collinear
collinear has broken its API twice to stop making the user pick thresholds.
A side-by-side editorial comparison of minimaxapprox and soilDBdata — release velocity, themes, recent moves, and the top alternatives to consider.
minimaxapprox found its Remez exchange had been silently returning the wrong minimax.
minimaxapprox computes minimax polynomial and rational approximations to functions via the Remez exchange algorithm. Version 0.6.0 is a correctness release of unusual depth: the exchange was redesigned after two structural defects were found that let it converge to a reference-local fixed point that is not the global minimax, with no warning. The worked example in the notes has atan on [0,3] at degree 3 reporting an expected error 1.8 times smaller than the returned approximation's true maximum error.
soilDBdata exists so soilDB's tests can run without a NASIS connection.
soilDBdata is a data-only package supplying NASIS and gSSURGO sample datasets as .sqlite assets, installed separately by soilDB's GitHub Actions so unit tests that would otherwise need database access can run. It began as a proof of concept carrying MT663 pedon and component tables used in soil survey coursework, and its most recent release adds a Marshall Islands FY26 gSSURGO dataset. Releases are infrequent and driven by what the parent package needs to test.
minimaxapprox computes minimax polynomial and rational approximations to functions via the Remez exchange algorithm. Version 0.6.0 is a correctness release of unusual depth: the exchange was redesigned after two structural defects were found that let it converge to a reference-local fixed point that is not the global minimax, with no warning. The worked example in the notes has atan on [0,3] at degree 3 reporting an expected error 1.8 times smaller than the returned approximation's true maximum error.
The redesign changes what drives the algorithm: roots are now located across the whole interval from an oversampled grid rather than only between consecutive reference points, each sign region takes its search direction from the error's own sign instead of a mechanically alternated schedule, and endpoints are always retained as exchange candidates. Around that sit a cluster of fixes with the same signature — a stagnation check tested in only one direction, a missing abs() in a coefficient test, ztol applied on the monomial scale when Chebyshev was requested — each one a silent wrong answer rather than a crash. Conditioning on ranges far from [-1,1] was also addressed, at the cost of a breaking change.
Nearly every fix here came from auditing places where the package trusted its own structure instead of measuring the error curve, so the next release most likely continues that audit into the rational-approximation path rather than adding features.
soilDBdata is a data-only package supplying NASIS and gSSURGO sample datasets as .sqlite assets, installed separately by soilDB's GitHub Actions so unit tests that would otherwise need database access can run. It began as a proof of concept carrying MT663 pedon and component tables used in soil survey coursework, and its most recent release adds a Marshall Islands FY26 gSSURGO dataset. Releases are infrequent and driven by what the parent package needs to test.
Development follows soilDB rather than leading it: assets get bumped when a soilDB version changes, and purpose lists are updated when soilDB adds a table. The one release that changed what testing is possible was v0.1.1, which added selected-set _View_1 tables alongside whole tables so both SS=TRUE and SS=FALSE code paths could be exercised. Four-year gaps between releases are normal here and do not indicate abandonment — a fixture package only needs to move when the fixtures go stale.
The recent addition is a new geography rather than a new table structure, so further releases most likely continue broadening dataset coverage as soilDB gains regions to test against.
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 minimaxapprox or soilDBdata.
collinear has broken its API twice to stop making the user pick thresholds.
nert put fourteen TERN datasets behind one dispatcher and called it stable.
scTypeEval judges single-cell annotations without needing a ground truth to judge them against.
medrobust made its partial-identification bounds usable by giving them confidence intervals.
probmed went from one probabilistic effect size to a family of them in sixteen days.
fluxnet keeps swapping its own plumbing — first a Python downloader, now a DuckDB backend.
See all minimaxapprox alternatives → · See all soilDBdata alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
Both compete on the same themes — r package — within Analytics. minimaxapprox and soilDBdata 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. minimaxapprox and soilDBdata 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 minimaxapprox alternatives in Analytics are ranked by recent ship velocity. Browse the "minimaxapprox alternatives" section above for the current picks, or visit /alternatives/minimaxapprox for the full list with editorial commentary on each.
Top soilDBdata alternatives in Analytics are ranked by recent ship velocity. Browse the "soilDBdata alternatives" section above for the current picks, or visit /alternatives/soildbdata for the full list with editorial commentary on each.