On Mon, Apr 8, 2024 at 7:19 PM Matthias Koeppe <matthiaskoe...@gmail.com> wrote:
> I need your help on these PRs. Please vote. > > Special expertise is not required for voting. You will find the comments > in these PRs instructive -- also as illustration for a (long overdue) > *discussion > about governance and review standards* in the Sage project. > > *1. Please vote +1 on both https://github.com/sagemath/sage/pull/36561 > <https://github.com/sagemath/sage/pull/36561> and > https://github.com/sagemath/sage/pull/37138 > <https://github.com/sagemath/sage/pull/37138>* ("Move metadata from > setup.cfg to pyproject.toml"). > These are trivial "chore" PRs. They update metadata of our > pip-installable packages "sage-conf" and "sagemath-standard" to the latest > format. > These straightforward and appropriately focused PRs have been held back by > months by *bundling the review of the PRs with unrelated issues.* I call > this "artificial friction"; see the discussion in the PRs. To help overcome > this artificial friction, please vote. > This is not true - the friction is not artificial. It is due to legitimate concerns of developers who are not interested in spending all of their time on ever growing "Sage the distribution", and/or who see little merit in Matthias' sagelib modulalisation project, which uses Python features (most of all, namespace packages) not universally supported by a number of important tools, such as Cython and pytest. Please vote -1 on these two PRs (there are more similar PRs around). This will force Matthias to reconsider his priorities, and enable other voices to be heard. So far, Matthias refuses to reassess the priorities of the project - instead he puts away the criticism as "abuse" directed at him. As a result of this friction, a number of developers have left or about to leave the project, or are discussing creation of a hard fork of Sage. In particular, I think it is urgent to re-access the need for sagelib modularisation in its current form. It appears to be harming the project, and its benefits are questionable. As well, it's urgent to make Sage more modular on the level of Sage the distribution - the "interesting" part, sagelib, is getting increasingly entangled in Sage the distribution (which is just a constantly growing pile of Jupyter, Python, ans Sphinx-related packages, which Matthias and few others are all too keen on hoarding). This in particular makes Sage harder and harder to package for Linux, and other, distributions (Packaging of Sage for Debian/Ubuntu and Fedora has been stalled for years already). > *2. Please vote -1 on both https://github.com/sagemath/sage/pull/37387 > <https://github.com/sagemath/sage/pull/37387> and > https://github.com/sagemath/sage/pull/36951 > <https://github.com/sagemath/sage/pull/36951>. *These PRs are about a > Developer Experience issue, namely the workflow on GitHub that notifies > developers when the HTML documentation is ready for inspection by PR author > and reviewers. Now a few developers have made it known that they are > annoyed by the notifications (whether received by email or the notification > tool on the GitHub website), and the PRs seek to turn off most or all of > the notifications. That *these notifications enable a productive > notification-driven development style, and that the notifications serve the > project's need for quality control on the formatted documentation*, has > not received meaningful consideration. > We are not an enterprise with full-time developers 24 hours a day ready to react to these endless notifications. They are spam for almost everyone, and should be turned off. Please, please vote +1 on them. What's happening on these PRs is exactly what I had cautioned about in > https://github.com/sagemath/sage/pull/36726#issuecomment-1820148873 > regarding the (then-proposed, now established) system of majority votes as > a conflict resolution mechanism for PRs. To balance it, we need the > involvement of the larger developer community: please vote. > No, what we really need is an honest general discussion on directions the project is taking. Right now it's going nowhere, to ever growing pile of bugs noone even talks about addressing (e.g. Pynac; Maxima; memory leaks related to Singular; etc etc), to huge package bloat, etc. Dima -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/CAAWYfq1RJSA_WNgvjdiSboxYJ2e7X3h2bwOYh_du82aekdtO0A%40mail.gmail.com.
