On 25 August 2024 15:33:41 BST, Marc Culler <marc.cul...@gmail.com> wrote:
>Dear Laël,
>
>On Sunday, August 25, 2024 at 2:29:44 AM UTC-5 "Dima" wrote:
>
>CertIainly Sage does build and run on all sorts of Linux systems. But it's
>getting more and more bloated to support macOS, and this gets in the way of
>Linux.
>
>
>It is true that Sage builds on linux, of course. It is false that Sage
>bloat (real or perceived) is related in any way to the fact that Sage
>supports the macOS operating system. The core issue is that parts of the
>Sage community have a very hard time accepting the idea that most Sage
>users are more interested in doing math than in compiling code, and would
>much prefer to be able to easily install a binary package which they can
>just use to having to build their own. The fact that it is now possible
>for macOS users to do that seems to make macOS the target of belligerent
>attacks from time to time.
I don't agree with "having very hard time" etc.
It's just not so - the problem is that the current state of Sage (the bloat,
related to the rather complicated macOS development model, coupled with
hesitation to properly package Sage for macOS Homebrew) makes it very hard to
package on a typical Linux distro, so that its users can easily install a
binary package. That such efforts are stalled on Debian/Ubuntu and Fedora is
very telling.
> The fact that it is now possible
>for macOS users to do that seems to make macOS the target of belligerent
>attacks from time to time.
Well, basically Sage the distro often runs in the way of sagelib (the only
component
really needed).
Attacks? No, it's merely pointing out the disregard of needs of Linux packagers
by a number of Sage developers.
As long as this is not sorted out, the future of Sage as the multiplatform
project looks bleak.
Last but not the least, you continue to ignore the need to install sagelib into
existing Python venv,
the need that arises whenever anyone doing maths computations with Python
merely wants to use sagelib.
For instance the latter is how one would typically use SciPy.
And SciPy gets about 2000 citations in refereed publications per year, without
offering any kind of SciPy the distro.
SageMath gets about 10 or 20, despite, as you claim,
being much more user-friendly.
Dima
>
>I have considered trying to build a binary package for linux, but I am
>currently leaning towards using AppImage rather than Snap. If you have
>experience relevant to choosing between those I would be interested in
>hearing about it (off-list, of course).
>
>- Marc
>
>
>On Sunday, August 25, 2024 at 2:29:44 AM UTC-5 "Dima" wrote:
>
>Please post your config.log
>
>CertIainly Sage does build and run on all sorts of Linux systems. But it's
>getting more and more bloated to support macOS, and this gets in the way
>of Linux.
>
>Dima
>
>
>On 24 August 2024 09:54:53 BST, "Laël" wrote:
>
>Except the latest conda binary version is sage 10.3 not 10.4 which lack the
>optional order= parameter on the log function on pari’s ffelt
>implementation.
>
>In my case, the finite field’s order is very hard to factor so it’s
>impossible to solve my problem in old version of sagemath which is why I
>can’t just install an old package.
>
>Sincerely,
>
>Le vendredi 23 août 2024 à 05:37:22 UTC+2, Nathan Dunfield a écrit :
>
>I recommend you install SageMath via Conda:
>
>https://doc.sagemath.org/html/en/installation/conda.html#install-from-conda-forge
>
>In my experience, this method "just works" even on seriously old and cranky
>Linux distros.
>
>Best,
>
>Nathan
>
>On Thursday, August 22, 2024 at 12:32:08 PM UTC-5 "Laël" wrote:
>
>Even by following this guide
><https://sagemanifolds.obspm.fr/install_ubuntu.html> on a fresh minimal ᴠᴍ,
>whether it’s on red hat or Debian or Ubuntu, the resulting sagelib doesn’t
>load :
>I’m meaning simply typing
>from sage import *
>
>on a python prompt triggers a segfault.
>
>Of course this means building docs fails. Here’s the logs attached with lot
>of warnings.
>
>x
>
--
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/D69D8243-8D9E-4B57-BC46-544C60C0A338%40gmail.com.
