I doubt bundling things together in some sort of pack would avoid every problem with python versions I could have. That doesn’t prevent a given python version from being marked as “end-of-life” and no longer receiving security patches.
Most software is produced and maintained for use solely by the company that produced it, and probably by people who are not experts in using python, so hiding the complexity that python versions and dependency versions are coupled seems like a bad idea, especially when one wants to limit the number of versions of the same software that is installed (and therefore re-use executable files to save disk and CPU usage and avoid accidentally using the wrong version of a program).
I have not interacted with flatpak in a professional environment, so I doubt I have been directly harmed by it. However, reducing the importance of quickly upgrading software after new versions are released is probably harmful overall: performing an upgrade will usually make development easier (so making it harder for me to pitch to managers that an upgrade should be done is harmful to my morale), and incentivizing having multiple versions of the same program accessible on the same system makes surprising problems more likely.
Yes. Lack of clarity on what python version to use causes me pain on a regular basis.
If only there was a way to bundle it all together, like in some sort of pack…
I doubt bundling things together in some sort of pack would avoid every problem with python versions I could have. That doesn’t prevent a given python version from being marked as “end-of-life” and no longer receiving security patches.
Most software is produced and maintained for use solely by the company that produced it, and probably by people who are not experts in using python, so hiding the complexity that python versions and dependency versions are coupled seems like a bad idea, especially when one wants to limit the number of versions of the same software that is installed (and therefore re-use executable files to save disk and CPU usage and avoid accidentally using the wrong version of a program).
I have not interacted with
flatpakin a professional environment, so I doubt I have been directly harmed by it. However, reducing the importance of quickly upgrading software after new versions are released is probably harmful overall: performing an upgrade will usually make development easier (so making it harder for me to pitch to managers that an upgrade should be done is harmful to my morale), and incentivizing having multiple versions of the same program accessible on the same system makes surprising problems more likely.