Wanted: A Database That Does Whats Needed

eLABorations: Current crop lacks vital organizational skills; Kapor's 'spreadsheet of the mind,' other offerings bear watching.

Last week I wrote about Microsofts Bluetooth keyboard and mouse and the role the pair can play in bringing some order to your desktop.

But as cool as Bluetooth peripherals may be, they can only play a very small role—a cameo, really—in getting and keeping your workspace organized.

After all, my computers desktop is always more cluttered than my desks desktop, and although its pretty easy to drag the contents of my virtual desktop into the trash, Ive kept looking for a more constructive solution.

From the first time I came into contact with a database product—for me, it was Symantecs Q&A—I was excited by all the great organizational stuff I could do with a database. However, this gave way to frustration when my initial design didnt do what I needed it to do, or didnt do it well.

By the time Id figured out the flaws, the data Id already entered had a sort of inertia that limited the changes I could make to the database without starting from scratch—my databases had a way of making me do what they wanted, rather than the reverse.

I wanted—and still want—to build a database rather than design one, the way you tear open a box of Legos, toss out the instructions and start building whatever you like.

This is why a headline from the Register last week, "Kapors open source spreadsheet for the mind," caught my eye. According to the article, Mitch Kapor, the author of the Lotus 123 spreadsheet, is at work on an organizational product with the spirit of a spreadsheet.

Spreadsheets are very popular, taking on many tasks thatd be better assigned to databases, because of the way they enable users to work with them—you take a blank sheet and start banging away, not necessarily knowing just where youre going when you begin.

The Register article, and the San Jose Mercury News story to which it links, dont go into too many details about what this spreadsheet for the mind will be, but heres what Id like to see: Id like to take a piece of data, perhaps that story from the Register, enter the text into a free-form database, and begin assigning attributes to it on the fly, such as the date I encountered it, the fact that it could be a column idea, and so forth. Those attributes would then be part of my database, available to be assigned to another bit of data the next time I try to catalog it. The database product would have to store its data in an open, accessible format, in case I want to switch to some other system in the future. This is what has stayed my hand every time Ive started to use Outlook and its category feature to hack together something approaching this.

I ran into the same wall with BeOS and its databaselike BFS file system, which fit the bill in many ways—but for the sad fact that Be is laying dead in some corner of Palm Inc.s headquarters (at least until one of the BeOS resurrection crews manages to revive the OS in open source).

XML seems like a good, cross-platform fit for an application like this, and, although details are also sketchy right now, Im holding out hope that Microsofts forthcoming Xdocs product can help fit the bill.

Until then, Im going to try putting a personal organizational system together based on ezPublish, an open-source content management system based on PHP, mySQL and Apache that stores data as XML. I came across ezPublish a couple of weeks ago, and Ive been very impressed with its polish and configurability—enough so that Ive migrated one of my personal Web sites from PHP Nuke to ezPublish.

Ill let you know how my ezPublish project goes in a future column, but in the meantime, Id love to hear how you are dealing with information overload on your desktops. ´

How do you get and stay organized? Drop me a line at jason_brooks@ziffdavis.com.