International Mac Podcast

Blog RSS Subscribe by email

Keaton On: Computing, Circa 2015-2020

Posted on Wednesday, March 31, 2010 by | Add a comment

Computers themselves have evolved drastically since the turn of the millennia, back in 2000 a single gigahertz was a lot, and a gigabyte of RAM was considered overkill. The Playstation 2 and XBox were still vaporware, Cellphones just made calls and PDA’s just remembered data, and neither WiFi nor Cellular Data Networks had really caught on.

This decade brought HUGE advances in the hardware side of computing, and yet the way we interact with our computers is nearly identical to how it was back then. In fact, the way we use computers hasn’t significantly changed since 1984, back when color screens were a novelty and the internet was only for the geekiest of geeks. Modern UI’s are often still based around the limitations and capabilities of 26 year old computers, and though we don’t really think about it, are now just a huge inconvenience in almost every situation. We’ve already seen some very successful attempts at getting rid of these vestigial interfaces (chances are you’ve got one in your pocket right now), and I think that by the end of this decade the software that we use now will be just as obsolete as the hardware of 10 years ago is today.

Kill the mice.

Look down at your hand, The majority of you will see 10 fingers (if instead you see bloody stumps, I would call a doctor or maybe the police). These are the 10 controllers that you use to interact with most things in the world, although in fact your whole body has 639 different controllers (muscles) that affect 230 different joints, giving you hundreds of control points and degrees of freedom. Now look at your screen, there is only one controller there, there is only one two-dimensional non-rotatable point that you can influence on your whole screen, and there’s only a handful of operations that can be performed on that point (left click, double click, right click, and scroll). And you wonder why your computer doesn’t feel natural to use? The concept of a cursor severely limits how quickly simple tasks can be accomplished on computers.

So why do we still use mice, touch panels have been around for years. The problem is ergonomics; unlike touchscreens, mice and trackpads allow you to rest your hand and arm while using them, and don’t cause your hand to obstruct your view of the screen. These are not problems for portable devices like the iPhone or iPad because you can hold them within the comfortable range of your arm, and the UI is designed with big buttons that don’t require hand-eye coordination to hit, however laptop and desktop computers are a different story. People want big screens, and they don’t necessarily want them right up to their faces, so we need physical separation of the user and the computer, but we still need it to feel natural. It’s a tough problem, the best solution to this I’ve seen is called 10/GUI, which uses a giant pressure sensitive multitouch trackpad positioned next to the keyboard. The video on their website is very well done and reiterates a lot of the points I’ve made, just in case you don’t believe me or have understood nothing I’ve said in the last 2 paragraphs.

That’s a solution to the ergonomic hardware problem, but one of the main reasons that we still use a single cursor is that every piece of desktop software written in the last 26 years is designed for a single cursor. We’ve already seen multitouch technology working it’s way into trackpads and even mice, but those are gesture based. It will take a huge effort and probably entirely new software and OS’s in order to finally see true multitouch work it’s way into laptops and desktops. Can it be done? Yes, in fact I would say that the iPad is the first foray into this, because I would actually count the iPad as a kind of laptop, and it definitely is capable of running fully fledged desktop-style software. We might even see the iPhone OS, or some future spinoff of it, replace Mac OS X in the long run.

Move the file system to the trash.

Hierarchical file systems are great if you’re a neat freak, but for the other 99.9% of us they just force us to dig through tons of junk to find what we’re really looking for. How many things are in your documents folder? For me it’s 319 different items, many of which are folders containing hundreds and hundreds more items. However I’ve probably only edited 20 or 30 of those items within the last month, and there’s only about one or two that I’m actively working on right now. I think the file system should be pretty much entirely replaced with smart folders where the user can assign rules for what goes where. Mac OS X already has a smart folder system, but it needs to be expanded, made more powerful, and it needs to more or less take over the OS. For instance, you could have a smart folder full of all the documents that are currently open in any application, and that folder could be the default thing that comes up in file open dialogues, that way you could pop an image from iPhoto to Photoshop to iMovie in a snap. Also, most people’s folders are generally organized by project, the OS could automatically clump related files instead of relying on the user to do it. The OS knows enough to organize a users data for them, but as of now it still generally relies on the user, which is a bad idea when your user is a file-slob like myself.

File systems also need to be more graphical. This is more for Windows, seeing as OS X already does a fantastic job with thumbnails and quicklook (which is not to say that it can’t be improved upon, but at least it’s not a new idea to mac users). Files shouldn’t be treated as lumps of ones and zeroes, they are content, they’re tangible stuff. Windows represents most types of file by the format that it’s in, which is not what people really care about. Also, all OS’s need to take the file system beyond grids into more interactive formats (Again, OS X is ahead of the curve with Cover Flow, but again, it could be taken further.)

Lock the Windows.

No, not the OS, that’ll definitely still be around in ten years, the concept of overlapping windows. It’s great for some things, and it’s probably the best way of multitasking on computers so far, but it’s tedious. All of the major OS’s have come up with their own ways of managing windows: Exposé and Spaces on OS X, Flip 3D and Snap-to-edge on Windows 7, and Compiz Fusion on Linux. All of these are novel approaches, but they all sort of dodging around the root problem, which is that people have so many windows open that windows get lost behind piles of other windows. If you’ve got two or three windows open the system works perfectly, but just 3 hours after re-starting I have 21 windows open, and I can sometimes get up to as many as 50. So how do we make it better? One idea I had is universal tabs, where instead of just having some apps use tabs, every app and every window would show up in one master tab bar at the top of the screen, and documents and files could simply be dragged between tabs. The problem is that then every app would have to be exactly the same size, and that would be very awkward for a lot of smaller apps and utilities, and also it would be more difficult to multitask. Another solution is the 10/GUI system where apps are arranged in a line, and you can simply scroll through the line. It seems like things like that would limit what you could do, but really it’s just a way of having the computer organize your work environment for you, which would be amazing for people like me.

Sync or sink.

It used to be that people only had one or two computers in their house, but now we’ve got iPhones, laptops, desktops, tablets, kitchen computers, and even computerized TV’s. In our (admittedly ridiculously geeky) house, we’ve got somewhere around 30 full-on working computers among 4 people. Each one of these computers is more or less an island, relying on the user to transport data to and from it. Managing all of your data between this many computers can be an incredibly tiring process, and though many solutions such as my beloved dropbox exist, none of them are cheap enough or universal enough just yet. There needs to be an easy way to sync absolutely everything between computers, ideally free and built into the OS (using one computer as a host, instead of relying on expensive hosting services), and the near-ubiquitous internet that we’ve got now is perfectly capable of that. All files, all data, and all settings should be available on all computers, no matter how big or small, that way they all become nodes of your personal life-management network rather than islands for you to jump between.

All shapes and sizes.

Finally, we’ve become used to computers showing up in all forms, but that’s a surprisingly recent idea. Computers used to be boxes plugged into screens and keyboards, and many still are, but that’s becoming a relatively quaint idea. We used to think of computers as destinations, somewhere to go when you needed to look up something on the internet, but increasingly the lines are being blurred between cyberspace and physicalspace, and now we expect computers to follow us around everywhere. This new idea has lead to laptops, mobile devices, and now tablets, and it won’t stop there. Soon we’ll see skinput, wristband computers that project data onto our bodies to save the physical space occupied by a screen. Or table-top computers like the Microsoft Surface, that allows us to collaboratively compute with others. Or touch-projection computers that can essentially project a touch screen onto any surface. In fact, soon computers will become more and more a part of everything, until you don’t specifically buy a computer machine, you buy a table, or a wall, or a house, and it’s equipped with a full interactive experience that follows you everywhere. Computers won’t be recognizable as computers for much longer, the ideas we have about what a computer looks like are becoming increasingly varied, until a computer can look like anything.





Comments are closed.