As others have said [uv] (https://github.com/astral-sh/uv) is likely a good option but since you’ve mentioned being a data scientist you might also check out [pixi] (https://prefix.dev/tools/pixi).
It’s built on top of conda so will likely have all the packages you might need.
It’s got quite a nice workflow, keeps things contained in the project directory, and adds a few conveniences over standard conda.
Sure, as a good admin he should certainly implement all industry best practices.
Of course, once he did that he’d still have the same fucking problem as unless he convinced the rest of the world to drop SPF the relayed messages would still be rejected.
I think it’s more about making things easy for his employees. His comment is just recognising that they already have personal gmail accounts so he’d like to allow them to use the same client for work email. Data privacy doesn’t seem to be an issue for him.
I do the same thing for my mail - rather than juggle between accounts I can just select from a dropdown which account to send as, and I see all my mail in one inbox.
His setup is complicated because he’s doing additional processing on the incoming mail for his domain - he can’t just hand it over to gmail, he wants to relay it. And because SPF breaks mail relaying he’s been relying on a workaround - he’ll just move on to implement RFC8617 instead now (assuming that gmail supports it - it’s still listed as experimental).
It’s not his SPF record.
The forwarding he’s talking about isn’t the same as you hitting forward in your mail client.
SPF only authenticates the first hop from the origin MTA. If you put a relay server in then you either need to disable SPF checking on subsequent MTAs or implement RFC8617. If you don’t then when subsequent MTAs check the original sender’s SPF it will fail because the message came from your relay.
Sorry, it didn’t seem like you were aware of them from the post above. There are plenty of reasons to stay with Windows, Linux lacking enterprise management tools just isn’t one of them.
People don’t generally care which OS they use as long as they can get their job done. We had one sub-division entirely on an immutable Linux desktop, another media unit was all-in on Apple products. As you say though, they’re outliers - simple inertia will keep people with Windows for a long time to come, their dominant position ensures it.
The cost vs complexity argument isn’t a compelling one either - there’s a reason so little of the internet runs on Windows.
If the taskbar position changes then the screen dimensions available to applications change - windows may need to be moved and resized. The applications themselves handle that. Of course, they need to be able to do this anyway so it’s not really an issue.
There’s plenty of enterprise management tools available - these tools all existed in the Linux world before their adoption to Windows.
There’s a bunch of different configuration management tools available:
Or you could go for an MDM (Mobile Device Management) solution:
These lists are not exhaustive.
The same tools that manage data centers full of servers can also be used to manage user devices.
Pretty sure a pedestrian-stepping-in-front-of-a-normal-tram’s day isn’t going to be any better.
Perhaps they shouldn’t be operating dangerous machinery if they’re not fully involved in what they’re doing.
They get moved to the init process (parent 1) if their original parent dies. The init process should always wait on its child processes so they’ll get cleaned up then. No reboot needed.
Once they’re zombies all they really exist for is to return an exit code for their parent - they’re no longer running.