so back in December, i switched from
Verizon to
T-Mobile. i had heard all the stories about how bad T-Mobile's network was, but I have come to despise the subsidized phone pricing model that dominates in the US and I really wanted to give T-Mobile a chance. I purchased a full price iPhone 6 from T-Mobile about 10 days before my Verizon contract expired and activated it using a temporary number in order to get an idea of what was coming. I've been rather disappointed to date but there is still a possibility that T-Mobile might redeem themselves.
I live in a fringy area for T-Mobile's towers. I see weak T-Mobile signals in my driveway (south side of the house) about 1/2 the time; in my office, which is on the other end of the house, T-Mobile signals are rare and not usable when they do show up. However, AT&T is supposed to carry the load for that; they partner with T-Mobile around here and have a usable network here in Averill Park, which i know from my years as an AT&T customer.
when i went to activate my phone, well, nothing doing. the first T-Mobile rep told me that the problem must be the brand new SIM that was preinstalled in the phone, and directed me to take the 1 hour round trip to the nearest T-Mobile retail outlet to swap SIMs. i combined it with another trip to reduce the pain level. the T-Mobile retail store was a trip; the salesmen were all quite sure that T-Mobile service never worked in Averill Park, and didn't want to replace the SIM. i pointed out to them that T-Mobile's published coverage map included Averill Park, and insisted on the SIM replacement (even though i was sure it wasn't going to work) because of the importance of jumping through the hoops properly when dealing with tech support. they seemed to be pretty oblivious to the issue of the marketing materials claiming service worked here and to the issue of proper interactions with tech support. i got the SIM replaced and left. as i expected, it fixed nothing.
the second call to tech support came about the time i actually ported my verizon number over to t-mobile (on the 15th of December). we worked through a bunch of stuff; the tech support guy got my phone working over the local wifi (wifi calling being a new feature to me); this means that usually my phone works within range of my wifi at home, which saves things (i work from home most days), but it's not entirely reliable. he also walked me through manual carrier selection, and we thus figured out that my phone can see the AT&T towers around here, but won't roam onto them, despite the partnering agreement. this is something that could be a provisioning screwup on either T-Mobile or AT&T's part. he opened up an engineering ticket and told me to wait at least 72 hours. i don't know the details of what he put in the ticket; hopefully it was about the failure to roam onto their partner's network.
So the holidays came up, and i decided not to pursue the situation until we returned from visiting family in NC. on the long drive there and back, i observed the phone successfully switching to AT&T in a number of gaps in T-Mobile coverage, which showed that this wasn't system wide and that my phone was not defective; all good information to have. so two days ago, i poked at T-Mobile's twitter help account with a stick, asking them what was up with the ticket. i also drove to the post office - not just to pick up my mail, but also to do the manual carrier selection thing and verify that i still couldn't roam onto AT&T's network - and i can't, so the problem isn't fixed (the idea behind driving to the post office is that there are no T-Mobile towers reachable from there, so there is no chance of a poor T-Mobile signal interfering with roaming.)
the word from the t-mobile twitter people came back - the ticket was closed, apparently because engineering saw my phone show up on t-mobile's network. which of course wasn't the actual problem.
so there's a new ticket open; we'll see what happens this time. but the closure of the ticket has me wondering if the performance metrics for their engineering group have to do with closing tickets and the length of time tickets are open, instead of actual successful solution of problems. because closing a ticket should not be confused with solving a problem; when you make ticket closure alone the criteria, you create a perverse incentive to close tickets regardless of whether solutions are found. anyone who has any experience in running call centers and engineering support is supposed to know this.