Skip Navigation.

broadcast as in scatter seed

I might as well tell you up front, I’m scatterbrained. Everyone who was ever alive is alive to me still. If space makes it so everything doesn’t happen in the same place, time ensures it doesn’t happen all at once. But in my mind is neither space nor time. Everything always is.

No comments - but you could add one! »

  1. I, too, was a heavy FileMaker user in the 80s/90s. Like wikis now, FM was revolutionary for the time — certainly for the graphical layout, but, I think, more for the flexibility of structure it offered. Adding or deleting fields was trivial, and could be done by a simple mode change — no need to rebuild the schema and define migrations and all the other crap that (still!) comes along with most databases. There was no need for a database administrator. Granted, you could end up with broken databases if you didn’t do it right. And there were, and still are, many other problems with FM that made it far less than perfect; notably, the easily-changeable model fell apart with networked databases.

    From both a contemporary tech culture viewpoint and an IT viewpoint, it’s easy to say “just use a wiki,” regardless of the project. It’s a small step up from passing around Word files. Perhaps it is a necessary branch of evolution: as more people force wikiware to be structured databases, complaints (like yours) will be heard, and said wikiware will evolve to become more architecturally interesting.

    After all, most wikiware is implemented on top of structured databases, and there is a already schema defining the minimal structure of page title and page contents. Other structure can be derived: a list of links leading out of the page; a list of sections, each with its own header; a list of included media; a text index for searching; a subset of recently updated items for an RSS feed.

    Perhaps what is needed is the ability to define additional arbitrary fields, along with an easy templating system. The fields themselves could be defined in a wiki-like way: anyone can add a field at any time, anyone can delete one; revisions can be tracked and undone.

    Hyperdata evolving from hypertext.

    Which makes me realize I should probably go back and read what Ted Nelson has to say about this.

    Comment by John Labovitz — February 12, 2007 @ 12:02 pm

  2. Perhaps there is progress after all. See Ars Technica for “Meet the uber-wiki.” The article describes a domain-specific wiki (on biology, in this case) that seems to go way beyond the average big-ball-of-links wiki. For example: structured data, grouped associations (becoming topics), explicit relationships, and annotations.

    Comment by John Labovitz — February 15, 2007 @ 1:52 pm

  3. […] I’m indebted to John Labovitz for his comments on my wiki lament and for pointers to two articles about efforts to build a better wiki. The first, “Meet the uber-wiki”, is about WikiProfessionial’s WikiProteins, a wiki with structured data, automatic updates of related pages and alerts. The second is a piece by Tim O’Reilly about a recent NSF award to the University of Colorado to research and create a “New Generation Wiki” that goes “beyond existing wikis.” (It’s gotta be vaporware when one of the specifications is “utilize new paradigms.” As if simply “using” them weren’t enough. “Use”, “utilize,” do you know the difference?) […]

    Pingback by The Participant Observer » Blog Archive » Mmm technology, not merely man-made, but made of men! — February 24, 2007 @ 10:32 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

You must be logged in to post a comment.