On 28/Feb/15 07:09, Joe Greco wrote:
Only partially. It is also a phenomenon of having built the first broadband networks with that asymmetry, which in turn discouraged a whole host of potential applications, which in turn creates a sort of bizarre self-fulfilling prophecy: broadband networks don't see much call for tons of upstream because it wasn't available, and so there aren't lots of apps for it, and so users don't ask for it, and so the cycle continues.
My point. It's not that folk don't ask for more uplink, but it's that they adjust to their situation because it's hard enough getting a sales person on the phone that knows what their doing, let along getting someone clued up to come install the damn thing. It's like cellphone toll quality - we've all accepted that if the call is unclear or drops, we simply ring our party back instead of doing something about it. We adapt to our network conditions where we know further argument will yield strokes and heart attacks. It does not mean we don't want better...
In many cases, users who had high upstream requirements have been instead working around the brokenness by, for example, renting a server at a datacenter. I know lots of gamers do this, etc.
A lot of my staff queue their uploads until they get to the office, where we have fibre to our PoP. That's saying much...
So even if we were to create massive new upstream capacity tomorrow, it might appear for many years that there's little interest. Consider streaming video. We theoretically had sufficient speed to do this at least ten years ago, but it took a long time for the technology to mature and catch on.
However, it should be obvious that the best route to guaranteeing that new technologies do not develop is to keep the status quo. With wildly asymmetric speeds, upstream speeds are sometimes barely enough for the things we do today (and are already insufficient for network based backup strategies, etc). Just try uploading a DVD ISO image for VM deployment from home to work ...
The current service offerings generally seem to avoid offering high upstream speeds entirely, and so effectively eliminate even the potential to explore the problem on a somewhat less-rigged basis.
Agree - but fundamental change like this doesn't happen overnight. Whenever we start increasing upload speed, there will be reasonable latency until users start to take advantage. So the sooner, the sooner. Mark.