Pretty much any IMAP mail provider combined with Nextcloud for sync will work.

Keep in mind too that Proton Mail has a commercial offering that should work with thunderbird. I have no experience with it.

We use our ISPs mail with our Nextcloud instance but long run will probably move to domain mail with maybe Fastmail for the mail component. Not sure yet. I will be interested in what you find.

Supply chain attacks also show one reason that using older software like Debian stable may be a better plan for things that matter. All new software versions need some time to be tested and vetted.

It also shows the importance of security in depth. That less is more in terms of code dependencies and complexity. That knowing dependencies is as important as knowing your code.

I would consider the xz incident to be a success. The supply chain attack was found pretty rapidly. We have already seen many of these and we will see more. Ones I remember off the top of my head include Linux Kernel, NodeJS, Python PyPI.

I would not over blow this. Security is an ongoing activity and all security is porous.

My wife and I have used GnuCash for 20 years. We used Quicken before that. Like GnuCash way better since it is actually double entry accounting. The major limitation of GnuCash is that it is not concurrent. So to people cannot be modifying the ledger at the same time. Not sure about viewing. It can be SQL database backed though I have never used that functionality. The other place where FOSS stuff probably lags is integrations.

Edit: Another area in which GnuCash is weak is basis tracking. Fine on accounting for gains and losses but for tracking the basis is limited. At least my version is. I am on 3.x which is the version in the repos of my near end of life Debian 10 distro and latest is 5.5. At least my version there is no way to show true returns either.

Why ssh? Does ssh use xz?

I think it depends on how good your Numpy build is. Lot of Numpys are not that well built so Numba seems to help there too in that case.

For a python library to be fast it needs to be compiled for your specific hardware, vectorized, with fast math, and auto parallel. Most are probably not unless you build them youself.

Nice thing about Solidworks is I think is used the ACIS kernel. Means it is directly compatible with a lot of other software.

Actually Solidworks is consider low to mid market. NX and whatever PTC calls their high end now are the main stream CAD systems as far as I know.

LibreOffice unless MS Office comparability is the most important thing. Then maybe try ONLYOFFICE.

I used to use Solidworks and NX some. I think there are similarities. That is sketch based. I admit though, not really learned FreeCAD either. On my list some day.

There are a bunch. There are fewer that are multiuser. Search alternativeto. Thing is people want more then double entry accounting. Electronic payment processing, reporting, payroll, AR, tax… Then how does it work with the professionals you hire.

Edit: For personal stuff, my wife and I use GNUCash. It does have small company features. I do not think it is concurrent though, but it can be used with an SQL backend though. We do not use the database mode so no experience with it.

FreeCAD is the best FOSS program I know for solid modeling. Librecad works for 2D.

Yes but it does have some editing and metadata management features. Personal opinion is everyone should start with a photo manager and then only use another program if you need more. On Linux Shotwell is another photo manager and Digikam is cross platform. I know my wife uses just Shotwell. She has never needed more. I use Digikam because it is a little more powerful and flexible but less elegant. My wife is into photos… me not as much too.

There is Darktable and Digikam too. Little different direction.

Gnome 3 under the hood but it is nothing like the Gnome 3 your thinking about. It is more like Gnome 2 as far as the UI.

There is still a Gnome 2 fork around also. Cannot remember the name. Used that in the early days but had some minor but annoying compatibility issues with some apps so went to Cinnamon which is based on the modern base to avoid those.

It is easy in itself. Like on Debian, just go into synaptic and install or use apg-get. Also if you can find a good how to to just follow.

The thing about QEMU is that it has a blithering number of options and reading the man page to get an idea is a major time sink. The other challenge is deciding how you run it and interface with it via the GUI, and file system. You can setup but there are various choices. Also it integrates with other useful commands too – the commands to qcow manage images is a separate command. Or you can work with direct images and use dd, loopback, mount, etc to work with them. The nice thing is you get great Linux integration. The bad thing (or maybe good thing?) is helps to be good with man pages, bash scripting, command line, processes, networking, routing, and the Linux system.

So do not think of it like an OS setup though you will probably do an OS setup in a VM, but qemu just a complicated command. In the end you’ll want to setup a folder tree, and some scripts to handle various things so it’s baked in. I use it that way, for flexibility but VirtualBox is much easier since there are menus for or that.

You could use cinnamon. Yes gnome tech based but quite different. I use that on my workstation and leave Gnome for my media center and laptop.

I know during COVID I tried Jitsi Meet eith a friend and we had quality issues using the FSF server. Zoom was way better. Have not conferenced since then. Plus way more people have Zoom installed, tested, and know how to use it.

Also Zoom, people can dial in by phone. My wife conferences with a friend where that is required.

Would prefer a FOSS alternative. Sadly above is our experience which shows why Zoom is way more popular.

People use Python a lot as a Matlab, Excel/VBA, or R alternative. That was my use for many years. Some of these are compute focused problems and if the dataset is large enough and the computations complex enough then speed can be an issue.

As far as loading packages and printing. Who cares. These are not computationally intensive and are typically IO bound.

Yes, I hate indentation as structure but I hate tracking brackets even more.

Same for me. I have used Python for most things since the late 1990s. Love Python. Have always hated the poor performance… but in my case mostly it was good enough. When it was not good enough, I wrote C code.

Python is good for problems where time to code is the limiting factor. It sucks for compute bound problems where time to execute is the limiting factor. Most problems in my world are time to code limited but some are not.

Python compute performance has always sucked.