The thing that people complain about most when it comes to using DAWs on Linux is that 90% of the plugins in the world are not available (at least, not without some relatively technical magic such as yabridge).
It is therefore quite curious to see people get all excited about a DAW on another "platform" where at least 90% of the plugins in the world are not available, and in all likelihood are even less likely to ever become available than they are on Linux.
There's certainly a role for a tool like this in education and for people who so far have no realized that they really need to have Pigments or fabfilter for their project. And yes, people do exaggerate the extent to which a specific plugin is needed. Nevertheless, the lack of ability to run essentially any of the existing 3rd party plugins would, were it a native DAW, be viewed as completely crippling.
The webaudio modules "standard" offers some hope here, and I suspect that within 2-5 years, plugin toolkits like JUCE will allow you to build not just as "windows/VST3" or "Linux/LV2" or "macOS/AudioUnit" formats, but also "wasm/webaudiomodule" (or something like it). However, given how easy the various Linux options already are with JUCE, and how few plugin developers choose to use them, I have to wonder if the massively larger size of a "browser platform market" would be enough to get them to add another platform.
I was a longtime JUCE user and won't hold my breath for them to support the web. They skate strictly where the puck was two years ago, not where it's going. I also wouldn't call their Linux support "easy" - it's not surprising to me very few JUCE developers even consider using Linux in CI, let alone as a supported target.
That said, I think there's something interesting about building out an audio platform with "no VSTs" as a constraint - about 6 years ago I was convinced that the web was a deadend for even middling complexity audio projects when I saw Bandlab at NAMM, and I was very wrong. It seems like the value of a DAW that you can fire up in a browser and instantly access all your projects/share them with your friends is more valuable than having no plugins and crashing after hitting browser tab memory limits. And looking down the road it frees you from the serious problems with native plugins and current plugin APIs.
I think you make an interesting point about the implications of the no-VST constraint. In the earlier days of Reason (before they had VST support or even Rack Extensions) it was great because I could work on a song on any system that had just Reason installed and anyone that also had the current version could open it as is. No installing plugins and no plugin compatibility issues between users. Away from the studio for a weekend? Just install on a laptop and use the dongle, no problem!
Creatively it was very freeing. Naturally, plugin envy eventually crept in and I was glad when they did add VST support, but I miss the ease of use and portability. And you got to know the stock effects inside and out which offered some streamlining in workflow.
If you miss the portability and the need to know the built-ins in and out, you likely might enjoy SunVox, with its utter portability, surprising richness, and the need to be inventive to eke out interesting sounds from standard blocks.
I recently used Bandlab when I was at my girlfriend's and the only thing I had with me was my company's laptop, which I don't install any audio software on.
I wrote and recorded a little song and published it withing three hours or so, just as an experiment.
I had to reload the UI a few times after moving too quickly and because of a janky internet connection, but other than that I thought it was a well designed tool. I think it's liberating to be shielded from the many choices you make when working in a "real" DAW, and when I don't have REAPER or Studio One around, I'll happily work with a tool like this to simply stay in the habit of producing music when on the road.
But they don't support the web as a target for plugins. It also took them years to acquiesce to that particular feature request (and imho, it was a strategic mistake because the only reason to use JUCE is to make the UI for audio plugins - every other thing they provide has low to negative value-add for professional audio dev shops).
Drivers and control panels for interfaces are also an issue on non macOs and windows platforms. I don't see hardware vendors changing that any time soon.
Don't control panels and surfaces send MIDI messages that could be processed in a somehow standard way? Or do they (predominantly) run proprietary protocols over raw USB data pipes?
Yes. Reaper does too. The thing that makes plugins incompatible on Linux is just that they aren't compiled for Linux. You can run some plugins through WINE via some tools, but it's more bug-prone than natively compiled Linux LV2 or VST
It is therefore quite curious to see people get all excited about a DAW on another "platform" where at least 90% of the plugins in the world are not available, and in all likelihood are even less likely to ever become available than they are on Linux.
There's certainly a role for a tool like this in education and for people who so far have no realized that they really need to have Pigments or fabfilter for their project. And yes, people do exaggerate the extent to which a specific plugin is needed. Nevertheless, the lack of ability to run essentially any of the existing 3rd party plugins would, were it a native DAW, be viewed as completely crippling.
The webaudio modules "standard" offers some hope here, and I suspect that within 2-5 years, plugin toolkits like JUCE will allow you to build not just as "windows/VST3" or "Linux/LV2" or "macOS/AudioUnit" formats, but also "wasm/webaudiomodule" (or something like it). However, given how easy the various Linux options already are with JUCE, and how few plugin developers choose to use them, I have to wonder if the massively larger size of a "browser platform market" would be enough to get them to add another platform.