HydroPortailStats
France's national flood statistics, ported out of Fortran and into R.
A side-by-side editorial comparison of estimatr and mize — release velocity, themes, recent moves, and the top alternatives to consider.
Fast design-based estimators for experiments, coasting on CRAN patches.
estimatr provides the design-based regression estimators the DeclareDesign ecosystem is built on — robust and cluster-robust standard errors, blocked and clustered randomization inference — implemented for speed rather than generality. The last three releases carry no substantive notes: each is a merge commit for a CRAN patch, one of them accompanied by a typo fix.
A numerical optimization toolkit that has been feature-complete and quiet since 2017
mize provides a configurable interface to unconstrained numerical optimization methods - line searches, gradient descent variants, quasi-Newton updates - usable both as a one-shot call and as a stepwise iterator. Its functional surface has not changed since the initial CRAN release in July 2017. Every release since has been a patch: R-devel compatibility, line search edge cases, and in 2026's 0.2.5 the removal of LazyData from DESCRIPTION plus deletion of some flaky tests.
estimatr provides the design-based regression estimators the DeclareDesign ecosystem is built on — robust and cluster-robust standard errors, blocked and clustered randomization inference — implemented for speed rather than generality. The last three releases carry no substantive notes: each is a merge commit for a CRAN patch, one of them accompanied by a typo fix.
Direction cannot be read from this feed. The release notes are unedited merge-commit messages, so the only signal is cadence — roughly annual, each release framed as a CRAN patch rather than as feature work. That pattern is consistent with a package whose estimators are considered finished and which now moves only when CRAN policy requires it.
On the evidence here the next release is another CRAN compliance patch, but the notes are too thin to support a confident read of what the maintainers are actually working on.
mize provides a configurable interface to unconstrained numerical optimization methods - line searches, gradient descent variants, quasi-Newton updates - usable both as a one-shot call and as a stepwise iterator. Its functional surface has not changed since the initial CRAN release in July 2017. Every release since has been a patch: R-devel compatibility, line search edge cases, and in 2026's 0.2.5 the removal of LazyData from DESCRIPTION plus deletion of some flaky tests.
The pattern is a stable library rather than an abandoned one. Fixes address real reports - a bracket_step error when the Schmidt line search exhausts its function evaluation budget, an incorrect gradient count under backtracking with a specified step_down - and the maintainer used one release note to clarify that backtracking behaviour differs depending on whether step_down is supplied, which reads as answering a recurring question. The five and a half year gap between 0.2.4 and 0.2.5 is the clearest signal: the package is maintained on demand, not developed.
Expect nothing until an R or CRAN policy change forces another compliance patch, which is what triggered the most recent release.
Other Infra & APIs 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 estimatr or mize.
France's national flood statistics, ported out of Fortran and into R.
Sign, zero and narrative restrictions brought into the bsvars ecosystem.
The grammar of uncertainty visualization, now drawing the uncertainty in its own estimates.
IP address vectors for R that hit 1.0 and then went quiet.
A column-key toolkit for stitching decades of ecological field data into one table.
Microsoft's automated forecasting framework, still mostly a one-maintainer effort.
See all estimatr alternatives → · See all mize alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
Both compete on the same themes — r-package — within Infra & APIs. estimatr and mize 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. estimatr and mize 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 Infra & APIs products to evaluate alongside.
Top estimatr alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "estimatr alternatives" section above for the current picks, or visit /alternatives/estimatr for the full list with editorial commentary on each.
Top mize alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "mize alternatives" section above for the current picks, or visit /alternatives/mize for the full list with editorial commentary on each.