IRC channel logs

2022-02-18.log

back to list of logs

<rekado_>hi there
<rekado_>are there any news on our tutorial application?
<civodul>hi!
<civodul>no, not yet
<civodul>notification of acceptance: 4th March
<civodul> https://www.isc-hpc.com/tutorials-2022.html
<rekado_>oh, okay
<rekado_>I thought I had missed something
<PurpleSym>rekado_: Going back to nbconvert: I see texlive-base is propagating inputs and if I do that for nbconvert it works too. I guess that’s how modular texlive works then, propagating packages?
<PurpleSym>(Actually, that might be because a Python package is calling xelatex…)
<civodul>i've seen people who don't know what to do with the output of "guix shell --check"
<civodul>to me, it seems there's a diagnostic, an explanation, possible workarounds, and links to documentation
<civodul>so i don't really know what else can be done
<civodul>yet, it seems to be insufficient or unfit to some (to many?)
<civodul>thoughts?
<zimoun>civodul: me, I am annoyed to have to check each time I run a “guix shell”. So, whatever is the behaviour, it is hard to satisfy many many users with many many different use cases. :-)
<civodul>"have to check"?
<civodul>the thing encourages users to do it once, not every time
<civodul>(the suggestion to run --check is printed only once)
<zimoun>no, “guix shell hello” exit “guix shell hello” print twice
<civodul>gasp!
<civodul>regression
<zimoun>but that’s expected, no?
<zimoun>“guix shell hello“ displays hint, then ‘guix shell hello –check’ dispplay the message, then ‘guix shell emacs’ never shows the hint again; except after the expiry time, I guess.
<civodul>ah no, sorry, i wasn't clear
<civodul>it's printed until you run --check once
<civodul>once you've used --check, it's never printed anymore
<civodul>yeah, what you wrote
<zimoun>yes, all is fine, I guess. :-) No bug.
<civodul>right
<civodul>i think it's a reasonable tradeoff between no hints (and you have to find about the option by yourself) and printing it all the time (and it gets really annoying)
<zimoun>I think it is fine to encourage user to to --check. The message is clear – bah I am not convinced people will read info but that’s another story.
<civodul>that's precisely the starting point of this discussion
<zimoun>yeah, I was just pointing that, because this behaviour (fine for most), it is annoying for me. Because I never run --check, so I always have the hint. Its’ fine because it is not possible to satisfy all the use cases. :-)
<civodul>yeah
<civodul>running it once is not too bad
<civodul>it takes less time that the discussion we've just been having :-)
<zimoun>heh! we are on Friday before vaccations. ;-)
<zimoun>I recently discovered this journal https://www.nature.com/sdata/ and so I have read this https://www.nature.com/articles/s41597-020-00638-4 about that https://www.aiida.net/ other said yer another workflow thing. :-)
<zimoun>I also just discovered (after the FAIR principles) the TRUST principles https://www.nature.com/articles/s41597-020-0486-7 :-)
<zimoun>and again The Dat Project https://www.nature.com/articles/sdata2018221
<zimoun>rekado_: have you tried it?
***zimoun` is now known as zimoun
***zimoun is now known as 047AACB9C