This idea isn't new. Does nobody remember Sveasoft? They tried this with regards to alternative WRT54G firmware, back when that was a thing. They ended up faili

This idea isn't new. Does nobody remember Sveasoft? They tried this with regards to alternative WRT54G firmware, back when that was a thing. They ended up failing in the market, to the point that seemingly nobody even remembers the attempt. Redhat seems to be betting on their size for their staying power. But that size also means there is much more demand for third parties to route around their attempt at creating an inefficient market speedbump. For example, that proposed solution of subscribing to Redhat through a cloud provider, then distributing the source code seems quite straightforward. If they tighten up on that, it would only seem to take a handful of small subscribers comparing the source code they receive to make sure there is no steganographic watermark, and then distributing source through an unrelated entity without fear of reprisal. I'm sure they will succeed at creating some level of speed bump for companies that want things to just work, but I'd say that ultimately this is just going to lead to another cycle of users moving to the new libre thing, and then Redhat needing to pay to reacquire those customers later.