* Talk at RSECon26 on Guix and ravanan
@ 2026-09-11 22:50 Arun Isaac
2026-09-14 13:59 ` Ludovic Courtès
2026-09-24 16:12 ` Simon Tournier
0 siblings, 2 replies; 4+ messages in thread
From: Arun Isaac @ 2026-09-11 22:50 UTC (permalink / raw)
To: guix-science; +Cc: ravanan
Hi,
I gave a talk on Guix and ravanan[1] yesterday at RSECon26[2], the Research
Software Engineering conference. My slides are at [3].
[1]: https://forge.systemreboot.net/ravanan/
[2]: https://rsecon26.society-rse.org/
[3]: https://zenodo.org/records/22682923
I have been working on ravanan for a couple of years now. ravanan is a
Common Workflow Language (CWL [4]) runner that uses Guix to provide
strong reproducibility guarantees and content-addressed caching. CWL is
a standards-based workflow language that is used in scientific
computing. You may have heard of other workflow languages such as
Nextflow, Snakemake and the Guix Workflow Language.
[4]: https://www.commonwl.org/
Based on conversations at the conference, I feel that Guix is slowly
getting on the radar of the research software engineering world. More
and more people have at least vaguely heard of Guix. Compare this with a
few years back when no one had a clue. Thanks to the community for
having carefully cultivated and cared for Guix these many years!
I also ran into Peter Schmidt from the Code for Thought[5] podcast. I
may be appearing on his podcast to talk about Guix.
[5]: https://codeforthought.buzzsprout.com/
In other ravanan news, the amazing Luis Felipe designed a logo for
ravanan. You can see it on the ravanan homepage.
Cheers!
Arun
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Talk at RSECon26 on Guix and ravanan
2026-09-11 22:50 Talk at RSECon26 on Guix and ravanan Arun Isaac
@ 2026-09-14 13:59 ` Ludovic Courtès
2026-09-24 16:12 ` Simon Tournier
1 sibling, 0 replies; 4+ messages in thread
From: Ludovic Courtès @ 2026-09-14 13:59 UTC (permalink / raw)
To: Arun Isaac; +Cc: guix-science, ravanan
Hi Arun,
Arun Isaac <arunisaac@systemreboot.net> skribis:
> I gave a talk on Guix and ravanan[1] yesterday at RSECon26[2], the Research
> Software Engineering conference. My slides are at [3].
>
> [1]: https://forge.systemreboot.net/ravanan/
> [2]: https://rsecon26.society-rse.org/
> [3]: https://zenodo.org/records/22682923
Glad you were able to talk at RSECon! The slides cover lots of
important ground for such a venue, and in a clear and direct fashion,
very nice.
> Based on conversations at the conference, I feel that Guix is slowly
> getting on the radar of the research software engineering world. More
> and more people have at least vaguely heard of Guix. Compare this with a
> few years back when no one had a clue. Thanks to the community for
> having carefully cultivated and cared for Guix these many years!
Good to hear. In fact Guix has been present at RSECon a few times now,
last year with Ghislain Vaillant and Julien Castelneau who ran a
tutorial.
> I also ran into Peter Schmidt from the Code for Thought[5] podcast. I
> may be appearing on his podcast to talk about Guix.
That would be nice too!
The ravanan logo is sweet, great job Luis, as always.
Ludo’.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Talk at RSECon26 on Guix and ravanan
2026-09-11 22:50 Talk at RSECon26 on Guix and ravanan Arun Isaac
2026-09-14 13:59 ` Ludovic Courtès
@ 2026-09-24 16:12 ` Simon Tournier
2026-09-25 15:04 ` Arun Isaac
1 sibling, 1 reply; 4+ messages in thread
From: Simon Tournier @ 2026-09-24 16:12 UTC (permalink / raw)
To: Arun Isaac, guix-science; +Cc: ravanan
Hi Arun,
On Fri, 11 Sep 2026 at 23:50, Arun Isaac <arunisaac@systemreboot.net> wrote:
> Hi,
>
> I gave a talk on Guix and ravanan[1] yesterday at RSECon26[2], the Research
> Software Engineering conference. My slides are at [3].
Nice! Cool.
> [4]: https://www.commonwl.org/
Well, this year is running faster that my capacity to tackle my TODO
list… As discussed back on Jan. / Feb. I have on my TODO list the item
about the “node” part. Somehow, this
make NODE=/usr/bin/node prefix=/usr/local
make install
doesn’t seem to me the end of the story. ;-) What I had in mind was to
see what could be done with the old standard of ECMAScript implemented
[1] in Guile. Between a coffee and a beer at Guix Days, Andy answered
that an easier and maybe better solution might be QuickJS [2] – a C
implementation that supports recent standards.
Somehow, it could be nice to embed this into Guile or ravanan instead of
relying on NodeJS. Bah I don’t know… It’s the item on my TODO list. ;-)
Another item on my TODO list is to try to run some workflows developed
by the CWL community with a drop-in replacement of cwltool by ravanan.
Well, comment on the gap that needs to be filled.
Cheers,
simon
1: https://doc.guix.gnu.org/guile/latest/en/html_node/ECMAScript.html
2: https://bellard.org/quickjs/
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: Talk at RSECon26 on Guix and ravanan
2026-09-24 16:12 ` Simon Tournier
@ 2026-09-25 15:04 ` Arun Isaac
0 siblings, 0 replies; 4+ messages in thread
From: Arun Isaac @ 2026-09-25 15:04 UTC (permalink / raw)
To: Simon Tournier, guix-science; +Cc: ravanan
Hi Simon,
> Well, this year is running faster that my capacity to tackle my TODO
> list… As discussed back on Jan. / Feb. I have on my TODO list the item
> about the “node” part. Somehow, this
>
> make NODE=/usr/bin/node prefix=/usr/local
> make install
>
> doesn’t seem to me the end of the story. ;-) What I had in mind was to
> see what could be done with the old standard of ECMAScript implemented
> [1] in Guile. Between a coffee and a beer at Guix Days, Andy answered
> that an easier and maybe better solution might be QuickJS [2] – a C
> implementation that supports recent standards.
That would be very welcome indeed! I wasn't too pleased myself to depend
on node. The CWL spec only requires ECMAScript 5.1. So, it should be an
easy bar to cross. Patches welcome!
> Another item on my TODO list is to try to run some workflows developed
> by the CWL community with a drop-in replacement of cwltool by ravanan.
> Well, comment on the gap that needs to be filled.
ravanan is already a somewhat compatible drop-in replacement for
cwltool. This is assuming you are invoking them like so:
CWLTOOL_OR_RAVANAN workflow.cwl inputs.yml
where inputs.yml is an YAML file that contains a dictionary mapping
input names to values.
However cwltool also allows directly specifying the inputs on the
command line like so:
cwltool workflow.cwl --foo bar --ham spam
And, it also generates nice --help messages when you run
cwltool workflow.cwl --help
ravanan doesn't support this yet. It should be relatively to support all
this. We just need someone willing to spend a weekend hacking on
args-fold for this. Patches welcome again!
Regards,
Arun
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-09-25 15:04 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-09-11 22:50 Talk at RSECon26 on Guix and ravanan Arun Isaac
2026-09-14 13:59 ` Ludovic Courtès
2026-09-24 16:12 ` Simon Tournier
2026-09-25 15:04 ` Arun Isaac
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox