My main/only concern is switching costs / lock-in to proprietary
services for open-source projects. If the service you are looking at
"owns" the content in some sense, such that you have to be a member,
pay, potentially pay in the future, or depend on their continued
existence for access to content, then I'd raise an objection as someone
who was once heavily involved and who continues to monitor development.
If there's something like a mailing list archive in a format that can be
monitored in real time and that would survive the termination of the
company/organization hosting it, then it's fine. Something like Slack
or Teams is a non-starter because the content can't be mass-downloaded
and hosted outside the service, so the service could suddenly impose
fees or terminate business if bought by a competitor, and we'd lose
everything. There are frequent references in the discussions to past
discussions and decisions. I would not want us to generate a bunch of
history and then loose access to it.
I'm not familiar with Discuss, and perhaps not everyone is. Could you
elaborate a bit about these concerns?
Thanks,
--jh--
Joe Harrington
Office of Research, Department of Physics, and Florida Space Institute
University of Central Florida
Founder of the NumPy/SciPy Documentation Project
On 4/3/24 10:04, Melissa Mendonça wrote:
Hi all!
We are proposing to move the development discussions from
scipy-dev@python.org to
https://discuss.scientific-python.org/c/contributor/scipy
The proposal is as follows:
- We start using Discourse as a replacement to scipy-dev@python.org
- We don't move the mailing list archive and history. That will all be
frozen at the current location
(https://mail.python.org/mailman3/lists/scipy-dev.python.org/). While
it is possible to migrate the history to Discourse, this makes the
transition A LOT more complicated including potentially exposing
private email addresses which is something we probably don't want to do.
- Posters will have the ability to post on the website forum or
through email. I am personally still getting used to the Discourse
interface, but it seems it is possible to interact with the forum
through email only if that is preferred, so hopefully those who would
like to keep the email workflow would not have to adapt too much. This
message was posted through email, for example:
https://discuss.scientific-python.org/t/this-is-a-test-message/1117
Note that there are two forums for SciPy in the Scientific Python
discourse:
- https://discuss.scientific-python.org/c/contributor/scipy (for
developer discussions)
- https://discuss.scientific-python.org/c/user/scipy (for user
discussions)
There is no expectation that maintainers would monitor the user forum.
I hope this is clear - let me know if you need more information. The
forum is already open so anyone can post there. Let me know your thoughts.
Thank you Stéfan van der Walt for the help setting this up and
Pamphile Roy for the previous work on setting up the forum!
- Melissa
_______________________________________________
SciPy-Dev mailing list --scipy-dev@python.org
To unsubscribe send an email toscipy-dev-le...@python.org
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address:j...@physics.ucf.edu
_______________________________________________
SciPy-Dev mailing list -- scipy-dev@python.org
To unsubscribe send an email to scipy-dev-le...@python.org
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address: arch...@mail-archive.com
