That’s my whole point.
Zwift is doing so much - continuous - data transfer across the inter-tubes that it is essentially a web-app, not a stand-alone application. It may have its roots as a standalone (classic) application, but, let’s face it. It is deployed on AWS or equivalent, doesn’t function without an internet connection, and makes use of a variety of internet protocols.
If that’s not"web based", then we have different ideas about what that phrase means.
The “canvas” element was introduced in HTML version 5.
en DOT wikipedia DOT org SLASH wiki SLASH Canvas_element
I, too, give up arguing with you. Clearly, there is some kind of communication issue between us. We are speaking the same language, but, somehow not actually delivering any form of intelligence (i.e. knowledge) with that communication.
Well, you are welcome to keep believing that for something to be web based, it must use HTTP, and HTML.
The rest of the people I work with, who actually write Web Based applications ALL disagree with you. Web protocols span several OSI layers, and include several protocols within each layer. Mostly none of them work without the others.
Thinking otherwise is kind of pedantic, and doesn’t reflect reality.
By the same token you would have to argue that any online multi-player game is a “Web app” and that is patently ludicrous. Data regarding the locations of other riders is communicated from Zwift servers to your local machine to be rendered within your execution of the program but, if you disconnect your device from your local internet connection, Zwift will run perfectly well, display the virtual world you are riding in, control and respond to your trainer and record all of the data generated by your ride. The only thing missing is other riders because the program has no data to show where they are … the program itself is installed, and running, on your local device.