January 31, 2026 at 7:25 AM, "Wilfred van Rooijen" <[email protected] mailto:[email protected]?to=%22Wilfred%20van%20Rooijen%22%20%3Cw.f.g.vanrooijen%40gmail.com%3E > wrote:
> > Thanks for your kind suggestions. I will check out the "Contributing" list. > > In my professional life I am mostly involved with numerical analysis (mostly > FORTRAN codes), programming (FORTRAN and Python), scientific publications in > LaTeX, and "casual scripting", mostly Bash and Python. In my career I have > also coded extensively in Perl and Visual Basic. I would say that about 90% > of my experience is on Linux systems (when I was a PhD student, I also did > some batch scripting on VMS, on a VMS cluster that was subsequently replaced > by a linux cluster). In other words, I am familiar with computer programming > but not the world's best programmer by any means. > > For GNU/HURD I will take a look at things like BLAS and LAPACK, those are > definitely needed for things like NumPy, SciPy and SymPy. > > In my opinion, for a desktop OS to be "acceptable", it must support at least: > - USB > - BlueTooth > - sound > - WiFi > > And there must be at least web browser. I don't know how the situation is > with GNU/HURD . Not that great. I use netsurf on the Hurd. But you have to fix some of the compilation failures manually at the moment. The best Hurd browser used to be LadyBird. Sergey got it compiled a year or two ago. Getting LadyBird running on the Hurd might be the best we could do. Though LadyBird is moving to use Swift. I don't know if the Hurd has ported Swift yet... > > Question: which packages are top priority for porting to GNU/HURD? I have > some experience adapting scientific software to various flavours of linux, > perhaps I can be useful in that area? > > I will investigate a bit further with the links in the "Contributing" page > and I will contact you all later. > > Thanks, > Shinichi > > > > On Sat, Jan 31, 2026 at 7:34 PM João Pedro Malhado <[email protected] > mailto:[email protected] > wrote: > > > > > Hello Shinichi, > > > > On Fri, Jan 30, 2026 at 08:28:03PM +0900, Wilfred van Rooijen wrote: > > > Thank you for your invitation to help development on the HURD. I would > > > like > > > to try, but there are few things that are holding me back: > > > > > > - I have quite a strong background in programming for numerical analysis, > > > things like finite element method and (large) systems of differential > > > equations. I consider myself fluent in FORTRAN, LaTeX and Python but I > > > have > > > little to no experience with C > > > > Sounds like we have a similar background :) > > > > > - I can definitely bring myself up to a level where I can bring a positive > > > contribution to GNU/HURD, but the question is how much time I'd have to > > > invest, in other words, I expect it would take a (very) long time before I > > > could become really productive > > > > > > - I am also reluctant because of the experimental nature of this software. > > > I am not sure I want to spend my time on something that will never reach > > > the level of actual application. I have spent 17 years of my life as a > > > professor working on things that are mostly irrelevant and useless, and > > > now > > > I want to spend my time on things that are actually useful. > > > > I can understand your reservations. Ultimately it is your evaluation to > > make. > > The number of developers is small, so there is some chance that individual > > contributions are significant. But it does take time to learn and be > > productive. > > > > > If I were to "join the team", what kind of work are we talking about? Is > > > it > > > feasible to write a complete USB driver, or something with sound? > > > > It is a free software project, so you get to choose what you want to work > > on ;) > > > > I think developing a USB stack would be very useful (I believe the idea > > would be > > to adapt work from NetBSD), but probably on the higher tier of difficulty. > > > > The following page lists a number of well defined projects useful in the > > short > > term: > > https://darnassus.sceen.net/~hurd-web/contributing/ > > > > There are things to be done within Debian as well (this is a Debian > > mailing-list after all), there is work to be done on porting different > > software > > to work in the Hurd. Perhaps porting different scientific software in > > Debian to > > the Hurd would be a good way to learn. The FORTRAN compiler should work, I > > don't > > know if there are particular challenges in porting FORTRAN software to the > > Hurd. > > > > Other people in the list may have other suggestions. > > > > Regards, > > João > > >
