saturn return to sender

This got really long and rambly and bitchy and needs to be edited, but after writing it all I refuse to not post it and it’s not worth the effort of editing.

First, del.icio.us/popular led me to DesignShack, ostensibly a paragon of and gateway to excellence in web design. It immediately pissed me off by mapping the arrow keys to navigate to pre-set pages. WTF? I sometimes scroll webpages with the arrow keys. Okay, yeah, nifty: you got the web page to respond to arrow keys. Woo-woo. Now why, exactly, is that a better design than just providing graphic navigation? At least provide some graphic feedback to let users know that their arrow keys now have this completely aberrant behavior. The site kept jumping to some page I’d never seen before because I just wanted to scroll the page down … arrgh!

Meanwhile, another del.icio.us link … this article is dumb. Guy says the “Mac style” of user interface has reached its limits:

Displaying commands in menus, toolbars, and dialog boxes works with a limited number of elements. But Microsoft Word 2003 has 1,500 commands, and users typically have no clue where to find most of them.

Um, hi. That’s because it’s crappily designed. If there were a sensible, straightforward data model behind a sensible, straightforward presentation of it, there would probably not be any problem with depth or breadth of manipulation options. Present a set of related operations under an articulate top-level description. Use visual cues as part of that description, if appropriate. Find innovative ways to let users drill through these hierarchies efficiently when it makes sense to do so. Make the data model consistent and apparent.

Maya remains to me one of the best examples of a seriously sophisticated program with a justifiably tough, but ultimately beautiful UI.

But more importantly, in the case of Microsoft, design your program sensibly from the beginning and don’t glom on features endlessly to a framework that was never intended to deal with them. 1,500 commands, huh? It’s like having a Swiss Army knife the size of your thigh. Dude. I recommend InDesign and Notepad or something. Your life will be easier.

Another WYSIWYG downside is that it forces too much manual labor on users and requires a stretch of imagination to envision results in advance.

Uh, wha …? It doesn’t per se force anything. WYSIWYG does one thing: provides immediate feedback to user manipulation. So … how is that bad, again?

His alternative?

The next version of Microsoft Office (code-named “Office 12″) will be based on a new interaction paradigm called the results-oriented user interface. As the demos show, the most obvious departure from the past is that menus and toolbars are all but wiped out. The focus is now on letting users specify the results they want, rather than focusing on the primitive operations required to reach their goals.

Sounds … intriguing. I envision a natural language interface. “I want a nice, clean report on the latest green building technologies,” you tell Office 12, leaning back in your Herman Miller Aeron, sipping your morning latte. “Give me a limited palette of green and burnt umber, with Art Nouveau borders a la Horta. Two columns, one inch margins, a series of appropriate graphics lining the top. Go.” It’ll be like having a design intern there in your office.

Not quite.

The new interface displays galleries of possible end-states, each of which combine many formatting operations. From this gallery, you select the complete look of your target — say an org chart or an entire document — and watch it change shape as you mouse over the alternatives in the gallery. The interaction paradigm has been reversed; it’s now What You Get Is What You See, or WYGIWYS.

Oh, good. More Microsoft-knows-better-than-you UI. That’s just what I was waiting for in a next-generation user interface. Limit the users’ options and pre-empt his or her decisions because the user is dumb.

AAARGH! I don’t want a list of pre-compiled tasks or outcomes. I want a model of how the tool works in my head so that I can use it as a tool, not a catalog of outcomes. I want the program to lead me into developing that model as the need arises.

Look, in my not-so-humble opinion, Microsoft’s UI design philosophies are exactly wrong. I don’t think they’re really philosophies thought through from the ground up so much as stop-gap measures to try to patch up bad engineering. Then they spend a butt-load of money in marketing and research to justify their “innovation.”

Microsoft’s idea of innovation is that when you ask for a sturdy power drill to build yourself a chair, they take you to Target and show you the furniture aisle. That’s not innovation, that’s capitulating to the lowest-common-denominator market forces. What you wind up with is software that excels at being mediocre and enabling mediocre content. In every way.

The sickening part is that this is demonstrably all the market demands.

We know from user testing that users often demand that other user interfaces work like Office.

Maybe there’s a name for this effect, but there’s a certain sense of ownership people acquire when they manage to fight through painfully bad interfaces. They learn to use a tool that is broken and they justifiably feel entitled to gloat and demand that their hard work transfer to other tools. They insist that other tools be broken in exactly the same way.

Form•Z is a great example of this. Its interface is completely abhorrent, due I’m sure to its 10+ year-old legacy codebase. But after fighting long enough, I found myself working reasonably efficiently with it. Grumbling at indefensibly stupid behaviors, and rebooting from crashes constantly, but that ctl-S keystroke after every model change just disappeared into the accustomed obnoxiousness of the workflow. It’s like getting used to a putrid odor.

Pro Tools has a similar legacy of bad UI, though it’s worlds better lately.

This kind of familiarity with broken tools in specialized domains creates a culture in the industry. If you mastered the broken UI of Pro Tools, you were valuable. You had a certain mystique. You were a shaman. Pro Tools is an awesome program in the hands of a master, went the common wisdom. (And not talking musical or aural design talent here: just competence with the program.) The beginner complains because he doesn’t know any better.

The Form•Z forum is packed with new users complaining about the “irregularities” in the program, as autodessys calls them, only to be chastized by the seasoned users. You just don’t understand the program, they say.

This is so, so wrong.

There’s no reason for the fight to begin with. There is no reason for mystique when it comes to using a tool. There is skill and talent, yes, but mystique? No. The act of using the tool itself should be its own training. The tool’s model of operation should make itself apparent, unfolding as the user’s needs increase.

Of course it’s extremely difficult to design tools that — first of all are designed rationally to begin with — and then exhibit their own rational workings transparently. Such an ideal is, yes, very hard to attain. But that guy’s article is supposed to present the cutting edge of interface design, and it’s completely off-base in this sense. A “gallery” is not a tool. If Microsoft is selling catalogs, then yes, “results-oriented” UI is appropriate. I prefer to buy tools, not pre-fab mediocrity.

Rant off.

delicate

This one I also quite like.

winter will be cold

One of the more plausibly professional-looking photos I’ve taken. I suppose it would be better if the leaves hadn’t been past their autumn peak. Still, I think it’s pretty spiffy. Most of the credit, of course, goes to the aspens, which were very tall and photogenic, and to some selective clear-cutting by the forestry service, which left these tall, photogenic aspens very much out in the open.

Update from previous post:

CNET has posted this story, proposing that video downloads are not coming anytime soon, mostly due to financial protectionism by satellite and cable companies like Comcast. I wonder if Apple might not sneak under their radar, as they have an “insignificant” PC market share. Of course there’s no reason an iTunes video store wouldn’t be cross-platform. Regardless, I wouldn’t put it past Jobs to shoulder his way to the forefront, toes of Comcast be damned. But then I wouldn’t put it past Comcast to do something dick-ish like purposely degrade traffic to/from Apple.

Come to think of it, though, a video service from Apple won’t happen until they have the hardware sales to back it up. They don’t make any money off of the iTunes store, they make money from iPod sales. They’ll have to have some sort of hardware device dongle to play videos before they can afford a video store. Particularly a cross-platform video store. They would make precisely $0 from renting videos to iTunes users on Windows machines.

I don’t know if an 802.11g wireless video device on its own would have high enough margins to justify such a move, unless they price it up in the iPod price range. Hmmm. I suppose they could.

* edit *
Oh, I get it. The mini will be re-incarnated as a wireless media server with video out. That will be the Apple video dongle. Cute! I like it!