Functional Form

Avoiding side-effects since 2004.

Wednesday, April 26, 2006

Who knows what we don't know? Should we assume anything is impossible? Check out this information about who controls the counting of votes in the United States.

Thursday, April 20, 2006

Tomorrow's Infrastructure

If this country were serious about maintaining an economic edge on the rest of the world for the next century, we would be taking economic stances that were less dogmatically free market centered. Many believe that we are successful because we have maintained a free market, and while that is true, the freeness of our markets is just one element in America's meteoric rise to power in the past century. We cannot forget the role of infrastructure. With highways, water systems, electricity, etc, America has allowed a high degree of regulation, and to great positive effect. These are the foundation upon which all further enterprise is built. Making certain guarantees about the availability and stability of such resources is critical to advancement. Now, as a new paradigm dawns, we are losing sight of the role of infrastructure in our current success, and we are failing to adapt our views about what should be considered necessary infrastructure in light of current realities. Because it has always been supplied at collective expense, the transportation network is taken for granted. Of course the road to my house isn't privately owned. Yet we have come to view communications as something that we must buy from owners, because providing it as infrastructure would be "infringing on the free market."

Yet with current digital radio technologies, it would be remarkably easy to network most major metropolitan areas in a federally-supported mesh network that would provide equality of communication at a reasonable public cost. Huge swaths of valuable radio frequencies are currently reserved for the highest bidder, a situation made necessary by fear of interfering signals turning the airwaves into unusable chaos. But scheduling airwaves in huge chunks via licensing is an outdated approach to the problem. Now ten customers in a café can all share the same narrow band to browse the Internet simultaneously with no interference. Digital technology has enabled more efficient means of sharing frequency than auctioning it on the open market. When the communicators are computers, they are smart enough to keep the peace among themselves.

Imagine: an America in which Internet connectivity is assumed anywhere. Information about our environment will be layered over our current reality, allowing unprecedented dynamism and freedom in human interaction. There's no more powerful accelerant to pour onto the economy. The efficient exchange of commodities relies on efficient dissemination of information. The more easily producers and consumers communicate, the better goods flow. And when consumers can communicate with other consumers without regard to time or space, the natural-selective processes touted by the free-marketists will also be amplified.

But the benefits don't stop with economics, at least not traditionally construed. Industrial capitalism has led us to view reality in terms of the exchange of commodities, the economy. But as we become more focused on the exchange of information, entirely new relationships with the world and metaphors therefor will develop. Just as the ubiquity of the automobile and freedom of transportation changed American culture drastically (albeit not necessarily positively), unabated flow of information is already changing us today. Like all paradigm shifts to come before it, the coming of the digital era offers a foundation for new solutions to a range of problems. And, also like other such shifts, it will undoubtedly bring replacements for the dilemmas it solves. But the solution to potential problems is not to cling to the laws of the universe that governed last century. We cannot stop the shift by denying it. We must embrace new technical possibilities, opening ourselves to new assumptions about how the world must work. Imagination must triumph over what we think we know.

Tuesday, April 18, 2006

Programming Language Update

Ranked by exposure level, though not necessarily level of mastery:
Java: a major NLP system for Hobbs, school projects, enterprise application development at SCME.
C++: School projects, namely operating systems and compilers course.
Scheme: Abelson and Sussman lectures / projects, statistical NLP project, lots of experimentation.
JavaScript: You know, for all that stuff.
Common Lisp: Natural language generation assignment for NLP.
Objective-C: Beginnings of a text editor, an interpreter, some tutorials.
C: School projects, robotics class.
Ruby: The requisite Rails experimentation and subsequent boredom, Lexical semantics project for which it was poorly suited. Former PHP Rubinistas: it's not that great! Don't drop your PowerBook in your rush to buy TextMate.
Flash ActionScript: Self-drawing interface from XML files for SCME, back in the day.
Perl: Lexical semantics project re-implementation, read the camel book, occasional shell scripts.
Scala: Denial that I'd have to program in Java. Some XML manipulating code / experimentation.
PHP: Not a fan.
Haskell: Part of The Haskell School of Expression.
MUMPS*: Dear InterSystems Caché, you're going to need a better disguise for you FORTRAN-era language guys.
Goo*: Object-oriented lisp-like language. Another Lisp to language in obscurity.
OCaml*: Latest prospect. I need a functional language that isn't Haskell or Scheme.
Io**: Cool looking prototype with no reserved words. What JavaScript could have been.
Arc***: We're waiting Mr. Graham. Looks like it really will be the 100 year language.
(* no coding experience, but more than an hour's reading)
(** 30-60 minutes of reading)
(*** vaporware)

Ranked by coolness:
Java + Eclipse + A billion .jar files

Ranked by lameness or other personal beefs:
BPEL (a language composed of lexically scoped hype)
Perl (has its place I suppose, a very small one, in my life)

Most active lately:

Hot on the radar:

Saturday, April 08, 2006

"If I were a butcher, I would call myself the cowww hunterrr." -- Emma Cunningham

Thursday, April 06, 2006

The girls and boys met.

After two semantics lectures, I am very happy that I'm taking this course. We're delving pretty deeply into the treatment of plural quantification, and it's intensely stimulating. The tool we're employing in our analyses is mereology, the proposal that groups of individuals can themselves be treated as individuals of type E. This extension of our ontology seems very natural to me, and it's right in line with Jerry Hobbs' advocacy of ontological promiscuity, in which anything that can be expressed in language can be represented as an object in the conceptual model. I'm gaining ever-increasing respect for the foundation Hobbs offered me in approaching these issues. The more I learn about other people's ideas, the more Jerry's impress me.

Wednesday, April 05, 2006

Don't put knowledge behind a wall

Hey academics: If it's not on the Internet, it doesn't exist. Post your content in journals I can access without paying $30 for the pleasure of a download.