First cut comment - obviously, this type of interface is much better than some horrible text based database lookup. I would love to see a tool that made it easy to build this kind of interface to time oriented data. However, it falls short for me in some important ways..
My feeling is that this interface is too cluttered and busy for the kinds of uses I'm interested in. It takes a lot of study to look at one of these diagrams to figure out what's going on. Lots of horizontal bars indicate "incidents", sure, but how serious are those incidents? I have to read the icons and the labels and such. LifeLines looks good for detail, but I want an interface that just gives me a quick overview first. Maybe LifeLines could be modified to do that, by taking some of the careful chart-oriented graphics and replacing them with something more fuzzy, organic.
I have the same frustration with this system that I did with Starfields - the system is too two dimensional, constrained. In this case it's even worse, since one dimension is always time you are forced into grouping or ordering all events in the other. For instance, the grouping of treatment vs. complaint in the medical example is clumsy - it's hard to know how they correlate. What are the other dimensions we could display information on? Make this selectable?
Gelertner's work on LifeStreams is definitely related work. He largely focusses on the storage and retrieval problem, but I know some of his group's papers have even presented this kind of visualization. Why was LifeStreams left out of the related work? Anyway, let's talk about LifeStreams in class.
Great ideas. I'm not sure what more to say here, since we've talked about them a lot here recently. In particular there are two Media Lab projects closely related to history in digital objects: Footprints and Patina.
One example of this history stuff being used in practice is Bonsai, a web interface for CVS repositories. It's used by large open source projects (such as Netscape Mozilla and GNOME) to help coordinate the efforts of lots of developers. You can see an active Bonsai at the Gnome pages. It's a really nice example of "Edit Wear" and "Source Code Wear", albeit without any sort of expressive display. Doing something expressive with Bonsai would make a nice project, and might even come in handy in our own effort inside the lab building Hive.
Ugh, this guy's page is ugly! The spiral binding and large font are very suggestive and specific, and the nonstandard Microsoft coding of quotation marks is obnoxious. His home page is even funnier, a sort of in-joke on Web technology. His pages are not to my liking. But it's his personal style, and that's OK.
Bricolage in web page design was greatly aided by the "view source" option in Web browsers. I'm very thankful that this feature has always been included in browsers - if you think about it, it's a weird thing to allow. But it made growing the web easy, as did the simple nature of HTML.
The home page has evolved an awful lot over four years. Has there been any writing about that? Oh, to have good archives. We should also talk about the vernacular of web page design.
I used to surf the web, look at random home pages. Now I only read people's home pages now when I've met someone who has already expressed interesting ideas online. For me, a home page is no longer the calling card - it's the background material.
I have a very ambiguous relationship to my own home page. It started out in 1994 as the class "show the world about me" site. These days I'm not so interested in doing that, though, and my personal stuff has fallen by the wayside.
Now my home page is supposed to be my professional page, a list of my brilliant accomplishments. I've collected a lot of academic detritus, papers and the like. All totally unorganized. I keep thinking I need to put my house in order, but then being overwhelmed by the amount of work it will take, especially to make it graphically snazzy. I also wonder now how much personal stuff to put up, whether it's appropriate, whether I risk prejudice that will affect me professionally if I talk about my personal life. So I think of having two home pages, one professional and one personal, but then I hate to present the idea that I'm two different people. I give up, explicitly constructing one's own identity is too difficult.
The main thing I want with my home page right now is more activity, more active process, more intelligence and interaction. That's what my Straum is about. I want my home page to be interactive, I want it to present active data, I want my home page presence to tune itself to whom is watching it. But that's a really big, abstract vision.
One option would be to create a home page with different facets to indicate my particular interests. For example, maybe we could have an old fashioned photo cube - each face then represents some facet of me.
Then my home page as a whole would be a collection of these cubes, arranged in whatever order made the most sense in the particular browsing context. It's a bit strange and literal of a vision, but it's a starting point.
| Nelson Minar | Created: November 3, 1998 |
| <nelson@media.mit.edu> | Updated: December 15, 1998 |