In 2009, Apple added a video camera to the otherwise perfected iPod Nano. It was a sign: While one might have had to be quite prescient to imagine exactly what would come after it, it was obvious that Apple felt it had reached the end of what could be innovated with that form factor. The 2010 reinvention of the Nano was the only logical next step, at least for Apple.
In 2009, Apple released Snow Leopard, whose emphasis on optimization and subtle improvement seemed to signal the same kind of state for the operating system as the addiiton of a camera signaled for the Nano.
I think it's safe to say that what comes next will be fairly big.
I have a suspicion that among other things, 10.7 will introduce some new conventions to the basic problems of window management (though I might be a little biased). Apple has been at the forefront of dealing with it ever since Exposé debuted with 10.3, but in the years since the iPhone, iOS has thrown this problem into sharp relief. Part of me hopes they might even announce something that bears a resemblance to 10/GUI's linear window management -- I've unfortunately proven I can't really do much with the concept myself so far.
I have a suspicion that among other things, 10.7 will introduce some new conventions to the basic problems of window management (though I might be a little biased).
For anyone not sure why mortenjock is biased, I recommend viewing his concept video 10GUI http://10gui.com/video/
The recent Magic Trackpad would tend to support your ideas here.
Off the top of my head, I guess I'd just count myself among the ranks of the inventor of Cover Flow, the creators of Konfabulator, and the creators of Quicksilver. It happens, and the best thing to do is just to double down and never stop thinking, exploring, and making. As disappointing as it might be, it would be an enormous validation.
My experience with the Apple recruitment process didn't tell me much about the company's plans for the desktop (I didn't get far enough to sign an NDA), but I did learn something important: Their interaction designers, the concept guys whose explorations eventually filter down into shipping products... are also developers. Apple hires the best of both worlds, idea people who don't even stop to sketch ideas out on paper but head straight to XCode and an NSWindow. That's literally their process.
The best thing that came of it was that it inspired me to set out to learn Objective-C and Cocoa. I'm not as far with it as I'd like to be yet, but it's opened my eyes to the possibilities of becoming a designer/developer. Even if, in the future, I'm still the archetypal designer looking for a technical co-founder, hopefully I'll be good enough that I can at least carry my own weight.
More importantly, you'd have an intimate understanding of the trade-offs required for various designs. Of course, the flip side is that you have to learn to ignore those trade-offs at times or else you start designing based on trade-offs instead of trying to push the limits.
FWIW, I'd love it if Apple did something like 10/GUI.
> I think it's safe to say that what comes next will be fairly big.
My feeling is that this unlikely because it doesn't seem like they've had enough time or resources to dedicate to something big. It's been one year since the release of Snow Leopard and since that year they've released an OS on a new hardware platform and a heavily updated iOS for the iPhone.
I think it's far more likely that this is an incremental release focused on user concerns with perhaps some other new software.
Just dropping some speculation here, as many are wondering what Lion can possibly offer that's new (and I don't mean a fix, but new).
Touch. Deep integration with touch. As you see with the release of the Magic Trackpad, Apple is retiring the traditional mouse paradigm and moving on to versatile and complex gestures to aid in window management. Four finger swipe for expose and show desktop has been godsend for productivity, as has three fingers up and down for switching between h/m files in xcode. Let's not forget swipe to go back and forth in browsers, pinch to zoom, rotation, etc... None of which is possible with the standard mouse.
And peruse through Apple's touch gesture patent library and you will see that they haven't implemented a fraction of what they have patents to. You have neat stuff like three finger pinch (possible to close/hide a window, save), non-adjacent finger scroll (ex. thumb and ring finger, note that Magic Trackpad is big enough for this one, but existing options on MBP are not), and several others involving rotation and unconventional finger combos. Expect for these gestures to be baked deeply into the new iMovie, iPhoto, and iTunes. Seriously, can you imagine alternate paradigms for simply grabbing video from a bin once you account for different numbers of fingers involved in the drag?
A preview of Lion with these features will allow them to sell the next generation MBP with a new, much larger trackpad to accommodate the new focus on touch and touch gestures. If Apple scores a hit with this new focus on touch, it will be another competitive headache for thin, underpowered clients that serve up web content and can't handle advanced video and photo editing, not to mention lack large surface areas for touch gesturing.
What's in it for Apple:
If you think about this from a competitive standpoint, most of Apple's competitors want to lift computing into the cloud--Google with Chrome OS, HP with WebOS, and even MSFT with Ballmer's pronouncements of "cloud speed". Apple's strategy, and profits, are vested in adding weights to computing to take advantage of powerful, local, GPU and CPU intensive tasks, thus slowing, or even reversing the shift towards browsers and web applications, in keeping us grounded in powerful, local machines. If Apple finally figures out a way to get your mom editing video with powerful and intuitive new touch gestures, how is a web application going to respond to this from the confines of a browser? How is Microsoft going to coordinate with clone makers to agree on standard sizes and performance of computationally complex multi-touch trackpads? Touch and media editing all the way.
Expanded trackpad gestures are a good idea, though geeks who are big in to keyboard shortcuts can already have a pretty great experience. You can do all your window focus and visibility with the keyboard. Add a 3rd party utility like SizeUp/Divvy and you have movement and reizing, too.
Like a bank that has become too big to fail, iTunes has become this massive hub that manages everything from audio and video syncing to mobile file and app management. Mobile device interaction not relating to multimedia needs to be offloaded to a renovated iSync, or better yet, Finder needs to be updated to integrate the iOS anti-filesystem. Photos, apps, app files -- all hierarchically listed with MacFUSE-like simplicity, but probably with a more context-appropriate UI than the standard Finder view.
I like the idea of Finder handling multiple types of filesystems. Seems like an elegant solution to this problem to me -- move device management out of applications and into the operating system, where that should be.
The elegance of that solution seems too much for Apple to ignore.
I wonder if this could prompt Apple to adopt Fuse as a means of filesystem abstraction in Finder?
Snow Leopard was all about the internals. UI changes were not a priority.
Apple has presumably learned much from their iOS work that will inform their thinking about how people want to interact with a device. I don't think we'll see a touch-the-screen interface, but we might see an interface that is less file/application oriented and more task oriented. It won't be a replacement, but it might be a new direction.
Oh, and some gradients and textures will be slightly different and 10,000 blogs posts will be written about that.
God, they fubared that. I can't count the number of times my wife complained about Snow Leopard's constant WIFI problems. It's sad when my iPhone has better connectivity then her MBP.
We need to get our iPad looked at, too. It constantly has problems, but my iPhone4 has no issues with local wifi connections. Between the iPad and the MBP, it's a crap shoot with connectivity.
I also complain to my wife that she needs to stop going on and offline all the time. She doesn't appreciate that. =)
Yes, and we've looked into that as well. We've tried multiple routers (we returned 2 prior to this one, thinking it was the routers at fault), and in each case the problem was always the same: iPad and MBP. Both of these have cases that can be searched for online, and lots of other people have these problems. For the MBP, it had something to do with Snow Leopard, as downgrading seems to help somewhat. For the iPad, their are other reports of it being a energy saving issue. We don't have issues with the iPhone 3GS, iPhone4, or the iMac when it's on the wireless.
We also considered the location of the router, but both the iPad and MBP suffer the problem even if they are in the same room as the router.
We also don't experience the problem on a netbook we have that runs both Linux and Win7. However, we don't use it that often, but it's never had the problem whenever we did use it.
It could simply be a hardware issue, but as I said, we aren't the only ones having this problem.
There were other wifi related bugs, too, fixed in 3.2.2. And people still continue to be plagued with wifi issues with the iPad. Just go to the support forums and see.
It has stupid limitations — files in use, can't put item to trash while trash is emptied. Still has only read-only support for FTP, locks up with WebDav, can't browse ZIP, etc.
Copying in Finder is much slower than with cp. On Samba drive it may need 10-20 seconds to replace 2kb file.
It handles permission errors erratically (sometimes asks for permission, sometimes fails, sometimes asks and fails anyway). UI for permission change is weak.
For me it was quite a disappointment because one of my main beefs with finder is the atrocious performance even on modern hardware and especially when network mounts are involved. Sadly pathfinder turned out to be even slower!
I don't need zillions of buttons and special "features". First and foremost I want a lean and fast view into my files, everything else is secondary.
As much as I hate windows; up to this day the windows explorer remains the most bearable graphical file browser that I have used.
I would love to see them release a Mac App Store. Of course, if they made it the only way to get new programs on a Mac, I would never upgrade, but I think providing a central market-place for software could be a very useful thing for a lot of users. Also, as a developer, I would love for there to be a way to sell apps that gives me the ability to reach a wide market, and also keeps me from having to run my own DRM system, or even my own payment system. The iOS app store is far from a perfect system, but I think this would still be a net win for developers. I think it would especially be good for iOS developers who need to have a traditional PC counterpart to their mobile app.
Of course, my pipe dream would be if they would release a Windows app store at the same time, with as little friction as possible for releasing apps on both platforms. I doubt that Apple would do that, but who knows.
- A new Finder. I want tabs (like TotalFinder) and two instances in the same window, so that drag+drop doesn't require multiple windows. Like this: http://dl.dropbox.com/u/12925/finder.png
Other than that, I'd like to see it continue to get faster. Booting, opening apps, etc.
Well, then I'm left with two copies and I have to go back and delete one. If OSX had a cut/paste, that might work better. Even so, with the way Finder currently works, it's easy to have 10-15 Finder windows going at any moment, so cmd+` can be a pain.
It has copy/paste for files, but you can't "cut" a file. This is a continual source of frustration for Windows power users who come to the side with cookies. My guess is Apple doesn't think "cutting" a file makes sense: you are actually moving the file or deleting the file. However, why it would suddenly make sense to "copy" but not actually have "copy" do anything is still confusing to me.
When I select "restart" (edit: at least when installing system updates), the computer should save application state and reboot instead of every application asking me stupid questions like "are you REALLY sure you want to quit web browsing?".
I want to select "restart", go to the kitchen, get my dinner, come back, and the computer be rebooted and my applications restored the way they were before.
Also, the developer tools should include a good package manager instead of having to use Homebrew or Macports (both of which are flawed in various ways).
The reason package management works better on Linux is not because there's a better package manager program. Rather, it's because there are hundreds of developers for Debian, Fedora, et al who put serious time into making sure the packages themselves work well. Few OS X users really care about Unixy packages, and even fewer actually work on one of the package managers, so bugs just don't get ironed out enough.
I don't see much Apple is likely to do to help. They could bless a package manager as official, include it in OS X, and pay for build-boxen and for people to work on it; but you'd still need a lot of community involvement, and that's not Apple's cup o' tea.
Apple could also just make a few things easier about building nix packages:
Currently the OS X linker will automagically look for libraries in /usr/lib and /usr/local/lib by default, this should become easier to turn off so that packages don't end up with hidden dependencies on stuff the user threw in /usr/local.
* A lot of libraries that are extremely common on other nix aren't included in OS X, eg: readline, iconv, gettext. But many of these are GPL3, so don't count on Apple being helpful.
Universal, backwards-compatible binaries are currently a PITA to build from the command-line. You need to specify the SDK, DEPLOYMENT_TARGET, and compiler separately; and then if you ended up using different SDKs for different architectures, you need to use lipo to combine the results. Some way to specify all this stuff at once, and use sensible defaults would be great: 'gcc --sdk=ppc/gcc-4.0/10.4 --sdk=x86_64/clang/10.6 ...'
One thing they can do is package-manage the Ruby/Python/Perl/Apache/vim/emacs/gcc/etc. that either comes built into Mac OS X or gets installed with the developer tools. Current package managers either have to work around that (Homebrew) or maintain an entirely separate repository (Macports).
The primary use of restart is to install system updates. Having my applications' states reset only means I have to go in and restore their states by hand every time Apple decides to release a new version of Safari.
I actually can't think of any time I've even used restart to reset my computer's state in the past couple of years.
Why do you think the updates require a restart? To reset the computer's state. I think the problem to be solved is updates requiring a restart (which is unacceptable in this day and age)
Yes, I have some idea of how computers work. And it would be fantastic if the OS could hot-load patches instead of needing a reboot.
It would also be acceptable if Safari could just remember what tabs I had open and reopen them instead of asking me stupid questions when I run Software Update and then canceling restart when I don't answer.
I'd think that could cause problems if the system frameworks or other libraries are changed by the update, but the apps are restored into RAM from the pre-update memory image.
Yeah you would want it to be a little smarter than I described, but not too smart. Web browsers would remember their tabs and windows, office programs would autosave and autoload temp files, etc.
I imagine there's a hundred tiny problems which make it infeasible, and it would require API changes, but damn would it be nice.
Yeah, I'm going back to Linux because I realize I miss apt-get more than all the goodies and bling OS X brings to the table. I checked and Ubuntu 10.10 will install happily on My Macbook 5,2 and I've checked my app list: Skype for Linux? check, Spotify for Linux? check, OpenOffice? duh, Chrome? yup, Tweetdeck? there's Gwibber, the list goes on ... Textmate I'll miss and Preview.app is super-duper but apt-get install I miss you!
In addition to Fink, there is MacPorts. Between the two, you can get a lot of things installed just as easily as on Linux. Of course, proprietary software is still proprietary.
Fixing all the problems that OSX is still plagued with. On a daily basis. Here are problems I deal with that are Apple related, almost on a daily basis:
* iChat and Bonjour. I cannot, for the life of me, get my iMac and my wife's MBP to chat, let alone do anything else iChat says it can do. I've not found a solution. I can occasionally see the MBP on my iMac, but I cannot do anything with it.
* Make iLife make sense. Videos are stored in iPhoto, but need to be transfered to iTunes to send them to the AppleTV. But I need to create them in iMovie. iMovie is for editing movies. iPhoto is for editing pictures, but also organizing movies. Movies are also organized in iTunes. Also, plugging in a devices starts up iPhoto by default. Setting this up silently isn't easy to discover.
* System Preferences. The best way to demonstrate the complete craziness of this is to explain that to disable the Remote Control, you need to go to the Security panel. I'm not even going to mention that fact that certain panels literally lie to you. No, that keyboard command you say I should use doesn't actually do what you say it does.
* Window Management. Sure, their are programs you can install for this, but it should be something handled by Apple.
* Fix Safari so I can log into their support pages. I'm mean, it's evil that I need to download Firefox so I can log into their support pages. Yes, I've cleared the cache in Safari, and cookies, etc.
Apple does great on the polish, and their system sits on top of a solid foundation, but in between that, the meat, is lacking.
Sure, OSX is stable, but it's still frustrating by default.
Have you tried playing with firewall settings? My wife and I had issues syncing our computers until we got the firewall setup right.
iLife isn't really part of OSX, even though it comes pre-installed. It usually gets a separate release on it's own so they can sell upgrades to people on older versions of OSX. The videos / iPhoto issue is annoying, but derives from the addition of videos to the iPhone. I'm not sure how they are going to handle that better.
It sounds like (based on this and a different comment) that you just don't like your Macs.
Oh, I like my Macs. I like the environment. However, it's not all roses and unicorn kisses as some make it out to be. The problem is when you want to do something remotely outside the default. Like, for instance, if you plug in your iPhone and you don't want iPhoto starting up and stealing focus, then it's a headache. Smart defaults work best, defaults that don't interrupt your workflow. Look at TimeMachine. It works without disrupting you. Awesome. It's their when you need it, but doesn't pester you. I complain, but their are also a lot of things I like.
As for the Firewall: I've tried so many things. Yes, the firewall was one thing I checked. I don't know. I spent too much time on it already that the thought of going back and spending more time is frustrating.
Like, for instance, if you plug in your iPhone and you don't want iPhoto starting up and stealing focus, then it's a headache.
Go into Image Capture and change the preference to not start iPhoto when it is connected. The reason it does this by default is b/c your iPhone has a camera so everytime you connect your iPhone the 'smart default' is 'hey there are pictures, let's suck 'em down 'n back 'em up!'.
I know how to do it now. =) I set it to auto-import and it does it without loading up iPhoto. I also realize why it does this. The problem is iPhoto does this, but their is no obvious connection for going into Image Capture from iPhoto to change this feature.
iLife and Safari are updated outside of system releases, so I wouldn't bet too heavily there. iChat will get an update though, surely with FaceTime support baked in.
I agree there are some core apps that have been neglected but overall usability in OSX in my opinion is still far better than the alternatives. The biggest issue seems to be that some applications have matured well beyond their original GUI so as new features have been added some of the simplicity has been lost. My guess is many of the core apps (Mail, iCal, iTunes, etc) will end up looking a lot more like their iOS equivalents.
I know iLife and Safari are updated outside the OS, but they are still shipped with new Macs. They are apart of the system you get when you open the box and plug it in. And they add to the frustration.
Essentially, I find it sad that my wife and I find it easier to email each other things rather than use something like iChat or file sharing. I can't count the number of times we're both online, but one can't see the other, or neither can see.
The problem is the foundation is mostly fine. They really need to start working on making the system easy to use though. Their are a LOT of areas were the polish is gone, and all you're seeing are warts.
Two things. One is a possibility, and one is a dream.
1. The Finder is going to change. What is becomes I don't know, but I think the filesystem will be further downplayed as Apple tries to take its vision of user friendliness even further. This is the possibility.
2. Window management goes away. Imagine software that didn't need its UI sorted and resized. Personally I hate even thinking about where something should rest on a screen. Given that there are tons of legacy applications out there that use windows in a variety of ways I don't see this happening. This is the dream.
-- Edit:
Graphics drivers that are on par with Windows should probably be added to the "dream" category.
How would window management go away in a desktop OS? By having only fullscreen apps à la iOS/Android?
I don't see that feasible or desirable. In a small format it works, but on a desktop I very often want to see multiple windows at the same time. (say, reading documentation with code, or multiple files open at once)
Sure, but the 10/GUI is managing the windows though. It's a different way of doing window management, but that's still what is it. I'm curious to know if ary had something else in mind.
The recent praise full-screen applications got at the 10.7 sneak peek doesn't resonate with me, but perhaps it will with others. What I want are applications that intelligently manage screen real estate themselves. My thoughts on this are still somewhat unpolished, but what I'd really like to see is automatic management based on need and importance. Clear, standard notifications (and as few as possible), cooperation between apps when two or more run on the screen, and a clear way to navigate between them. All the while there would be no title bars or min/max/close buttons to be found.
This isn't a call for more iOS in OS X (although it seems to be going that way) so much as it is a call to acknowledge the limitations of user ability and patience. It's something I think about a lot.
- Tighter integration/control of external stuff: AppleTV, Airport Express, etc.
Not that I have any hope of seeing this, but I would extend it to iPhones, iPad, etc. for things like sharing clipboards, using the computer's mic/audio to pick up the phone, direct access to the iPhone's camera from the desktop... basically make the relationship between the computers and devices more symbiotic.
File Vault is a poor hack. Needs logout to free disk space. You can't sudo into file-vaulted account. It makes Time Machine nearly useless: you can only restore entire home directory at once. It won't backup anything in home while you're logged in.
I must admit, one thing I've become absolutely dependent on in Windows 7 is when you press Start you can do a quick lookup of the program you're after by typing a few characters.
Sure OSX has the Finder and Spotlight but it's not quite as quick and convenient, in my experience.
But my big request is this: implement the TRIM command in the OS to open up the SSD choices. This is long overdue.
"Two thumbs down." I'm not rating your comment or spotlight or the start button. Rather, that's my mnemonic for Command-Spacebar, which brings focus to the spotlight search field. I can launch Terminal in 5 keystrokes: CMD-Space t e ENTER. Is the Start button really significantly better?
"We started off this article making it intentionally limited in scope as we weren't expecting, in a OS that doesn't support TRIM, to find anything all that interesting. What we found was the exact opposite: an OS that doesn't appear to be affected by SSD performance degradation [...]"
an OS that doesn't appear to be affected by SSD performance degradation
That may be true, but there isn't enough information in that article to tell. I think their other theory — that the Samsung SSD is essentially pre-degraded — is more likely to be correct.
You might try this: From the Finder, Cmd-Shift-A, which opens the Applications folder, then your first few characters, then Cmd-Down to open.
It seems like a lot of keystrokes, I know, and you have to switch to the Finder and it only works on initial characters (e.g. LIghtroom, not LightRoom) -- but since there's no search involved it "feels" extremely fast, which is often the thing that trumps all else.
The two buttons needed to trigger spotlight are right next to each other, so you can bring it up by rolling your thumb from one to the other, which is at most 50% more inconvenient than just pressing the first button.
The command key does need to register first, though. How easy that is depends a bit on what style of keyboard you have.
With the old full-travel desktop keyboards, it's a bit awkward.
With the pre-"chiclet" laptop keyboards, where there is no significant gap between the keys and the edges are simply beveled, it's pretty easy and smooth.
I don't have a chiclet keyboard handy at the moment to test, but it probably isn't any more awkward than the full-travel keyboards.
To reply to the point raised by several people: yes I know about CMD-Space (this is what I meant by Spotlight but I guess that can mean more).
Somehow it just doesn't seem as good. Maybe it's the single keypress. Maybe it's that Start also has pinned and commonly used items on it (CMD-Space is just a box you can type in so really it's half the functionality of Win7's Start button).
That's where I'd put my money, too. Several weeks ago I was sitting at my laptop and I wanted to scroll down on a web page I was reading. I instinctively wrapped my hand around the bottom-right corner of my monitor and brushed the screen with my thumb.
A week later I was in an Apple store and a young kid walked up to an iMac and said "ooh, I bet this is a touch screen". It wasn't, of course, but it made me wonder if Apple hasn't been quietly training a generation of people to expect to be able to touch the screen. Maybe that's what the much-maligned glass displays has been about: user preparation for the inevitable day.
I tend to think that Retina on larger screens (iPad sized and larger) still a ways off for cost and gpu/battery reasons.
I mean, doesn't a 30" monitor require dual-link capability to feed that resolution? A retina display on even an iPad sized screen would approach that type of resolution.
The dual-link thing only applies to DVI, which Apple doesn't even use externally anymore, let alone internally. Also, the success of AMD's Eyefinity shows that modern GPUs can handle very high resolutions effortlessly for anything other than complicated 3d rendering. 2D and simple 3D (of the sort that might be used by a compositing window manager) are too easy.
The only limitation is that producing very high DPI displays in large sizes with acceptable defect rates is extremely expensive.
A 15" laptop would need something like 4000×3000 pixel to equal the current iPhone’s resolution. (A 30" screen would accordingly need something like 8000×6000 pixel.)
I don’t even know whether that‘s currently feasible below the price of a small car or something like that. I wouldn’t expect 300+ ppi (“Retina“) screens anytime soon but resolution independence sure would be nice. (And later maybe some 200+ dpi screens.)
A Mac "Retina" display wouldn't necessarily have to be 326 dpi like the iPhone 4's: computer displays are usually quite a bit further away from the user than iPhone displays.
I wish they would get rid of the absurd "mount a disk before you run an installer" paradigm. Installing an app downloaded from the web, Windows vs Mac:
Windows:
1 - Click a download link
2 - Yes, I'm sure
3 - Run
4 - Next, Next, Next, Finish
5 - Delete the setup file
OS X:
1 - Click a download link
2 - Yes, I'm sure
3 - Open a finder window
4 - Open .dmg file, and it mounts
5 - Open the mounted image
6 - Run the installer (or drag it somewhere, depending on the app)
I agree the description is a little unfair, but the point is sound.
This is how it should work
1. User clicks download link
2. Browser downloads a standard cross-platform .zip file, decompresses in a temporary directory
3. Browser notifies OS.
4. OS looks in Info.plist for instructions on where to install, possibly mpkg for more complex installers
5. OS prompts user for permission to install and possibly run the app, or passes user over to Installer.app for complex installs
6. Temporary files are deleted
From the user's perspective it would work like this
A lot of application DMGs feeling they have to give the Finder window a custom background image directing you to drag the .app to /Applications/ does indicte the paradigm is a bit shabby.
Disk images are one of the most confusing things about day-to-day Mac OS X use, no question. The next time you're at a relatively non-technical friend's house, apple-click on Firefox or Chrome in their Dock. Dollars to donuts, it resides on a disk image.
Apple got rid of disks, why not get rid of the images too?
One could go in almost any direction, like bible codes[1], by trying to infer things from the nature of the particular cat in question.
For example, "Lions are unusually social compared to other cats,"[2] which means that MacOS X Lion could have social networking as its main feature. Just pick what you hope for the most and find a way to tie it to the critter.
10.0 ("Cheetah") was at least supposed to be cheetahy--i.e. really fast--at least compared to the Public Beta. That was the main focus of improvement there.
I'm not really looking forward to that. With each upgrade comes that little bit of agony about the unix underpinnings: what are they going to do to the default python install? Ruby? ...
Add MacPorts (/opt/local/bin) to the front of your path and you have fully updated versions of all of your favorite UNIX goodies such as Emacs, Clang and Python.
In 2009, Apple released Snow Leopard, whose emphasis on optimization and subtle improvement seemed to signal the same kind of state for the operating system as the addiiton of a camera signaled for the Nano.
I think it's safe to say that what comes next will be fairly big.
I have a suspicion that among other things, 10.7 will introduce some new conventions to the basic problems of window management (though I might be a little biased). Apple has been at the forefront of dealing with it ever since Exposé debuted with 10.3, but in the years since the iPhone, iOS has thrown this problem into sharp relief. Part of me hopes they might even announce something that bears a resemblance to 10/GUI's linear window management -- I've unfortunately proven I can't really do much with the concept myself so far.