I looked into #meshtastic on a desktop again and I'm sad to report that it looks like it's really immature (at best). The TL;DR is that the best case scenario has the software working with a single #LoRa USB adapter and that looks like it's far from certain.
If you're interested, I show my work below.
There's a #Linux native app, which you'd think would be the end of this thread, but unfortunately not. Almost everything here is about emulating Meshtastic traffic, not transmitting and receiving with a real #radio https://meshtastic.org/docs/software/linux-native/
There's a single sentence saying it can use hardware, but no more info there. Tine to start digging. 🧐
They link to https://github.com/geeksville/framework-portduino which says it works with "the Pine64 Lora USB dongle". Maybe that uses the same hardware as the LoStik? Let's see.
Well, the only USB LoRa device that pine64 seems to sell is the Pinedio: https://pine64.com/product/pinedio-usb-lora-adapter/
That page says "The Pine64 USB LoRa Adapter for PineDio ecosystem, suitable for SBC application." So, in other words, not suitable a desktop operating system. But it gets worse. That page goes on to give a big red warning "Software for receiving and sending LoRa messages via this adapter already exists, but at the time of release no SBC operating systems have implemented support for it."
Not compatible with the LoStik. Like, at all.
What does all this mean? It means that Meshtastic on a desktop is likely not ready for developers to tinker with.
What about tinkerers writing their own code that implements the meshtastic protocol in a way that will work with and USB LoRa device? I'm afraid that's not currently easy either. This is because Meshtasic is not a protocol, it's a piece of software. The documentation has great diagrams https://meshtastic.org/docs/introduction/ but no detailed specification.
What this means is that for someone to write desktop software which is compatible, they first have to read through the Meshtastic source code and try to figure out what the code does and why. Since there's no protocol, these internals could change at any time.
I leave you with this though. There is hope. It's likely that the core LoRa part is fairly simple and the WiFi/Bluetooth parts don't need implemented on a desktop app. So it should be something a single developer could do in ~20 hours/week for a couple months. Much less if they are a Meshtastic developer who is already familiar with the code base.
The same is true for documenting the software and writing the specification.
So it can be awesome. It's just not there yet and we just need a volunteer to take it on. ✊