Seapegasus Blog

Wednesday April 05, 2006

Don't Write on the Backside of Paper

Didn't we learn that at school? "Don't continue writing on the backside of the sheet, it's tidier to take a new one!" *squinting over to the stack of scribbled notes on my desk* Hmmm. I mean, come on, at school I also used some made-up sign language to 'discuss' results during exams, and do I still do that today? No. Times change. As soon as we come of age, we write on the backsides of paper as much as we want, and even after midnight, hah! Eat this, school. :-p

Of course nowadays, we don't write on paper anymore: Memos arrive via email and we print them (the so-called paperless office.) How come school never taught us useful stuff like how to save big bucks by printing duplex and psnup -9? (Are schools sponsored by the paper industry?) And that's only paper: With the increase of possible screen resolutions, we can even use our screen's real estate better each year. I mean, come on, there were times when having a one-dimensional commandline was the Next Big Thing! I say "one-dimensional commandline" because IIRC, when the third dimension is this all aroundish kind of thing that you see when you look out of the window, the second dimension is a plane, and the first a line (and the zeroth dimension a dot?).

What a command line interface can do in one dimension is a very good start. You could enter commands to create and edit files, browse, move and copy them, just as we do today. When memory allowed longer lines, you could attach a bigger screen and let the line wrap. Information could now be spread over several rows and columns on the screen. That way, the commandline mutated into something new -- a 2D interface. Quite a clever use of just a simple line of text, considering that the alternative was... punchcards or something.

Hold on to your hats: Text on its way to the 2D interface.

The first demonstration of what would be possible if we switched from commandline to an innovative 2D user interface with a mouse happened as early as 1968 (!). As soon as the hardware for these textual 2D interfaces was available (and later even for visual 2D interfaces like Apple Finder and Microsoft Explorer), people came up with more and more interesting ideas of how to use 2D space: Menus drop down and make room when you didn't need them anymore; drawers with once-a-day favorites pop out, and sticky taskbars with once-an-hour favorites follow you everywhere you go; stack and tile windows with lists of files in them, and the interface even lets you drag and drop items from one to the other -- Awesome!

This kind of visual 2D interfaces has already been awesome for over 15 years straight. =-) So will the next logical step be 3D interfaces? The current hardware can take it, and the 3D software for interfaces seems to be on its way. But still there are no 3D applications in our everyday live, and most people don't see where they should come in.

To get a picture of what 3D applications could look like, I looked at this demo of Project Looking Glass and some screen shots of possible applicatons. There is also this silly divx/avi where somebody tried to implement a Sun version of a Minority Report-like interface. :-D Allright, some good ideas there: Writing notes on the backside of a webpage is cool; a rotating CD browser is not too bad; UML diagram designers may look forward to viewing diagrams in 3D (only to realize that they'd need 4D to understand them); and it may be funny for a while to rotate a movie player and watch the movie from behind for a change. But is that all? Where are the innovative killer apps? :(

Well, we will have to write them. But what is there I can do in 3D that I can't do as well in 2D? On the other hand, what tasks are there in a 2D visual explorer that we can't as well accomplish on the commandline? The tasks may be the same, but maybe we find a cooler way of doing them. We don't know yet when or whether 3D will catch on (I demand this sentence be quoted by historians in 10 years!), I think the current plan is that companies provide the hardware and software (Looking Glass for Linux, Looking Glass for Windows) and then see whether someone installs it and ... happens to come up with a useful application.

Charlie Hunt for instance already started working on a Looking Glass module for the NetBeans IDE! He plans to set up a project and announce it in his blog so other can participate. The idea behind his LG project? "Any Looking Glass application and visualization is possible in the NetBeans IDE and the NetBeans Platform. Think of the possibilities! Imagine being able to debug an application in three dimensions, being able to view multiple threads and their relationship with each other as a program is executing with the current executing thread rotating to the foreground." And what I personally am looking forward to doing most -- we may finally be able to turn the Source Code Editor around and write on its backside! :-)