I don’t see how The Great Metro Experiment will turn out any other way.
The initial word was that Microsoft, at long last, had gotten the memo and was going to deliver a true RISC tablet. The bonus would be that theirs will run a new version of Office for ARM!
This summer they released a preview of Office 2013 that runs on their shiny new OS. Well, sort of. It runs on the “Desktop” environment within Windows 8. It also runs on Windows 7 computers.
That means it isn’t really a Metro app. Instead, Microsoft uses the euphemism metro-style. Metro-style uses the new fashion statement of the Metro API, but runs in Desktop. It requires a context switch from the Metro environment to the Desktop environment in order to operate.
Do you think anyone will notice?
No one who’s been watching should be surprised. They still don’t have a proposal out on how Metro itself can be scaled to support the productivity applications that their customer base actually needs.
The subsequent review in Ars noted as such. It's very diplomatic, but their reviewer is no dummy. Peter Bright, their Microsoft man, added a supporting article about Office 2013 on a Windows 8 tablet. I haven't tried this — I've only tried Office 2013 on an Intel desktop — but I can imagine that Mr. Bright speaks the truth:
Using Office 2013 on a touch machine is, at least in the public preview version now available, a tremendously frustrating experience. Even with the auto-hide ribbon, the Office applications are simply too complex to cope well with half their screen being covered up by an on-screen keyboard, and their interfaces are far too big for a simple band-aid such as “make the ribbon spacing a little larger” to be anywhere near adequate.
Unfortunately, as soon as one ventures beyond mere reading, the experience becomes unsatisfactory. Finger users attempting to make edits will find themselves regularly dumped into interfaces simply not designed for imprecise input, and even if they stick to the “main” user interface (the ribbon and pop-up toolbars), that interface works poorly. The interactions with the on-screen keyboard are frustrating and the interface is cluttered, leaving too little of the working area actually visible.
As things stand, far from being a valuable feature of Windows RT, the Office 2013 applications threaten to make it worse.
Microsoft did their best, but the flagship productivity suite doesn’t co-exist peacefully with the new Metro matra. What we’ve got for now is a forced marriage: Desktop is here to stay.
This is nothing like what Apple did. When they finally released their Mac OS 9 replacement, they included a compatibility environment, Mac OS “Classic”, that ran legacy applications until the third parties had time to rewrite them for their new Unix-based OS. After a few years, it disappeared.
This time, Microsoft’s “legacy” environment is the home for the updated version of their flagship productivity software! Apple has never done anything like this: Once they shipped their Unix-based OS, they never looked back. All the new Apple apps were designed only for the new operating system.
What does that say about Windows 8 as a whole? What’s happening is remarkable; they’ve reduced their own customer base to second-class citizens. I’m sure they’re eventually going to notice.
What’s Wrong with Metro?
Force-feeding their new tablet OS to their existing customer base is what I think will result in the Great Train Wreck of 2012. This Metro-style version of Office is part of their unique approach to tablets.
It’s so different from anything that has been done before. When Apple replaced their original OS with Mac OS X, they vastly improved the foundation on which professional applications could be built. After all, it was Unix under all the fluff.
That’s for starters. As I played with the Preview release of Metro, I concluded that there is a second-level problem that is hardly ever addressed. Marco Arment brought it up in a blog post.
Metro’s design principle in this area seems to be to let “the content” be the design, but that doesn’t work for a lot of app types. And even for apps that incorporate appropriate “content,” customers expect more from the design of the controls and non-content screens. (Believe me.)
If designers create beautiful, rich, iOS-style Metro interfaces, they’ll look garish or out of place. And if they follow Metro’s lead instead, there’s a good chance that everything will look stark, bland, sterile, and undifferentiated.
He sure ought to know, as his company ships a 3rd party app that is superior to the Apple counterpart. That’s the idea. The API should be rich enough so that independent developers can implement great ideas of their own or do more engaging or more powerful versions of the first-party apps. That’s what Instapaper is. That’s what Downcast and Plex are.
Marco noticed the same thing that I did. When I finished playing with Metro, it actually reminded of a deliberately “dumbed-down” environment. Specifically, it reminded me of those notorious “Dick and Jane” readers that millions of kids were subjected to. “Dumbed-down” readers were all about deliberately limiting readers to a controlled vocabulary.
The motivation for the reform was based on the fact that a controlled vocabulary also led inevitably to insipid, bland, and vapid stories. A critic noted that:
By 4th grade, the Russian children’s reading vocabulary was nearly 10,000 words, while their peers in American schools had been exposed to a carefully controlled vocabulary of fewer than 1,800 words.
That’s why. My reaction to this initial version of Microsoft’s tablet OS is to dumb it down. Unfortunately, the intended effect is to produce exactly the type of apps that Marco talked about:
“...stark, bland, sterile, and undifferentiated...”
In a nut, that’s what’s wrong with Metro as the new API for all of Windows. At first blush, it looks like it isn’t the kind of rich, fertile, innovative API that inspires developers to extend it to meet exciting new challenges. The design goals of Metro don’t inspire developers to do their most creative and imaginative work. It’s “See Spot run!” as an API. By imposing this interface on their own productivity customers, Microsoft is asking for it.