August 16, 2002
Simon Gronlund is earning his Master of Science in Computer Science at the Royal Institute of Technology, Sweden, as an adult student. He also teaches Java and computer-related courses at the college. When he isn't tampering with his Warp 4 PC, he spends his spare time with his two boys and his wife.
If you have a comment about the content of this article, please feel free to vent in the OS/2 eZine discussion forums.
There is also a Printer Friendly version of this page.
What is natural to you?
Why is that we arrange our files as a
tree? In OS/2, having more partitions gives you more trees,
while the 'nix world mounts the partitions in a bigger
tree. Perhaps Mac users don't think they use trees but
they are there, although you might plow through a bunch of
folders first. The trees seem to be here to stay.
The načve answer is that an unordered bunch of files soon becomes unmanageable, you cannot find the needle in the haystack. It will also soon become impossible to find unique file names.
Organizing files in families seems more natural since we work that way from childhood. Advertising in one pile, bills in another pile and personal letters in a third. Your e-mail client is pre-organized the same way: drafts, inbox, outbox, sent mail and trash, and you may add folders of your choice. Your office is furnished with shelves filled with clearly labelled binders, each containing its family of loose-leafs.
Since computers mostly are invented and developed by scientists, the loose-leaf binder philosophy has had a great impact on how to organize files. Naturally the operating system has been separated from user software and personal files. And naturally applications can be kept separate from utilities and tools, as my letters aren't mixed with my wife's.
But is this way of organizing the computer natural to computer illiterate people? Due to the common errors people make and how they seem to be lost whenever they have to find out the location of a lost document, I tend to say: No this is not the only and the most natural way!
Humans tend to organize stuff in piles, binders and folders. But we never put a folder within another folder. You could find the largest (physical) file cabinet, yet it still does not support nested folders. Neither does your bookshelf or your desk.
Due to limited space and our human limits we cannot manage an unlimited number of folders/binders/piles. Due to limited space, our computer desktops do not afford unlimited numbers of folders.
Okay, but use less nesting then. That is a start, but does it solve the problem in the long run? People using computers more and more over time will soon have an unmanageable magnitude of files. What about big enterprises? Can they benefit from a lower tree height? No.
I do not propose a new and revolutionary design with this. There is a lot of work going on, one project weirder than the next. But at least these researchers are looking for a better way of managing information. Let us see to it that OS/2 (and eCS) will not be the last one on the wagon when computers move forward.
In the long run we will see a new standard evolve. Maybe we will see some efforts from various locations. Can OS/2 be one of these? Can we develop something useful today?
What about a file system more like a database? It would need to be searchable with an interface that people are already familiar with. A Google-looking Open-file dialog perhaps. Save is nothing more than a Save button, it wouldn't even need a dialog except if you explicitly wanted to add key words or anything else to the document description.
A search for "insurance", "accident" and "demolished" will maybe provide three hits. But the document date or a short excerpt will reveal which one you where looking for so you can open it.
You never now where the document resides, but it will be somewhere. You do not need to know how the software piece actually works. The essential part is that you work with another kind of interface. The software piece might even build its own tree depending on which software produced the document, or what the document is about, but that is not important.
A Google-looking interface might not be the best choice, I don't know the correct answer. But I sure do know that my wife, a clever girl, but also an artist rather than a technician, finds computers awkward to work with. Don't make the mistake of thinking she doesn't like computers, she uses my computer and she loves to play with Photoshop and such software. But there are so many things I take as natural that she does not. And she is right, they are not. I can use them because I am maladjusted to the nerds' world.
This column is not written because I need another and better interface myself. I am rather content with today's graphical user interfaces (GUI) and have learned that the more minimalistic they are, the faster they work. And I work faster along with them.
This column is written with the following mantra: if OS/2 ever will find its market share it needs to afford at least as good a GUI as its competitors. Or better. The average user does not care about stability issues, nor other technical features you might face them with. Object oriented workplace shell does not mean anything to them, as it does not to some of this columns readers.
Neither do tomorrows decision makers bother much with technical data, as today most software vendors offer rather good quality. They are concerned with how people can manage their tasks. And correctly they leave it to the technicians and engineers to solve the technical problems. If anyone can afford something that is clearly better than average, that might be their choice.
Hopefully rumors of good operating systems will trickle into the ears of decicion makers. And rumors, where do they originate from? From stuff that common people like. But nerds like you and me, we are not common people, OS/2 needs to hit the broader mass with its need.
Anyone have something up your sleeve? Show it then.
Anyone have ideas? Write an article. Let the debate live