of planned new capability.
Other Analytics products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either collinear or skylight.
A distribution catalogue that grows by one family at a time, and rarely breaks anything.
College football's open data client hit v2 — and now reports how many API calls you have left.
The USA phenology data client rebuilt its entire stack and stopped handing users -9999 as a number.
GeneNMF rebuilt how it derives meta-programs, changing every result it had produced.
Publication-ready psychology tables and plots, tracking APA style as closely as the software allows.
A spatial-statistics utility package exists to be depended on, and is built accordingly.
See all collinear alternatives → · See all skylight alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. collinear and skylight are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. collinear and skylight are shipping at a similar cadence (velocity 0.0 vs 0.0, both within Sparkpulse's "active" band). For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
Top collinear alternatives in Analytics are ranked by recent ship velocity. Browse the "collinear alternatives" section above for the current picks, or visit /alternatives/collinear for the full list with editorial commentary on each.
Top skylight alternatives in Analytics are ranked by recent ship velocity. Browse the "skylight alternatives" section above for the current picks, or visit /alternatives/skylight for the full list with editorial commentary on each.