“Maybbe apple need a time machine so they can tell them selfs not to b so lame”
I think this wonderful Engadget comment captures the zeitgeist of the morning of June 11, where Steve Jobs gave the boringest WWDC address I have ever seen. To recap:
Doesn’t quite compare to the Intel announcement, the theatrical death of OS 9, or even Leopard’s semi-underwhelming introduction last year, does it? Unless you’re a hopeless Apple apologist. Even still, when I bitch in chatrooms and message boards about how OS X sucks and we should expect better, people still say “WELL WHAT DO YOU REALLY WANT THEM TO CHANGE ANYWAY??” As a result, I have come up with a bunch of ideas that can be seen as a mixture of what Leopard should have been and what new OS Xes should bring. Since Leopard’s ship has sailed, I will bill it as my featureset for 10.6 Liger.
You may notice I hold a bit of contempt for other lists like these. I have no problem with people who come up with great ideas to refine parts of OS X’s UI, but I think often people rely on. My first few features are supposed to be a lot more “revolutionary”, and as a side note would make an awesome demo for attracting customers in a way that “imagine if Safari had extensions!!” sort of ideas never will. I hope they’re interesting, though of course many of my ideas are influenced by others.
Summary — Scrap the Finder, restart, and do something we haven’t seen before.
I’m gonna start with something I can’t offer a solution to. The Finder has been pushed as far as it can reasonably go, without major changes to the paradigm behind it. People have written interesting things about how the Finder fucked up spatial navigation and seems generally lost and confused. It’s functional as a file manager, for sure, but surely we’ve learnt something in the last 23 years since the Mac first came out.
Some people respond to the Finder’s problems with this kind of madness. You can see above the results of a technical approach to fixing the Finder.. “Oh, add terminal support! Make sure you can customise the colours used in alternating rows! Add support for multiple sidebar sets!” This suits people, but they’re typically the sort of greasy nerds who don’t understand good software, or who have really specific needs a common man’s Finder could never address. So you can see what I don’t want Apple to do.
What should the new Finder do? Define a new paradigm. Use some elements of icons and folders, and still work in the world of hierarchial file systems, but allow people to manage things visually, like Stacks but done properly. Use 3D. Use (Core) animation. Use metadata.
Imagine if the process of locating documents for modification\moving\viewing involved rifling through this screen:
Where you could see the documents you’ve been working with today, or yesterday, or in the last month. With small visual cues like individual thumbnail size to show which you spent more time on, or round outlines to show what you’re already working on. Or if you didn’t like that, click “Projects” to view stacks of files put together that OS X has determined are related to each other (through content, position in the file structure, and usage patterns) or simply click “relevance” to have it give you a selection of all the files that seem relevant to what you’ve been doing in the last few minutes (“Oh, you put in a USB drive? You seem to copy your recent photos to that a lot, let me bring those to the top.”)
Okay, so I do have some ideas for the Finder. But it’s this sort of thinking that is sorely missing from the Finder, and (disturbingly) from the mockups I see other people make. Finder cannot be fixed by allowing you to split panes, adding SFTP support and arbitrarily changing window dressing. It can be fixed by using Apple’s clever people to push the boundaries of what we expect from our computers as we manage documents.
Summary — Give Spotlight a natural language parser and make it a lot smarter\fuzzier under the hood.
So, what enables cool UI ideas like that? It came in 10.4 - Spotlight. Not just a search technology, it introduced the concept of a metadata store for each file in OS X, and I think there’s a lot of room for improvement here.
Spotlight is great technology. Apple seem to be working well on that. I encourage them to make it faster, and to remove problems with the 10.4 implementation of it like the lack of support for single files containing multiple search records. But beyond that, what Spotlight needs is to be more usable. It needs to gather more and more metadata - whether that be by logging what’s used with what, how long it’s open, etc, and it needs heavy reworking with the UI.
Fixing the terrible lagging problems and the “mystery meat” search window would be a start, of course, but I think a great Spotlight will go hand in hand with Finder. Search belongs in Finder (duh, just read the name) but it deserves to be done a lot better.
I’m not going to provide a visual mockup here, as it’s so simple. I want you to imagine the primary usage of Spotlight to be through that menu item. No new buttons or a sidebar or anything, just that search input - but a much more powerful one.
Right now you can type “linguistics nominative” and you get your assignment about nominative\accusative systems, but that’s all. If I type “linguistics uni” it will find nothing, nor “language work”. Why not? Because it’s braindead! Why shouldn’t OS X know that my assignment’s in a folder called “Uni Work”, or even that while I was working on that assignment the other night I used the slang “uni” when talking to my friend over IM? Why can’t it associate the word “linguistics” with “language” and come up with a relevant result even when nothing matches exactly? Stupid, stupid.
Also, I don’t know if you noticed, but computers are getting really great at parsing images and speech now. Why aren’t all my audio files transcribed to a private text file? I’d love to type “podcast microsoft” and find all my podcasts where someone said “microsoft”. If I downloaded a huge wallpaper pack and decided I want something new, why isn’t it smart enough to figure out what “pastel colour wallpaper” means? (For those of you not following, that translates to filtering out all wallpaper-size images in colour which have pastel hues.) Why not “long noisy camera movie from last week” for that long noisy movie you took on your digital camera last week?
This is where Spotlight needs to go. I think it’d be possible to at least get a start on this with 10.6. Just give it a vocabulary, make sure it gathers loads of info, and code some sophisticated processing algorithms that constantly work behind the scenes to enhance your ability to find the stuff you want. It’s the sort of thing that could put Apple way ahead of anybody else if they tried.
Summary — Use the sensory mechanisms of modern Macs to adapt to different contexts of usage.
I’m not sure if I’m the only one who noticed, but Macs are pretty good at noticing what’s going on around them now. By my count, a MacBook Pro can:
That’s everything except smell and taste, with a bit of a double-up on the “sight” and “touch” ones there. Apple should do something with this.
A rudimentary idea, first. Let’s say you want to switch network settings between the office and home. What’s different about the office and the home? The MacBook knows that home has an airport network with a unique name, so it’s not hard to solve this problem with that sort of intelligence.
But let’s say we move it to the next level. Perhaps the MacBook notices that the outside world is brighter, noisier and involves a lot of movement, especially when it’s in its briefcase. Home has a warmer, dimmer light, is quiet (besides the music it can hear its friend the AppleTV making), and it’s usually sitting still on a desk, or getting warmer than usual on a bed. When it’s in the city, it sees tens of access points, and has a lot of bluetooth phones coming in and out of range.
This sort of information I envision in some sort of “Location Manager” in the system preferences, where you have some predefined settings like “Home” and “Work” and “School” and “Out and about” to adjust settings in - perhaps energy saving options, automatic muting, applications to transparently prefetch for faster loading, metadata to save. Perhaps we could search for “documents saved yesterday in the city” with this technology? Have iChat tell people where you are, or have iCal record where you have been? Adjust fan speed depending on the amount of ambient noise that’s available to cover it?
This idea has more to develop about it, but I think you could do exciting things if you played more with it. Again, more intelligent software in the background, constantly monitoring and recording activities, and doing its best to associate things with each other.
Summary — Implement resolution independence and push it. Down our throats. Relentlessly.
Okay, they’re already doing this. But they’re doing it wrong. How should resolution independence be addressed?
Well, the day Liger is released, Apple say “okay guys, all of our Macs are running 160dpi displays now.” and update everything - their studio displays, their MacBooks, the iMac. And so a new era is born - where you measure a computer not only by its CPU speed, graphics and screen size, but by its dpi rating. At the point Apple moves beyond the 100dpi standard, Macs can once again stand out as super unique. Linux is a long way from getting its act together as far as resolution independence goes (hell, it’s still a long way away from getting it’s GUI together) and Windows won’t have it done for another two versions. Apple, as the only hardware+software PC company around nowadays, is in a prime position to not only become the supplier of computers that subjectively “just work” and look pretty, but the sole supplier of computers that display text and photos with amazing sharpness above all other computers.
Sure, Apple has to wait for LCD manufacturers to make more high-dpi displays, but there’s gonna be a 2-3 year gap where all the Linux and Windows people have a mix of low dpi screens and high dpi screens with tiny text, and Apple could pounce on this by quickly adopting a product line of all high-dpi screens with the only OS around that can supply them with big, crisp hybrid bitmap and vector images that look amazing and are readable by people older than 16. This is a big missed opportunity, and I hope with the next OS X Apple take this chance.
There are a few short ideas I have that are not necessarily minor, but harder to explicate into several paragraphs.
Oh, yeah, and when they introduce this new OS X, they should stop making fun of Windows so much. Has anyone noticed that the amount of “Redmond, start your photocopiers”-esque banners at WWDC seem to have a proportionate relationship to the amount of ideas Apple will borrow from Windows that year? I find nothing wrong with teasing MS, but it seems that it makes Apple unhealthily self-satisfied, and in a world where Vista still has mostly everything cool about Leopard, they can’t afford to be this way for too long.