5G is coming - at least to America. For all the very-understandable skepticism, it’s clear now that the US’s big four wireless carriers are deeply committed to launching 5G, and they’re in a network arms race to be the first to deliver on the promise of the next generation cellular technology. And it could make buying your next phone a lot more annoying.

If you can remember way back to the dark days of 2011 and 2012, when 4G (commonly known in the US as LTE) was just getting its start, something happened on US carriers. And that something was, by and large, very bad. There were suddenly a lot of 4G phones - that only worked on a single provider. Verizon had the HTC ThunderBolt and the Samsung DROID Charge, Sprint (remember WiMax?) had the Epic 4G Touch Whatever, AT&T had the Samsung Skyrocket, and T-Mobile had a pink can and some string because its network was basically still prehistoric.

The transition to LTE was tumultuous and took years to iron out. We’re still feeling the ripple effects today (for example, VoLTE doesn’t work across all four carriers), even though we’ve finally reached the point where smartphones that work on all four carriers are relatively common. Samsung’s Galaxy S and Note phones, the all the Google Pixels, and even Moto’s cheap G and E series smartphones work on basically any US carrier. They may not support every band and every feature, but they make buying an unlocked smartphone feasible, something that hasn’t been true in the US market historically.

5G is probably about to take a wrecking ball to this little interoperability bridge we’ve constructed, though, and while carriers and phonemakers will do their best to make it sound like a small compromise in the name of progress, the truth will be - as it so often is - far more complex.

Carriers are already making it clear they don’t all have the same 5G vision. AT&T and Verizon are pushing for much higher frequency 5G in consumer devices to start with (but, of course, at very different frequencies), allowing for massive bandwidth on networks that cover very short distances and require multiple antennas for a device to effectively receive signal (as such signals are easily blocked, even by your hand). T-Mobile, by contrast, is in the process of deploying low-band 600MHz 5G that it wants to use as the backbone of its next-generation network (which it’s not even clear will work with Qualcomm’s X50 modem). And Sprint is doing… something else (2.5GHz 5G). All of these approaches will all require their own specialized modem and antenna setups, and there’s absolutely no incentive for carriers to share their toys here - why spend the time or money? - so they won’t.

That means the 5G phone you buy at Verizon will all but certainly not work for 5G on AT&T, or T-Mobile, or Sprint. And the same goes for all the 5G phones on all the other carriers, and it’s impossible to say for how long that will continue to be true. With LTE, it tooks years before even Apple was able to effectively untangle iPhone carrier SKUs. And while the technology may exist now to build a relatively interoperable 5G phone for these early networks, it’s unlikely that will happen for a while. They’re still evolving and changing, and no phonemaker wants an entire product launch to be beholden to a single operator’s goal posts. Sure, 5G phones may still work across networks just as well for LTE (we can hope!), but as 5G deployment widens, nobody’s going to want a phone they know is missing out on vastly superior connectivity for a lack of compatibility.

Early 5G adopters will also probably be teaching these carriers and phonemakers hard lessons which those consumers, in turn, will pay for. Prices for 5G phones will be high, and the differences between the first and second generation of 5G phones will probably be significant - meaning that early 5G phones could have pretty nasty depreciation curves, especially if battery life or connectivity prove to be less than stellar. Verizon’s HTC ThunderBolt was the prime example of this in the LTE era, with its atrocious battery life, middling performance, and high MSRP relegating it to near-doorstop status in terms of resale value.

So, will the benefits of a first-generation 5G phone be worth the headaches? Maybe to the right person with the right use case in the right place, but probably not for most of us. We’ll be keeping a close eye on 5G phone launches in 2019, because there’s a strong chance we’ll be seeing a lot of one-off carrier-manufacturer collaborations that lightly rebrand and redesign existing phones specifically for the purpose of showcasing 5G technology. Non-5G phones will continue to launch, too - probably even a lot of high-end ones. By 2020, though, it’ll probably be 5G or bust here in America, and we can only hope carriers make quicker progress on working together than they did with LTE.