This weekend, I set aside my G2x and bought my third Nexus S.

I'm not bragging, mind you. I certainly can't afford to own three $500 phones (I sold the first two). Quite the opposite, actually: after having recently owned both the G2x and the Atrix, I bought this phone with a feeling of resignation, defeat, and disappointment that I've been unable to find a proper replacement for a phone that launched nearly half a year ago. And while I'm well aware that half a year isn't very long at all for a normal smartphone-owning human being, I think we can all agree that I'm not normal. And frankly, I probably wouldn't have the job that I do if I felt comfortable about owning the same phone for six months.

So what keeps bringing me back to a somewhat cheap-feeling phone that lacks external storage, Wi-Fi calling, Gorilla Glass, and HD video capture?

I've thought a lot about that over the past 24 hours, and I think it has everything to do with quality of the overall experience, not just the hardware or the software alone. The Nexus S is my constant. It does everything pretty well, rather than a few things extremely well and the rest poorly. What I've come to realize -- for me, personally -- is that I'll take a reliable device with middling specs over a monster with crappy firmware every time. What good is a dual-core phone with a laggy UI?

Going forward, my advice to manufacturers is simple:

  • Don't launch a buzzword feature until it's more than just a buzzword. "Dual-core" is a prime example. The only practical advantage that I've found to dual-core so far (on Android, anyway) is that you get marginally better graphics in a handful of games. My Atrix and G2x were no faster than my Nexus S in day-to-day use -- and in many cases, they were slower. I get the impression that Android, as a platform, isn't yet optimized to take full advantage of multi-core processors, so for the moment it's little more than a checkbox for manufacturers to tick off. 3D cameras and displays are in the same territory.
  • Don't launch a product with obvious problems, even if you plan on fixing them down the road with a patch. First impressions mean everything. If you're launching a high-end smartphone, you're going to get a rush of power users flocking to it right away -- and if the browser freezes every few minutes or the battery is inexplicably charging at a glacial pace, that's a problem. Those power users are going to talk to each other, and they're going to talk to their less technically-inclined friends. The product's image is going to be tarnished well after the update rolls out -- you are working on an update, aren't you? -- and your entire brand is at risk of coming along for the ride. It's not worth it.
  • Keep it simple. Favor quality over functionality. I don't have any proof of this sitting in front of me, but mark my words: if I ever find out that an awesome-sounding product (take the HTC Sensation 4G, for example) could've launched sooner had it not been for tweaks that needed to be made in the manufacturer's software customizations, I'll flip a table. Of course, it goes without saying that loading stock Android requires less engineering effort on the part of the manufacturer, runs a far lower risk of harboring bugs, and makes it easier to roll out updates quickly... but that's a well-trodden argument that never seems to sink in with anyone.
  • Don't be fooled into believing that you can't differentiate in meaningful ways through industrial design alone. Take the Nexus S with its curved, blacked-out face, the iPhone 4, or the positively distinctive Nokia N8 (yes, Symbian^3 is a nightmare, but the N8's hardware is stellar -- and you won't mistake it for any other device). Deep software customization exists thanks largely to the wrongheaded belief that the endless parade of big touchscreen slabs makes it difficult for manufacturers to leave their fingerprints through hardware design, and I think we've got enough counterexamples at this point to demonstrate that it's a false assumption. Heck, the Nexus S is little more than a massaged Galaxy S -- but you'd never mistake one for the other.
  • Trumpet your commitment to Google's 18-month upgrade policy at every opportunity. Every couple weeks, one of my non-techie friends asks me when the next iPhone is coming out, because they don't want to buy an iPhone 4 if it's going to be replaced in a few months. To me, that's evidence that even "regular" consumers are thinking about this stuff. One of the reasons I love the Nexus series is that it's directly connected to Google, and I know they're pushing updates just as quickly as they can. I've always got the latest and greatest (true, it took the Nexus One a little while to get Gingerbread, but it still came faster than everything else on the market). Manufacturers should be making it crystal clear that they're a part of this upgrade pact and that they take it seriously.

I'm not proud of this Nexus S in my pocket, and I can think of at least three or four phones on the market right now that I'd theoretically rather have. Ultimately, though, this crazy spec sheet arms race is doing nothing but confusing customers, lowering average quality, and creating frustration where none needs to exist. And at least in my case, it's slowly bankrupting me.