Berkman Web of Ideas: Everything Is Miscellaneous
This coming Wednesday I’m holding another in a series of open sessions on, well, ideas I’m interested in. This time will be a little different because I want to try out a presentation I’m giving at the TTI Vanguard meeting in SF in a few weeks. The title is “Everything Is Miscellaneous,” and I’m really not yet settled on what I’m actually going to say. But here’s the blurb:
For 2,500 years, knowledge was shaped like a tree. It had a root,
branches and leaves. Now that we’re digitizing all the information we
can lay our mitts on, it’s becoming clear that trees make sense within
the constraints of the real world but are far too limited when it comes
to organizing information in the digital world: Trees assume a leaf
really should be on only one branch, favor neatness over mess, are owned
by the people who own the knowledge, and assume the universe can be
known ahead of time. We are instead rapidly inventing new principles of
organization, from faceted classification to bottom-up
folksonomies. If we change the most basic principles of organization,
what will happen to knowledge and to the institutions that take their
shape from knowledge?
The session is open to anyone. It runs from 6-7:30pm at the Berkman Center’s Baker House in Cambridge (map). Best of all: Free pizza.
Categories: Uncategorized dw
Just to be clear, by ‘coming Wednesday’ you mean the 26th?
SORRY
You’ve been tested….
Keep in mind that faceted classification was invented before the digital era. In general, I wouldn’t associate tree like structures with non digital information at all. I would blame more recent technology for imposing that structure on us.
http://www.slais.ubc.ca/courses/libr517/winter2000/Group7/facet.htm
Everything Is Miscellaneous
Looks to me as a similar process to the global village process. Humans aren’t neatly divided into and limited by geographical locational locations as they were in previous centuries. Greater interconnection brings new, more complex relationship topolog…
Sure, but Ranganathan-era paper technologies can’t do the rapid sorting and tree-building we can do digitally, so faceted classification didn’t hits its stride until it went digital. Am I wrong about this?
I’m not sure it’s hit its stride yet. I think we’re still working on interfaces that make good use of it from the end users’ standpoint.
Have you looked at older classification systems that weren’t meant to sort out physical objects? I’m thinking of the International Classificatiton of Diseases which has its roots in the 19th century. It’s used today to track a lot of public health data info. I’m not familiar enough with it to know if it qualifies as “tree like” but something tells me it wouldn’t survive that long if it were strickly so considering the advances in medical knowledge. I’m mainly familiar with the description of it in Bowker and Star’s “Sorting Things Out: Classification and its Consequences.” If you haven’t read it, I think that book, along with Lakoff’s Women Fire and Dangerous Things would be of interest to you right now while you’re thinking about classification systems.
As a database programmer, I’m often consumed with data structures from the presentation and context of data to its electronic repository and design.
What struck me in reading about both you and Tanya’s interpretation of traditional classification systems is that the focus seems to be on those systems and nothing more.
It’s obvious that these systems are limited. I would proffer that even today’s knowledge-based systems (more connotations to that word exist than are herein expressed) continue to be quite limited in terms of the scope and understanding of data.
I often think of data in terms most eloquently expressed by the great psychological anthropologist Gregory Bateson who once wrote the following (I’m paraphrasing here): “Look at your hand. What do you see? You see 5 digits but that’s not what is really there. What exists are 5 relationships between these digits.”
Being able to qualify the RELATIONSHIPS among and between data structures (to me) epitomizes what it means to begin extrapolating some semblance of truth from data.
For technology to accommodate this objective will require data structures to create and manage complex, multiple relationships within data structures far beyond what exists today. However, neural networks present the greatest possible hope for the future success of this effort.
And even if such a future accomplishment were to be achieved, there still would remain one of the most perplexing of constructs in linguistics to comprehend – namely the Hermeneutic(s) (depends on how you view the term which is exactly the point!) of language. Simply defined, Hermeneutics represents methodological principles of interpretation.
Most respectfully submitted for your perusal and Hermeneutic reaction,
Brice Richard
Database Programmer
Washington, DC
Brice, I agree. In fact, it sometimes seems to me that our growing infatuation with tagging is based on the notion that we’ve somehow routed around hermeneutics — just tag it with this short word and now it’s findable — when in fact we’re just postponing the hermeneutics. Then we’re going to have a heck of a time facing up to the rich contextual ways in which tags mean things.