Here’s Pascal-Emmanuel Gobry, writing about iCloud:

Here’s how Google and Apple’s vision of the cloud differ: for Google, the cloud means cloud + web; for Apple, cloud computing means cloud + software, with the internet stuff happening behind the scenes.

All of the cloud computing services Google offers to consumers, like email, word processing and spreadsheets, happen within the browser. To Google, the point of cloud computing is to replace desktop software with the web. 

Many — maybe even most — observers share this view of the differences between Apple’s and Google’s approach to cloud-backed software. I think this view is wrong. Where Gobry writes “for Google, the cloud means cloud + web”, he glosses over what “web” is. Web apps like Google Docs and Gmail don’t somehow obviate the need for client-side software. They just change where the client-side software runs, and what APIs it is written against.

In both cases, there are client-side user-facing apps, and back-end services and canonical storage running on servers in the “cloud”. We don’t talk about “AJAX” much any more because it’s become so ubiquitous. But what it means is client-side software written in JavaScript, running in a web browser, communicating with a server using APIs.

Client-side apps and cloud-based servers. Apple’s primary focus is on native Cocoa Touch and Cocoa apps running on iOS devices and Macs. Google’s primary focus is on HTML/CSS/JavaScript apps running in web browsers. Google is not getting away with less work. If anything, they’re doing more work, because it is harder to create good user experiences inside a web browser. Where Google benefits from its strategy is reach — Gmail and Google Docs run anywhere with a PC-caliber modern web browser. Cocoa apps run only on Apple-made devices.

Neither company is dogmatic about these priorities, however. Apple has web apps at MobileMe — which I believe will soon become web apps for iCloud. Google has native apps written for Android (and don’t forget, for iOS, too). But there’s no mistaking where each company’s primary focus is.

The mistake — perhaps this is where Josh Topolsky went wrong — is to think that what you see in your browser when you type gmail.com is Gmail. It’s not. It’s a web-based client to Gmail. Admittedly, it is the flagship client to Gmail — the place where you can manage everything regarding Gmail. And users can and even should think of it as being Gmail. It’s a complete encapsulation of a powerful fast email service. But a major chunk of it is a client-side app written in JavaScript. (View source in your browser and see.)

I’m biased, insofar as I consider Apple’s strategy more appealing than Google’s. But that’s because my interest lies in having the best possible user experience — the best-looking UIs, the lowest-latency responses, the smoothest animation, the most elegant designs. I share that interest with Apple. Google’s interest is in reaching the largest possible audience. That’s why I chose, at the outset of this paragraph, to say that I find Apple’s strategy “more appealing than”, rather than, say, “superior to”, Google’s. Apple’s strategy is correct for optimizing the quality of the user experience. Google’s strategy is correct for maximizing the number of users for its apps.

But don’t make the mistake of thinking that Google’s strategy involves less work. That’s what I meant last week, when I wrote:

Google’s frame is the browser window. Apple’s frame is the screen. That’s what we’ll remember about today’s keynote ten years from now.

There’s a simplicity-based argument in favor of web-based apps. Nothing to install, no data that the user is responsible for managing and backing up. Apple solved this with the App Store, though — local native software with truly simple, obvious, easy installation and complete encapsulation of data. Yes, web apps are one solution to the problem of Mac and PC users having to be, to some extent, system administrators. But Apple’s App Store model shows that there are ways to solve that problem without eliminating native apps.

Josh Topolsky, in his piece questioning Apple’s commitment to web-based apps, wrote:

There is no native application for the Mac or iOS that replicates the shared document editing of Google Docs; there’s no mail application that exists for the Mac which will allow me to access my important information from anywhere in the world with or without a device in hand; there is no photo sharing service for iOS or the Mac which is as flexible or accessible as Flickr.

The first clause isn’t true, I’d argue. Shared document editing is not an inherent advantage of Google Docs being web-based; a native client could do it just the same. As the document changes, those changes are reflected live in your browser by way of a stream of API calls between your browser, executing on some device in front of you, and Google’s remote web server. Shared document editing is a difficult problem to solve and Google Docs deserves credit for solving it well, but it has little to do with being rooted in a web browser.¹

The third clause is about public sharing: web sites versus web apps. And agreed — Apple hasn’t addressed how iCloud will replace MobileMe’s galleries in this regard, and Flickr is better than MobileMe’s gallery anyway.

But that second clause — “there’s no mail application that exists for the Mac which will allow me to access my important information from anywhere in the world with or without a device in hand” — is key to Topolsky’s argument. Access anywhere. Where his argument falls down is “without a device in hand”. Many people may depend upon shared or even public PCs to access their Gmail and Google Docs accounts. But that’s a different audience than Apple’s. When am I without a device in hand? Never. As with MobileMe, there will be web app interfaces for iCloud for those times when all you have is a web browser on someone else’s machine, but Apple’s vision for “access anywhere” is “iPhone everywhere”.

I love the cloud. Last month I factory-reset my phone, and all my emails and Market-bought apps downloaded back on to it automatically. When I switch computers, Chrome will set itself up with all my bookmarks and passwords. And I don’t know what I’d do without Evernote giving me remote access to every thought I’ve written down and website I’ve clipped over the past few years.

The cloud is fantastic — and yet it’s flawed. What happens if I need to read something stored there while I’m on a long underground journey? Or if I’ve got a 300MB file but am on a slow Internet connection? Or when the server storing my entire MP3 collection goes down? Sometimes you can’t beat portable storage. But I don’t want to carry a USB hard drive around with me all the time…

Enter the Atrix.

What is the Atrix?

The Motorola Atrix is a dual-core Android smartphone which came out earlier this year. The tech specs are all well and good(dual-core 1GHz processors, 1GB RAM, 16GB internal memory, 5MP camera), but they’re not the interesting part of the phone. This is:

Webtop

The monitor, keyboard, and mouse are all interacting with the phone, as if it were a desktop tower. Motorola call this setup Webtop.

Even cooler is the Lapdock:

Lapdock

It’s an empty laptop shell into which you can plug an Atrix. As with the Webtop, the keyboard, mouse, and screen all interact with the phone itself, rather than with any internal computer components.

So What?

I said I don’t want to carry a portable hard drive with me all the time, but of course I’m already carrying something much better: a smartphone. My Android has gigabytes of storage, and can automatically sync my emails, photos, MP3s, notes, and anything else I want it to, automatically. I don’t have to worry about having a working internet connection at the point I want to access a file; I’ve got every file I need in my pocket.

Now, I can’t say yet whether the Atrix itself is any good; though I’ve seen it up close, I haven’t had any hands-on time with it. But I am very excited about the possibilities. Imagine walking into a library or campus, sitting at a computer desk, plugging your phone into a dock, and having everything instantly set up exactly as you like it. There’d be no need to worry about restrictive security settings, or having to wait to download your bookmarks, or wondering whether the computer’s got your favourite web browser installed, because you’re using your computer — yet without having to lug a laptop around!

This feels like a more plausible future than having computer labs all running off the cloud.

Speculating Wildly

Continuing this trend of convergence, what might we see next?

Maybe we’ll get detachable lenses that fit smartphones, so you can take DSLR-quality photographs, edit them directly on your phone with an Android photography app, and post them to social networks via 3G.

Or perhaps your phone will merge with your car key, keeping your favourite radio stations, seat position, GPS locations, and aircon settings the same if you share cars.

There are plenty of possibilities for this approach in the gaming world too. In 1994, Nintendo’s Super Game Boy cartridge allowed you to plug your handheld Game Boy into a SNES, so that games could be played on a TV with a separate controller. It was a wasted opportunity, but with mobile games being so popular and Sony showing interest in the form of the “PSP Phone”, I think we’ll see someone try this again.