>
> As a tangent, this is why I am also skeptical of virtual environments and
> cloud-based solutions. These make the instructors' lives easier, but three
> months later those students are going to open a spreadsheet to work on
> their new dataset and not launch a VM or try to work on AWS.
>

I think Steve makes many great points here about getting the big picture
workflow rather than mastering details in a few days.  At the risk of a
further tangent though, I think the oft-echoed idea that students can leave
a SWC workshop with "all the software" installed locally, and are therefore
freed from all future "command-line bullshittery" pain is a bit of a myth.

I agree with Steve that virtualization can purely be a crutch for the
instructor that leaves the student with no idea how to access or deploy the
tools they have just learned.  But perhaps it is worth remembering that it
is also a skill that can be taught, applies to real research contexts, and
pays dividends in areas like scaling and reproducibility; just like other
tools we teach.

I sometimes feel there is a concern that virualization or use of cloud is
advanced wizardry and it is ignorant and elitist to expose students to that
pain when they will surely work locally in non-virtualized environments.
But it is worth observing that these tools are becoming easier and more
powerful every year; and that the view that everyone must master the
ancient art of POSIX-style command-line bull-shittery to install and
maintain their own local environment smacks of it's own elitism.  In either
case, local installs or virtualization, I think we must bear in mind these
are skills that must also be taught and cannot simply be done for the
students once and for all time.

Wow. Apologies for the tangent; maybe I should learn to start a new thread.
Apologies also if I've made straw men rather than capturing anyone's actual
views here. But I think dealing with infrastructure is an important issue
and as many on this list have much greater expertise than I, your thoughts
or counterpoints are greatly appreciated.

- Carl





>
>
>
> _______________________________________________
> Discuss mailing list
> [email protected]
> http://lists.software-carpentry.org/mailman/listinfo/discuss_lists.
> software-carpentry.org
>
_______________________________________________
Discuss mailing list
[email protected]
http://lists.software-carpentry.org/mailman/listinfo/discuss_lists.software-carpentry.org

Reply via email to