Tuesday, July 10, 2012

The 3D desktop and the dangers of unchallenged ideas

A follow-up of sorts to my last post on my use of Gnome3 and the struggles overcoming it's user interface problems. After a few updates, Gnome3 "safe mode" has become usable. Workspace switching now works but the top and bottom panel stiil can't be modified, not easily anyway. But it is usable and familiar. In fact, I use it for about 90% of the time. Some things don't work all the time, though, like sftp integration. But that could be PCManFm's problem, the file manager. But since it's the default choice, Gnome therefore shoulders some blame.
This whole episode reminded me of something that happened to me in the late 90s. I was asked about what would be the next step be in terms of desktop interfaces. I had postulated that given the proliferation of graphic cards with 3D capabilities, the next step in desktop interfaces would take advantage of that. It was logical because Microsoft Windows tend to drive the hardware requirements up. By requiring that PCs running Windows have 3D capabilities, PC makers would then make it a standard. Which is in their interest because this would drive new PC sales. It happened before with CD-ROM drives so why not 3D graphic cards?
The design was based on the concept that the desktop would exist in a 3D world. Application windows would "float" in the air above a horizon. By moving within the 3D world, you would move and zoom in and out of the windows. For example, if you are working on a spreadsheet and you want to see a document  to verify some information, you would "move backward", making the spreadsheet window and it's contents smaller but still visible. Now you can open a document at the "normal" size. You would cut and paste in between them as you would normally do because you can still see the spreadsheet and use it even though it was smaller.
A horizon on the bottom of the screen would hint the idea of land. It would animate according to your movements within the world, giving the sense of movement when there are no windows opened. I think that it could also have features like hills and bodies of water to introduce the idea of unique areas. 
This would be useful when you create a workspace (in the semi-literal sense) by grouping windows together within an area of the 3D environment. By moving between the groups, you would be moving between workspaces. The different features of the horizon would help you remember where the groups are. To make movement easier, double-clicking on the horizon would bring up a "signpost" that would point to where those groups or windows are. You can move manually or click on the signs on the post to jump to the area.
Since the windows would exist in a 3D space, you would be able to scroll the "world" left and right.  Scrolling slightly to the side would eventually make the visible applications start to move off the screen. This is useful if you need to see only part of an application window while opening and working on another. Or you could stretch the window further to see more of the application. This would be useful for spreadsheets and graphical applications like painting or CAD application where you could want to open the windows into a space larger than the actual screen size.. 
So what's the point I'm trying to make here?

I thought of all that in the space of about 15 minutes. Once I presented it to other people, there were many problems that were discovered. While the consensus in the room, wrongly, agreed that 3D hardware would be standard on all PCs, not many agreed that the desktop should go 3D. First was the relatively low resolution of 3D display sat the time. It was accepted that when switching from 2D to 3D, the resolution would drop significantly. This was because the GPUs at the time were not powerful enough to render 3D at acceptable desktop resolution. In fact, what was more desirable was the opposite, an even higher resolution. Second was that although the concept was cool, the application was limited. The people who would benefit from the virtual desktop resolution, the graphic professionals and accountants, preferred a second screen as an answer. Third was that there would be problems quickly navigating within the 3D world with a mouse. Even most first person shooters today use a combination of keyboard and mouse or a special mouse with about a thousand extra buttons.
My point is that although my idea sounds smart and cool inside my head, it took me a short time to realize that there were many problems with it once I shared it with others. The more people I shared it with, the more varied the responses were. I didn't agree with all of them but I did try to take their position to see if I thought they were valid. Sometimes by doing this, I could see something different than they did. 
It was clear that the Gnome UI designers were only listening to themselves when designing the UI. Even though they were sharing it among themselves, there could have been a echo chamber effect, of ideas being amplified without actually being scrutinized. Their position of exclusion made it even worse. This is demonstrated in their attitude that "if it doesn't work for you then it's not for you". Not everyone is using a Mac Book. Worse of all, it was a continuation of an existing product loved and used by many. Perhaps the "we know better" attitude of Apple seems to be rubbing off on the people who are using them.
I understand their effort to be original and demonstrating leadership. But to be a leader means being challenged. Invention will always be great to it's inventor. Only when the invention is used does it show it's true value. It's great to think of new stuff and try out new things. This is what open source is about. But open source is also about community. Not thinking of your community mean you're straying away from that concept.The truth is, everybody gets it wrong sometimes. It's time to swallow some pride and really lead. True leaders learn from their mistakes.


No comments:

Post a Comment

Recently Popular