Actually I said that wrong. Its not really ethernet per se I see going, although I do see its current version going and being replaced at some point with a newer protocol version...But what I meant to say is I see CAT 5 and 6 dying out sooner rather than latter vs coax. (fiber, cat 7 etc) Coax will always be around because in the end, its perfect for all rf signals from an antenna like ota and radio. Its not going anywhere. And internet is not going to replace all things rf in the air in our lifetimes. And it also works excellent for Ethernet! IN fact I think its cheaper to use coax for ethernet than making ethernet work as coax, if it even works well enough for all possible scenarios for rf transmission like Coax does.
And actually, you proved my point on the IP test Directv did. It makes no sense because its so much more complex to move to an ip system than to advance a coaxial system that we have now. Much easier to setup and control and operate both for a tech and for hardware etc. Even in your dswim speculation, reality is if they did move to ip in that realm, they'd likely still do it all over coax using moca technology instead of Ethernet.
Well, it is hard to argue against cat5/cat6 being replaced before long, given how quickly networking standards, and the cabling required to support it, have evolved. But despite that I think we may be at the point where the ethernet and cabling technology used in the home evolves much more slowly, and begins to lag far behind the state of the art. I don't see a good case for 10 Gbit ethernet in the home, let alone 100 Gbit. Maybe someday, but not for a long long time. Today 10 Gbit ethernet in the home is a solution looking for a problem, and that's already easily doable with cat6/cat6a.
The technology for RF cabling has evolved far more slowly than cabling for networking because there has been little evolution required. RG59 replaced 300 ohm twin lead for convenience and RG6 replaced RG59 because satellite uses a higher frequency ceiling. There is nothing driving a replacement of RG6 on the horizon because there's not going to be any evolution in the need for passing RF signals around the home. Instead, the need for passing RF around the home (as opposed to initial distribution from the source) will simply go away over time. That is, you might still need RG6 to get cable to your home from the pole, or from your dish to one location in the house, but you won't need RG6 run to a bunch of rooms in the long run.
Think of it this way. The "analog window" keeps shrinking. It used to be that you watched content on your TV that was analog from the moment it was filmed/recorded to the time you watched it. It was never digital data at any point. Over the years that has changed and digital has taken over on both ends, squeezing towards the middle. Today we receive content that was filmed/recorded digitally and processed digitally all the way until it is broadcast to you - as modulated digital data on an analog RF carrier. Your antenna/dish receives it, you pass that RF around your home, but once it gets your set top, it is converted back to the original digital data and displayed on your digital display device. Thanks to HDMI, even the link from the set top to your TV is no longer analog RF.
Do you think that the process of shrinking that analog window will cease? I believe someday it won't be converted to digital data at each TV location (i.e. in individual set tops) but in a single location. Like the Genie already does - of course, the Genie clients use coax, because that's what people have today, and that's what Directv installers know how to do. But they communicate with the Genie using IP, so they're already moving down the path towards making coax irrelevant. The wireless Genie client (if they ever release them) goes down that path further, and becomes the first Directv set top since the ill-fated H20i/HR20i to not need the RF input. It won't be the last.