wrong

John Gruber is wrong. It doesn't happen much.

Yesterday's Daring Fireball took on a useful thought experiment proposed by Marc Andreessen. Here's the experiment -- extracted from a Michael Copeland piece in Wired by way of DF:

Let’s say we all grew up in tech world where we only used tablets and smartphones. Then one day, someone comes up to you with a 27-inch display hooked up to a notebook. You could have everything you have on your tablets and smartphones, and then some. Except you don’t have to download anything or update it. Everything is the latest and greatest, and just one click away. If you are a software developer, there are no gatekeepers telling you if your latest creation is approved, or when you can add the latest flourish.
“We would be like, wow, that’s great,” Andreessen says from his office at venture capital firm Andreessen Horowitz. “It’s why in the long run the mobile web is going to dominate native apps, and for the same reason that on the desktop the web dominates apps. When the web works for something, it works way better in a whole lot of ways than a downloadable app.”

Gruber dismisses this argument with two of his own. First, that the desktop is over forever, baby. Imagining a user with a 27 inch screen is like imagining a unicorn. Second, that web apps suck.

Gruber is probably right on both counts but it doesn't matter. Andreessen's 27 inch display doesn't have to be a desktop. It could be a TV. It could be a car.

ASUS announced an 18 inch Android tablet that doubles as a PC display (or vice versa) last month. It might be great. It might be a piece of garbage. I don't know. Beyond a certain size, gadget verdicts and sentences start to diverge. It's a hassle to administer the death penalty. The ASUS mondo-slab might be found not guilty by reason of insanity.

The overwhelming majority of Android and iOS devices sold to date share an essential property. They fit in a trash can. You can even chuck most of them into the skinny receptacles at Starbucks.

TVs don't fit or belong in the trash no matter how deserving the software. At some point, the

oft-caricatured smartphoning, ZipCaring generation will realize that it costs them $25 in rental fees just to go get the ZipCar to drag a dead gizmo to an appropriate e-waste drop off point.

The idea of a functioning gadget that stops working when its bores its manufacturer is less funny when it leaves a corpse in your living room or in your ride.

9to5mac reported earlier this week that Apple's advanced life support team were prepared to call it on the original iPhone on June 11. I have Macintoshes made before the iPhone was born that are still attractive and useful. I don't own a car newer than that iPhone. Until this moment, I thought of my second generation Prius as a pretty new car.

Large creatures live longer. This has to be the case for gadgets as well. If we are going to have connected gadgets larger than trash cans, then the web wins. I don't think there is any market for third party native apps tailored to these gadgets -- even if there should be. Things, by their nature, come in too many shapes and sizes to be worth targeting with native apps.

It doesn't matter that web apps suck. Essentially all software ever written sucks. That's why web apps suck. Native apps suck too. Web apps that advertise their native app counterpart might be a case deserving special scorn. Arguing about which sucks less accomplishes nothing. In fact, the only constituency for a pronouncement about which, on average, sucks less is middle managers who don't write and don't use the software they commission. These people might be half of the reason that software sucks in the first place. Independent developers and users have no problem arriving at their own conclusions without a debate.

Cisco sees a $14 trillion market in the 'Internet of Things'. If Cisco is right, then almost none of those things will be smartphone or tablet sized. If they were, they would be killed outright by smartphones and tablets or become them.  Anyone bringing a new smartphone-sized product to market in any segment had better make sure that it is edible, disposable, holds a smartphone, or is a smartphone in some combination. 

I have an ad supported flashlight app on my iPhone. The app puts the ad right below the 'power switch' for the skeumorphic torch. Here's a truth about the universe: Nobody who urgently needs an ad-hoc flashlight has ever wanted to click on an ad at that moment. Right now, the app tempts with a 'Free Game' ad for something. Seriously?

You can't do ad supported apps on Things. You can't easily monetize the user by mining photos and contacts on Things. You can't monopolize the user's eyeballs with Things. Combine those and you have nuked most of the technical rationale for native apps over web apps or web apps in Cordova wrappers on Things. 

That's the genius of Google Glass. Every refrigerator can become an ad-encrusted refrigerator. Every cordless drill can become an ad-encrusted cordless drill. Even if Glass were powerful enough to render the web app irrelevant -- the web still wins. Glass only works when you're looking at something. It's easier to be that something with the web.