On Sat, Aug 23, 2025 at 1:59 AM Jesus Martinez Garcia <askat...@gmail.com> wrote: > > When it's for me, for research, we're allowed to get an admin account (after > jumping a number of admin approval hoops). I still found it difficult because > the VPN is installed by deffect and cannot be turned off even by admin and it > blocks a number of things (on top of my head package install, there may be > other things). I learned to manually kill its access every time I restart via > Services but there's a limit as to what one can expect the average user to > know, figure out (or how much they dare to break security policy).
This is not a VPN what you describe, but some sort of spyware. VPN (Virtual Private Network) is a tool to access intranets, it has nothing to do with preventing you from doing things on your computer! > > In any case, that's beyond the point. Sure, one can "demand" but the actual > nature of the power relationship at many unis is a different one. "Demand" meaning that in your course proposal you set out the need to have WSL, by spelling out options: SageMath - needs WSL; Oscar - needs WSL; Macaulay2 - needs WSL; Magma - needs $$$$ in license fees; Singular, SymPy, Pari/GP, GAP - not powerful enough for the task at hand; HaskellForMath (https://hackage.haskell.org/package/HaskellForMaths) - well, why not... And of course you spell out the advantage of Python as preparation for the real world jobs (TM), etc etc etc, which rules out Pari/GP and GAP (and, well, Oscar, even though Julia is popular, but not nearly as much) And you seek input and support from CS Dept on this. Cheers Dima > I'm sure our CS department would prefer Linux running in the computers that > their students use and tried demanding this but it didn't work, which makes > it difficult to run certain courses. I currently don't teach but when I do > again, if I want to run classes with Sage (increasing the teaching base), I > will sadly require a non WSL option (which, as I say, we still have as they > kept 9.1 although I suppose at one time it'll go). I also know that our > students would likely not manage to successfully do the current installation > (or give up). I know of the difficulties that the previous Windows version > posed. Perhaps Sage cannot simply run on Windows without WSL just like > programming (or maths) is not for everyone. But we were asked for feedback, > so that was mine, based on the situation here, not so much my preferences. > > I agree with the problems that MS products pose. I wish the issue stopped > with Sage. Email via Outlook is a daily nightmare (and expensive) but again, > we don't have the power to force IT go out of it (or the 'we' is not large > enough or has more survival battles to fight). Please, don't shoot the > reporter. > > Best wishes, > > Jesus > ---------------------- > > > > On Fri, 22 Aug 2025, 21:42 Dima Pasechnik, <dimp...@gmail.com> wrote: >> >> On Thu, Aug 21, 2025 at 10:29 AM aska...@gmail.com <askat...@gmail.com> >> wrote: >> > I use Sagemath on Windows (essentially because my university only supports >> > Windows, but also because I prefer Windows to other systems and I cannot >> > really deal with having separate machines for programming and everything >> > else). My university provides Sagemath on version 9.1 on the old Windows >> > installer (and I fear they discontinue it once they find out that it is no >> > longer supported). Using it via WSL doesn't work unless one has admin >> > privileges and knows how to disable our VPN (which is something we are not >> > meant to do, and that I am pretty sure they don't know I know how to do). >> >> It's just a non-starter. "My employer only allows MS-DOS, ChatGPT, and >> typewriters" sort of thing. >> >> WSL is an essential part of Windows. Demand WSL support from your employer. >> They need to understand that a lot of research software does not run >> on Windows, and won't ever, most probably. >> What does WSL have to do with VPN, or admin privileges (yes, one >> sometimes might need to be an admin on the WSL vm, it does not mean >> admin on the Windows box). >> >> >> > >> > I am saying the above because my feedback would be to get as close as >> > possible to "no WSL" as many universities tend to disable it. If I >> > understand correctly from your link, approach D may get close to this? If >> > you have specific requests for feedback (i.e. if you want me to try >> > something), let me know and I can do this in September. >> >> Approach D is very far from being done. It requires years of dedicated >> person-year work, and the persons in question have to be very capable. >> Unless Micro$oft or some other big company forks out a sizeable grant >> for this to happen (and other funding just should not go there, >> because more working on Windows software ports is mostly bringing more >> benefits to their shareholders, the benefits to the society as the >> whole from ports to Windows are negative!), it probably won't happen. >> Life is too short to be wasted on banging one's head against horrible >> M$ engineering problems. >> >> Dima >> >> > >> > Thanks for making new Sagemath options for Windows users! >> > >> > Best, >> > >> > Jesús >> > >> > On Monday, 18 August 2025 at 18:45:30 UTC+2 seb....@gmail.com wrote: >> >> >> >> > A big thanks, Marc. >> >> >> >> +1 >> >> >> >> > I am looking forward to testing it with my colleagues, students and >> >> > friends, especially those on Windows. >> >> >> >> BTW: Eight months ago, we reactivated the sage-windows repository, which >> >> was once used to provide Windows installers for Sage via Cygwin. The goal >> >> was to create a place where we could share the current status of our >> >> efforts to provide easy access for Windows users and finally make such >> >> installers available from there. Unfortunately, work on the repository >> >> has stalled, primarily due to a lack of feedback. >> >> >> >> There are currently four approaches to obtaining installers for Sage on >> >> Windows, again. At least the approach marked with "C" still works for all >> >> Sage releases available on DockerHub. I suggest adding AppImage as a >> >> fifth approach. >> >> >> >> >> >> GMS schrieb am Samstag, 16. August 2025 um 11:10:33 UTC+2: >> >>> >> >>> >> >>> A big thanks, Marc. I am looking forward to testing it with my >> >>> colleagues, students and friends, especially those on Windows. >> >>> >> >>> Best, >> >>> >> >>> Guillermo >> >>> >> >>> On Sat, 16 Aug 2025 at 10:56, Eric Gourgoulhon <egourg...@gmail.com> >> >>> wrote: >> >>>> >> >>>> This is great news! >> >>>> Thank you very much for making this possible! >> >>>> >> >>>> Eric. >> >>>> >> >>>> Le vendredi 15 août 2025 à 16:34:29 UTC+2, marc....@gmail.com a écrit : >> >>>>> >> >>>>> The 3-manifolds project is now distributing a beta version of >> >>>>> SageMath-x86_64.AppImage . >> >>>>> This means that SageMath 10.7 can be installed and run on any version >> >>>>> of linux which is compatible with manylinux2014 by downloading one >> >>>>> file and marking it as executable. The compatible linux systems >> >>>>> include Ubuntu 24.04, Debian 12 and the default ubuntu image for >> >>>>> Windows 11 WSL2, none of which are currently providing any sagemath >> >>>>> packages. >> >>>>> >> >>>>> This continues our effort to make sage accessible to the large group >> >>>>> of users which includes most students and most of their professors and >> >>>>> which consists of people who are not interested in, or not capable of, >> >>>>> learning how to compile sage or maintain 3rd party package managers -- >> >>>>> people who expect to be able to just download a program and run it. >> >>>>> While, as a reader of this email list, you probably do not belong to >> >>>>> that group, we are hoping that you do have an interest in increasing >> >>>>> access to sage and that you might be interested in testing the >> >>>>> AppImage and reporting any issues you encounter. >> >>>>> >> >>>>> For those who are not familiar with the AppImage format, it is simply >> >>>>> an ELF binary executable file containing a very small program with an >> >>>>> arbitrarily large squashfs filesystem appended. The small program >> >>>>> does a fuse mount of the squashfs filesystem and then execs a main >> >>>>> program located at the root of the filesystem. The squashfs should >> >>>>> contain all library dependencies of the main program, and all dynamic >> >>>>> libraries should be loaded from an rpath pointing into the squashfs. >> >>>>> That way the AppImage has no external dependencies. >> >>>>> >> >>>>> - Marc >> >>> >> >>> >> > >> > -- >> > 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 visit >> > https://groups.google.com/d/msgid/sage-devel/afeebaa6-19f8-4d46-b829-17be28f84db5n%40googlegroups.com. >> >> -- >> You received this message because you are subscribed to a topic in the >> Google Groups "sage-devel" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/sage-devel/mbuGsrmkhpE/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> sage-devel+unsubscr...@googlegroups.com. >> To view this discussion visit >> https://groups.google.com/d/msgid/sage-devel/CAAWYfq0v7t7ew8pi%2BZFJgSEP9YC4doYh0LBJMKzj7x_H7%3DWCwQ%40mail.gmail.com. > > -- > 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 visit > https://groups.google.com/d/msgid/sage-devel/CAJvvKwvPE1C3%2B3kg6T_52WnJ5e22HT1ZbF4dSt7%2BQtFfJxN1SQ%40mail.gmail.com. -- 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 visit https://groups.google.com/d/msgid/sage-devel/CAAWYfq1M9PpjAGBN0JRyB1bm_n8FRJYEV_Z_bzDjg2WMyPLaeg%40mail.gmail.com.
