Wait - so TMO monitors your data usage and throttles you down when you exceed your cap on a phone, but doesn't do this for hotspots and trusts the hotspot to police the data cap?
Unlimited smartphone users are in a different situation, they have "unlimited" un-throttled mobile data usage, and 7GB of Mobile Hotspot usage.
To accurately measure Mobile Hotspot data usage, your phone connects to a separate data access point when Mobile Hotspot is turned on - but there are people who are using third-party tethering applications that simply use the 'standard' access point, allowing users on "unlimited" plans to skirt around the 7GB Mobile Hotspot limit.
T-Mobile EXPLICITLY advertises the "unlimited" plan only applies to usage ON YOUR PHONE, not on any tethered devices. They have every right to curb people using their service in this fashion.
Actually, I think it's more common to simply tell Android to not send tethered data to a special "tethering" APN.
Frankly, I bought my phone, I can configure it how I want to, and I see no reason to allow T-Mobile to configure my phone for me.
As a TMO subscriber I've never had a problem with TMO's "unlimited data". What's the problem with throttling down once you've hit your cap? You still get data, just at 3G/4G speed and not LTE. IMO if you're pulling that much data from the tower then yeah, you should probably pay up because you're screwing it up for everyone else.
This seems like a totally easy problem for TMO to fix, but I wonder why they haven't done it until now.
I'm on the fence about what I think about the announcement itself (because of the implications around enforcement), but this is a pretty untenable argument if you want to get into "promises", or, who's really breaking their contract.
Contractually, the promise is unlimited data without tethering. Explicitly. If we want to talk about who's breaking what promises by rooting phones to bypass tethering data limits, it's actually the "omg hackers".
The only clue they have that these people were tethering was that they used 2 TB, but they may just as well have been watching a lot of Netflix on their tablet.
That's not true, and it's kind of uncharitable of you to assign malice and technical incompetence to T-Mobile. There can be all kinds of differences in the tethered traffic patterns vs. native phone traffic. For example:
- Differences in IP TTL of packets
- Differences in TCP options, advertised windows, and so on
- Differences in packet ordering / RTTs (if you've got interleaved TCP flows one which is terminated at the phone and one which isn't, you'd see the terminated packets being acknowledged sooner)
- Differences in the application layer data (for example the user agent field)
The people who are camouflaging their tethering would of course try to apply countermeasures against as many things as possible. But if they miss one kind of fingerprint, it could be used to detect the tethering. That'd be the case if it's a signal that'd be hard to deal with by a realtime automated system, but requires some kind of expensive offline analysis.