tealeaves
A leaf-temperature model that finished its job in 2020 and has stayed finished
A side-by-side editorial comparison of rgm and saros — release velocity, themes, recent moves, and the top alternatives to consider.
A microbiome network model that got itself un-archived by deleting the dependency that killed it.
rgm implements the random graphical model for microbiome interactions across related environments, published in JABES in 2026. The package was archived from CRAN in February 2026 because of its dependency on huge; the recovery release drops that dependency entirely, which cost it the graphical-lasso warm start that used to seed the initial graph — the default is now an empty graph, with warm starts left to the user. A post-processing function returning ggplot diagnostics arrived in the same release.
Survey reporting automation moves past charts and starts writing the sentences about them.
saros generates survey reports semi-automatically, turning questionnaire data into the plots, tables and Quarto structure a report needs. Version 1.6.0 adds txt_from_cat_mesos_plots(), which produces textual summaries of two categorical plots and calls out where groups differ significantly, plus crowd_plots_as_tabset() to fold a set of plots into Quarto tabsets with computed heights and download links. An interactive interval plot type arrives alongside.
rgm implements the random graphical model for microbiome interactions across related environments, published in JABES in 2026. The package was archived from CRAN in February 2026 because of its dependency on huge; the recovery release drops that dependency entirely, which cost it the graphical-lasso warm start that used to seed the initial graph — the default is now an empty graph, with warm starts left to the user. A post-processing function returning ggplot diagnostics arrived in the same release.
Three tags shipped inside two hours on one day, and the notes are candid about why: 1.1.0 held the actual work but was never released, 1.2.0 restated it under a higher version to signal the size of the change, and 1.2.1 answered CRAN pre-test feedback. Beyond the archival recovery, the visible work is housekeeping that had accumulated — a shadowed rmvnorm() definition, roxygen import tags that were silently emitting nothing, leftover C++ template scaffolding, and build artifacts under version control. The diagnostics function is the only genuinely new user-facing capability in the window.
The immediate task was restoring availability, and that is done; the open question the entries raise is whether losing the graphical-lasso warm start affects convergence in practice, which the new diagnostic plots are positioned to answer.
saros generates survey reports semi-automatically, turning questionnaire data into the plots, tables and Quarto structure a report needs. Version 1.6.0 adds txt_from_cat_mesos_plots(), which produces textual summaries of two categorical plots and calls out where groups differ significantly, plus crowd_plots_as_tabset() to fold a set of plots into Quarto tabsets with computed heights and download links. An interactive interval plot type arrives alongside.
The package consolidated its interface first and is now spending that consolidation. The embed_* family collapsed into a single makeme() generic with S3 dispatch in 1.2.0, and every output type since has been an S3 method rather than a new exported function, which is why 1.5.0 and 1.6.0 could add violin plots and interactive interval plots cheaply. The 1.6.0 notes are heavy with internal modularisation, splitting makeme() into argument setup, crowd processing, assembly and validation without touching the public API. Text generation is the newest direction and the one that changes what the package produces.
Expect the textual summary work to extend beyond the two-plot categorical case it currently handles, and the tabset and download-link helpers to spread across the remaining output types. The public surface should stay stable while the internals keep being split.
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 rgm or saros.
A leaf-temperature model that finished its job in 2020 and has stayed finished
A credential platform assembled two or three pull requests at a time, never a headline
Search without knowing the field — SigNoz keeps lowering the cost of not knowing your schema
A NOAA Fisheries colour palette that ships when the branding guide changes
A genetic-mapping mainstay that now points new users toward MAPpoly at load time
Relative-risk regression that converges where glm fails, under an unreadable tag order
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. rgm and saros 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. rgm and saros 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 rgm alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "rgm alternatives" section above for the current picks, or visit /alternatives/rgm for the full list with editorial commentary on each.
Top saros alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "saros alternatives" section above for the current picks, or visit /alternatives/saros for the full list with editorial commentary on each.