One Big Library.

Here's something I wrote two years ago:

"sometime by summer 2009, Apple will introduce the MacPad, a keyboardless tablet computer with a wholly reimagined UI based on multi-touch."

By many estimations, this is on track. All the stuff they've added to the 3.0 OS seems to make it possible to do almost everything you can do with a computer with an iPhone. Add a multicore processor, bluetooth, a plugged-in device API, and ZeroConf support (finally!) and this is getting very close to a general computing device.

Which is what I want. Only a little bit bigger.

The challenge here seems to have been to find the right balance between shrinking the mac laptops and growing the ipod/iphone. When I wrote about "MacPad" I presumed they would be shrinking the laptop. But now it seems more like they're growing the ipod/iphone. They've built up the restricted OSX environment on the phone now so that it can handle most any kind of app except for a terminal, a decent IDE, or a decent word processor. An "iPhone Pro" at 8"x5" wouldn't get quite there, but with a shape and features just like the iPhone, but bigger, it'd be just shy of that. A "MacPad" smaller than the Air in a tighter package with hideaway keys and multitouch would be just enough, but might make a crappy phone.

Like plenty of other people, I'm still convinced this is coming soon, but I'm not sure whether it'll be one or the other or both. Can't wait to find out, though.

Just got yet another new mac. Here's what the first 24 hours looked like.

DONE
system updates && tweak java prefs
istat menus
natsulion
textmate
scrivener
colloquy
adium
bashrc
vimrc
firefox w/plugins
  firebug
  webdev
  zotero
  jsonovich
  neo diggler
  elastifox
  switchproxy
privoxy
jungle disk && grab useful docs from s3 cache
quicksilver
audio hijack pro
skype
nodebox w/libraries package
processing
quicklook plugins
  betterzipql
  qlcolorcode
xcode inc. x11 sdk
instantshot
macports
ipython
readline
camino
wget
sqlalchemy
bzr
hg
git
screenflow
solr
couchdbx
adobe reader
vmware
  xubuntu iso, config guest, update packages
boxee
arduino
mysql
docs
  python-2.x
  nodebox
  sqlalchemy
  processing
transmission
vlc

DO
iwork
OO.o
tunnelblick
django

After that, it's a very useful machine.

There's lots of flashy and useful stuff in OSX 10.5 (aka Leopard) which received a lot of attention when it was first released. After several months using a lot of these (time machine in particular is a great addition), I've stumbled onto what I think is my favorite, and the single most useful new thing, and I don't even know for sure if it's new, or if it was ever mentioned in the release hype. It's this:

The finder now sports shortcuts for a time-based view of files you've used somehow. If you're like me and use a lot of apps all the time, and download a lot of stuff all the time, and muck with a lot of different files all the time, the "Search for Today" shortcut (highlighted above) is dreamy. It's like a Lifestream but without the failed proprietary windows-only app: it simply gives you the fastest way to get to "that thing i just used/downloaded/moved/executed/whatever."

A finder window dedicated to keeping this view open is now a fixture on my desktop, and I use it constantly. Wonderful!

Attached at a link at the bottom of this entry is a dashboard widget that lets you keep a list of items from unalog right on your OS X desktop. You can see 5, 10, or 20 recent items from the feed as a whole, or for any particular user or tag. Here's an example, for the tag "xkcd":

It was pretty easy to do using Dashcode and its RSS widget template. If you want to try writing your own widget, you'll find it easier if you already know some javascript+html and have some minimal event-driven UI development background to figure out which events need to happen and when. Or you could just grab some existing widgets you like and do what they do. I'm attaching the dashboard project source for this, too, in case you want to take this one apart.

If you use unalog and OS X, give it a try!

There are some interesting things that the iPhone do at Macworld seems to be telling us if you look beyond the immediate. Like, I have no problem believing that they'll capture 1% of the cell phone market within 18 months, because so many people with older players and crappy phones will add up the totals and say "why not get an iPhone for the same amount?" I'm not worried about Steve Jobs indicating that they'd hold development close to the vest. Instead of 1000 flowers blooming, you get interested developers funneled through some sort of licensing and certification and testing program so your iPhone doesn't crash a lot. People want reliability, and they're used to it in a phone, so there's every reason to fear user-mod-driven crashes.

But I don't think any of this is the point of what iPhone has to say to us. Beyond iPhone as *product* is iPhone as *inflection point*. The Apple ][, Macintosh, OSX, and iPod were similarly inflection points that meant nothing looked the same afterward. If we presume that iPhone doesn't even have to succeed to have the same magnitude of effects, it gets easy to guess what Apple will do next and be confident that you'll be right.

First off, they'll refresh the hard-drive iPod lineup with similar widescreen, multi-touch UIs (it certainly wouldn't help the Shuffle). They'll need time to work the kinks out with the early adopters, first, so this might not happen for a year or even two. But what can they possibly do to the current iPod line to make anybody satisfied with their somehow now-antiquated look and touch-wheel UIs short of dropping prices dramatically? I'm guessing we'll see a high-end hard drive iPod with widescreen and multi-touch sometime in 2008 at the latest.

But that's just the start. The vision of replacing hardware input devices on semi-limited use consumer devices is *huge*. In recent years there have been several attempts to make the keyboard "soft": like the projected-light keyboard, and the Optimus keyboard where the keys all have their own color Uis in them so you can soft-swap the keyboard for gaming or i18n or other purposes. The thing is, have you ever seen one of these in person? Do you know somebody who owns one?

No, you don't, because they're both cool but ill-fated one-trick ponies. If you've watched Jeff Han's TED talk video you'll know that the multi-touch UI is a huge departure from what we know and so broadly capable that it might just change everything we know about UI development. And for Apple to grok that deeply enough to devote so much engineering and PR willpower into pulling this one little phone off says that they're betting multi-touch UIs have legs far beyond limited-use devices.

During the Macworld keynote Steve Jobs made the point about how they've "learned from the iPod". They take this stuff more seriously than most anybody out there, and surely one of their goals with the first iPhone is to learn from it so they can figure out just how far multi-touch can go up the food chain. My hunch is that it can go all the way - that *every* computer form factor we're used to using today (laptop, pda, desktop) can be ultimately improved dramatically with multi-touch UIs, and that Apple already has a roadmap for doing just this in mind.

Why do I think it can go all the way? I'm not sure I can explain it well, but here goes. I'm a bit of a dabbler with UI paradigms - I'm not a 3D-savvy coder so I've always been a few steps removed from trying lots of things out myself, but since I took a course in Information Visualization with George Furnas at UMich during library school (1996 I think) I've tried to keep an eye out for improvements and implementations with large-scale takeup, and tried repeatedly to work on fun new UIs myself. Here are some images of ones I've been involved with.

So far, I'm not sure I've seen many, and the ones I've worked on while at Yale (DubMed, above, a circa-1998 java applet for visual query construction, and the MeSH Visualizer, below, a circa-2003 multipane vocabulary browser with AJAXy popups and highlights and scrolling in SVG) were cool and fun, but ultimately rather limited in potential.

Sure, the cognitive coprocessor model is basically a given at this point - you see it in sliding AJAX widgets, in the OSX dock, in UI widgets like menubars sliding open and closed. And the fisheye view model is relatively well-known too. But have you used *any* 3D UI at any length, ever? Or do you navigate your filesystem in treemaps? I know there are high-end visualization libraries out there, and that lots of data mining applications have glossy visualizations. And there's touchgraphs and lots of stuff like Grokker's map interface but, seriously, do you or anyone you know use those with any regularity?

Back in Pf. Furnas's class we were able to poke at the Pad++ environment, which appears to have lived on in Piccolo. This is way cool stuff, but I'll tell you, there's a huge limitation with all of these paradigms, and it's the mouse/keyboard/window/scrollbar input device model. We can go up and down and left and right and float in XY space really well but navigating zoomable 3Dish spaces like these just don't work well. Did you ever try navigating a VRML world with an SGI spaceball? Grokker's responsiveness to the mouse is great but its circular anchoring counters the way our brains are trained and our devices are forced to work today - X and Y axes and scroll a lot. These variants didn't work that well ten years ago, and I don't know of any widely-used interfaces that really are better now.

There's a reason google's simple search box and paged result lists worked so well and we keep coming back to them after trying stuff like grokker. It simply fits what we're used to doing with our input devices better, and therefore lets us get at what we want faster, period.

I think multi-touch UIs can solve many of these problems. I'm *certain* multi-touch could have made DubMed and the MeSH Visualizer a heck of a lot more engaging. And I'm pretty sure that once a few million people get a feel for it, it'll take off like wildfire.

I think Apple thinks so too.

If I'm right, where does this lead to? After refreshing the iPod lineup, what does Apple do next? I'm thinking it's tablet time. Imagine how cool a multi-touch UI tablet would be. No more stylus, right? And the bigger the touchable surface, the more powerful and wide-ranging the soft, "swappable" UI widget set could be. With a simple paper-sized tablet you could even project a keyboard easy enough for common typing tasks - though it wouldn't have to be QWERTY, come to think of it. :)

So remember, you read it here first - sometime by summer 2009, Apple will introduce the MacPad, a keyboardless tablet computer with a wholly reimagined UI based on multi-touch.

(Er, hold on a sec... /me googles... yeah, okay, there's this, but they didn't know about the multi-touch bit. So nyah. :P)

That's as far as I've gotten with this - I'm still deliberating over whether notebooks could be wholly replaced with tablets. It's hard to know. Maybe if a text-input method for multi-touch is developed that's as good or better than QWERTY keys with loving tactile response, the answer's yes, and if not, I'm not so sure, but I wouldn't count it out. One of the big rumors leading up to Macworld was ever-bigger screens, so maybe something like huge touch screens with voice recognition is the way up. Maybe the "minority report" interface is really coming soon.

I'm not sure. But I'm not selling my Apple shares anytime soon.