Ricardo Wurmus <rek...@elephly.net> writes: > Liam Hupfer <l...@hpfr.net> writes: > >> Ricardo Wurmus <rek...@elephly.net> writes: >> >>> My work on Guix is dominated by bulk upgrades of hundreds of R >>> packages. There is no point in submitting the upgrades >>> individually. In fact, my work on Guix has become so dominated by >>> R upgrades that I barely ever get to do much else. This means I’m >>> unlikely to review more contributions and increased friction in >>> performing these massive upgrades would only make it less likely >>> that I can perform them at all. >> >> What if the CRAN packages lived in a separate channel? > > The remaining R packages live in separate channels that have no > quality control: guix-cran and guix-bioc. > > What is available in Guix proper does have quality control and these > updates require significant work to ensure that the packages remain > usable. Don't let the number of commits fool you: my contributions to > R packages in Guix are not the results of just running a script and > calling it a day.
I appreciate your work and I think it would be valuable if you can share what can be improved in [1] based on your experience. I can see an issue with bulk updates is that, while it's easy to prepare the patch series, every requirement added into the review process will be multiplied. It's still fine for a dozen of patches, but beyond that there will be much more stress. Is it possible to divide the changes into chunks that can be reviewed independently? So that the workload at one time can be more reasonable, and this leaves room for others to join in. Thanks [1]: https://guix.gnu.org/manual/devel/en/html_node/Bulk-Updates.html
