h a l f b a k e r yFree set of rusty screwdrivers if you order now.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
I can't think of any kind of data that can't be represented as a tree. Even modern relational database tables are symmetrical trees.
A lot of menus on modern applications are arranged in trees. The windows registry is a tree. My proposal is for an interface that represents all of an application's
data as a tree. Forget about what window or tab or button that you have to get to to view or modify data. You now have an abolute path to all data in an application.
If you wanted to get real crazy, you could show the entire computer system as a tree.
(??) An example tree
http://www.clustan....ClusterDataTree.gif
This is just one example of a data tree [lawpoop, Apr 16 2005]
Windows registry screenshot
http://images.googl...safe%3Doff%26sa%3DN
Imagine one of these for all programs. [lawpoop, Apr 16 2005]
Concept Maps
http://users.edte.u...lanzing/cm_home.htm
"Concepts and sometimes links are labeled. Links can be non-, uni- or bi-directional. Concepts and links may be categorised, they can be simply associative, specified or divided in categories such as causal or temporal relations." [reensure, Apr 17 2005]
CMap Tools
http://cmap.ihmc.us/
To answer, "How are concept maps browsable?" [reensure, Apr 17 2005]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Destination URL.
E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)
|
| |
Although I'm not totally positive I understand what you're getting at, I think I get it and it sounds pretty baked to me. I can't name anything specific off the top of my head, but LOTS of applications use a tree-based interface. Some prime examples of this are Linux networking configurators, and a program I once used years ago to create one of those press-4-to-speak-to-nobody type automated phone systems. |
|
| |
You have tuned into what I'm getting at, but this is supposed to be *comprehensive*. Meaning, there shouldn't be anything left out. |
|
| |
Current applications have a mis-mash of menus, trees, windows, tabs, buttons, fields, dialogue boxes, wizards, etc. You can go nuts trying to find or remember where a particular setting is. Sometimes settings are left out! |
|
| |
With a comprehensive tree, all the data is in there. |
|
| |
I don't get it. Illustrate. |
|
| |
Have you browsed the windows registry with 'regedit'? That's basically what I'm talking about. All data in any program can be arranged hierarchically, and that arrangement is called a tree. |
|
| |
OK. Utterly baked. For example, in regedit. |
|
| |
This still sounds like an interface that's already popular to me. |
|
| |
Regedit is not comprehensive. It only applies to the windows registry. |
|
| |
Trees are not very popular interfaces. What common desktop apps have you seen a *comprehensive* tree (I'm not talking about the trees in the various title bar menus -- note that this idea is a *comprehensive* tree) |
|
| |
I understand where you're coming from [lawpoop] but I have trouble with the concept of a tree as "comprehensive", insofar as tree elements could be (but are not generally) displayed as decision endpoints. Seeing where files, or even their calls (e.g., Windows™ Shortcuts), are placed into a tree does little to enlighten me beyond what I see when looking at a file folder view in Windows™ Explorer. |
|
| |
I find that concept maps are more illuminating and they may be made to layer effectively in order to explore their hierarchy as well as their order. I have linked to a resource site for concept maps. |
|
| |
For web interfaces this is very common - its called "Breadcrumbs" Because you can easily backtrack where you came from. (Halfbakery -> Ideas -> Computer -> Data_Organization) |
|
| |
(There's an MS caricature showing a person on a planet with a mouse pointer next to him, and two Martians asking him: Think well, what did you do to get here...) |
|
| |
In 1999 I was CTO and then VP R&D at a dotnet-bubble startup called Generize. The idea was that any information can be depicted as a tree, via XML, and that its easy to automatically create an intuitive user interface for creating that data, viewing it and manipulating it, either by example or by templates and rules, themselves saved in a tree, and mapped via XSL.
We actually made some large institution's websites and (early) web applications, such as forums, and content providing, customization and personalization using this idea. (Some are running to this day, code sold to large company). |
|
| |
In some of the views you could see the path to any data or action being accomplished, and could retrieve the data, action or view using that path. |
|
| |
Since MS Office moved to OpenOffice which is XML, and much of communication is done with SOAP which is XML, and applications are moving to XML (with MS XAML stealing from Mozilla XUL and W3's XSG) You'll soon be able to do this automatically. |
|
| |
But you will need to hide certain parts which would not interest the user. (Once you have a button, you are not interested in the
1. Borders: 4
1.1 Top
1.1.1 Length: 10
1.1.2 Width: 1
1.1.3 Line-Type: Solid
etc |
|
| |
I lead a team that developed such an application once (Unfortunately, the IP is not mine to link to, and it has not been released to the public. It may be resting on a gov't shelf somewhere). |
|
| |
It had an interface to write "adapters" for such things like file systems, database systems, etc. It had a filter engine to annotate branch nodes and cull out portions of the tree. |
|
| |
It turns out that looking at the file system as a tree is very, VERY boring (when you zoom out to screen width, the individual nodes are less than a pixel wide, and the tree doesn't even reach the bottom of the screen -- it depends on the file system, of course). |
|
| |
Things get a bit more interesting when you look at everything as a _graph_. We had to have specialized dynamic graph viewers which allowed the user to define rules for how the nodes got clustered together. |
|
| |
For a similar application, check out Protege. |
|
| |