Re: What I want
Originally posted by suzerain
My own personal opinion is that people are now sufficiently acquainted with computers that we can stop all these stupid real-world metaphors. A "file" in computer parlance is not really a file; it is simply an address where data resides.
Replace "people are" with "I am" and you might be correct. In my experience, however, the vast majority of computer users are
not beyond real-world metaphors; they cling to them voraciously and are completely lost when those metaphors are violated.
I think we need to take the BeOS system-wide database thing even further, and instead forget where things are grouped, and just have all groups be dynamic searches.
Why completely discard heirarchal organization in favor of endless searches? Seems silly to me. Heirarchal organization is just so easy to implement, and so easy to ignore if you don't want to bother with it (just put everything in your "Documents" folder ...) that it's just plain silly to discard it altogether.
You know, say you create a Microsoft Word document which is a letter to a potential employer. In my Finder, you'd save that to groups you've assigned (basically just pieces of metadata), rather than choosing a "fixed location" (like now).
As an example, if you didn't want a core heirarchy, the file "name" you chose would just be in the Documents folder. You only have to use folders if you want to. I mean, DOS 1.0 didn't even have folders!
So, you give the file a name, and the save dialog box allows you to just pick as many groups that you have defined as you want, like "Work", "Cover Letters", "Company XXY", and the document is "saved" in all of them at once.
An excellent idea (meta-data on files), but, again, this does not mean that folders must or should be forgotten.
This whole "spatial Finder" thing was useful back when computers were a new paradigm, because it related to the real world, but we've now reached the point where continuing these metaphors is an exercise in futility, and is really holding computers back, rather than helping them become more useful.
Just my unsolicited opinion on all of this.
Well, I don't see how spatial concepts are "holding computers back". Layering additional functionality over the top of a solid spatially-oriented core makes far more sense to me.
Take a ride in a modern jet and you'll find the same basic controls you'd see in a prop-driven plane 75 years ago. The controls back up to electronic sensors and such, and pulling back on the stick doesn't connect directly to the flight control surfaces via cables and pulleys, but still the controls are the same. Why? Because, in a desparate pinch, many aircraft will backup to manual systems, because every school child knows how to operate a joystick, because the original controls were designed in a human-oriented manner.
The idea here is simple: when all else fails, the fundamentals still apply. If you saved a file and can't seem to find it via your searches, you can still go in and manually scour the folders for it. While it's nice to just say "Work reports, most recent" and get the most recent report you worked on, sometimes things get mis-labeled and mis-filed, and sometimes indices get out of whack (programs continue to have bugs).
IMHO, a lot of the "spatial Finder" stuff is overblown, but we have lost the clear distinction between our spatially-oriented foundation and the browsing-oriented abstractions. The result is a measurable loss of usability. By "measurable" I mean via formal usability studies: have a dozen subjects do common tasks and measure how quickly and easily they end up doing them.
The simple fact is that most people out there are not database programmers, and don't think of their files as "an address where data resides". They think of files as physical, tangible objects somewhere behind their monitor. No matter how "spatial" or "abstract" their interaction with these files is (ie, the Finder), they continue to think that there is somewhere some physical object that holds their data. Now and in the future, most people will use the computer as a tool, and should not have to adjust their way of thinking 100% to be in line with that of their tools.
In most fields the cutting edge is abstract as can be, but concepts are quickly reworked so that they can be understood in common terms. Scientists and engineers strive to come up with concrete metaphors for their concepts, because that is the fundamental key to progress. I find it odd that there is a strong tendency in computer circles to do the opposite.
