I own two Raspberries 1, a Raspberry 4 8GB and a Raspberry 5 8GB. I wouldn't recommend the 4 as a full-fledged desktop replacement, but the 5 has been very smooth so far.

I'm currently using the latest Raspberry Pi OS Lite and installed KDE on top.

I was lucky then with the 4 A400 I'm still using. I also have 3 BX500 that have been very reliable.

Kingston A400s and Crucial BXs have been very good as cheap SSDs in my experience.

I haven't used source-based distros, but I've installed Linux on a couple of older Macs. You will probably need to search for proper drivers for the Webcam and Wifi. Other than, you won't have any mayor problems.

My own example. I still have an ancient netbook lying around. It runs on an Intel Atom N270, which is only 32bit / i386. It came with Windows XP and I quickly switched to Mint, when it was still supporting 32bit.

I think the last Ubuntu release supporting i386 was 18.04 (around 2018) and all other distros started to drop i386 support after that.

AFAIK Debian is the only major distro still fully supporting i386. And a Debian based distro that still supports i386 is MX Linux. My ancient and crappy netbook is running MX Linux right now.

My 'weird' example. I have a Raspberry 5! It's ARM and very new. It runs its own distro, Raspberry Pi OS (Debian based), and Ubuntu does also fully support it. Right now if you try some other distro, it probably won't even boot unless you start tinkering a lot with it.

So Debian is definitively a choice for very old hardware. And the odd ARM SoC has usually at least some custom Ubuntu build that runs with it.

Perfect!

Good idea! But the borders don't look good with a darker background. I'm seeing this right now: https://0x0.st/H0JJ.png

Same problem here!

django-tui 23.9 adds an interactive shell for running django ORM queries: https://fosstodon.org/@anze3db/111608276020943202

Part 3: https://www.paulox.net/2023/12/11/database-generated-columns-part-3-geodjango-and-postgis/

TalkPython episode about Memray: https://talkpython.fm/episodes/show/425/memray-the-endgame-python-memory-profiler

Part 2, PostgreSQL: https://www.paulox.net/2023/11/24/database-generated-columns-part-2-django-and-postgresql/

Part 3, GeoDjango & PostGIS: https://www.paulox.net/2023/12/11/database-generated-columns-part-3-geodjango-and-postgis/

PyQT / PySide are huge, but they have been very good in my experience coding cross platform desktop programs. macOS, Windows and Linux (even on ARM) are very well supported.

The error you would get is a timeout. The easiest way to wait longer is increasing the timeout of your connection. I think the default value for a timeout is 5 seconds.

Look at ForeignKey -> Arguments: https://docs.djangoproject.com/en/4.2/ref/models/fields/#arguments

In the sample in the linked page, deleting a Person object would also delete Order object/s linked to that particular person via the ForeignKey

edit: although the syntax is usually on_delete=models.CASCADE

I use VSCode for coding, but if it's a small script or pure text files, then I use Geany.

Follow-up by Paolo Melchiorre https://fosstodon.org/@paulox/111262287902120294

https://github.com/pauloxnet/uDjango

More on μDjango: https://www.paulox.net/2023/10/26/udjango_micro_django/

IMHO the Django Girls tutorial should be your first step: https://tutorial.djangogirls.org/en/

I was also a Pro user and I'd gladly pay again for a Pro version.

Ultra was too much for me on Reddit and here it's the same.

Thanks!

I'm having a similar experience. Almost all developers (mostly Python/Django) I was following on Twitter are on Mastodon and being able to follow hashtags is great. The servers are stable and I kept the very first android client I tried (Tusky).