The slave of some defunct economist

In 1936, Keynes wrote, “Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slave of some defunct economist.” In 2008, no defunct economist is more prominent than Keynes himself.

via What Would Keynes Have Done? -Gregory Mankiw

Influence apps posted to ReadWriteWeb

Two influence applications were listed in a post at ReadWriteWeb: my Random Walk app, and a quiz app called Taught or Not.

RWW says: “There are big gaps in the data, which is understandable, but the interface is so much harder to use than Wikipedia’s that there’s reason to be concerned about expectations of substantial human editing.”

I agree about “there’s reason to be concerned about expectations of substantial human editing”, but I think they’ve done a good job working on the UI since last year. A collection of data is more abstract than a collection of encyclopedia articles – it makes sense that less people would visit or care to contribute to an abstract project. It’s also harder to assess the state of some datapoints than the state of an article. There are linguistic and stylistic cues that help signal unedited, “original thought” content on Wikipedia.

Flare Genealogy of Influence viewer

Martin Dudek made a really slick visualization of the Freebase influence data using the Flare library.  I think it’s the best so far: color coded influence/influenced-by/peer, a bio and image on the right, and a history of who you’ve looked at on the left.  Martin says it’s still a work in progress.  Link

Thinkbase: node-edge visualization of Freebase

This is cool.  Christian Hirsch at the University of Auckland put together a node-edge visualization of the Freebase graph.  The properties are represented as grey nodes, and the topics have icons depending on whether they are typed person, film, book, etc. (Link)

Powerset demo pulling Freebase data, including influences

I think this is very cool.  I heard Powerset’s latest demo was pulling Freebase data.  I went on Powerlabs and it pulls up Genealogy of Influence data for “Who influenced Thomas Jefferson?” above the other results from Wikipedia.

While this makes me excited about the future of both companies, it reminds me of a question I asked on this blog few months ago (Metadata spam on the semantic web):

I think it will be an interesting problem to block spammers, if something like Freebase or Twine becomes as useful as Wikipedia is now: as a trusted knowledge repository and therefore highly ranked search result.

This will especially be the case when spammers can figure out how to reverse engineer data to jump to the top of a search engine results page.

Joseph Priestly, Chart of Biography

I want to find a full image of this chart and put it online:

Joseph Priestly, Chart of Biography (1765)

A Description to the Chart of Biography (Google Books) has lots of defensive posturing. It reminds me of things I’ve said defending the genealogy data.

The proper employment of men of letters is either making new discoveries, in order to extend the bounds of human knowledge; or facilitating the communication of the discoveries which have been made already, in order to make an acquaintance with science more general among mankind. Вut few are qualified to make new discoveries of importance: a considerable share of natural genius, opportunity of making experiments, and a favourable concurrence of circumstances are requisite to it….

This Chart, which is about three feet in length, and two feet in breadth, represents the interval of time between the year 1200 before the Christian Eга and 1800 after Christ, divided by an equal scale into centuries….

It will easily occur to all my readers, that my greatest difficulty must have been the proper choice of names to fill this tablet of fame; and some degree of solicitude is certainly unavoidable when a man voluntarily assumes the province of the arbiter and dispenser of every man’s reputation, and when he sees all the dead pass, as it were, in review before him for that purpose. But this is no greater presumption than is implied in numerous other works, and in fact no more than one man’s giving his present opinion of others.

The nature of the design necessarily assigned some limits to the width of the columns and though in some ages there was room enough for all the candidates for fame, if I would have inserted them; in others, and particularly in modern times, where no reasonable column would admit a tenth part of the candidates, it, must require no small judgement to decide concerning their respective pretensions. With respect to this, I can only say that I have acquitted myself with all the impartiality of which I was capable.

Thankfully, with JavaScript, Flash and dynamic web pages we can get around the limits of space. From Wikipedia on the Chart of Biography:

Priestley organized his list into six categories:

• Statesman and Warriors
• Divines and Metaphysicians
• Mathematicians and Physicians (natural philosophers were placed here)
• Poets and Artists
• Orators and Critics (prose fiction authors were placed here)
• Historians and Antiquarians (lawyers were placed here)

Priestley’s “principle of selection” was fame, not merit; therefore, as he mentions, the chart is a reflection of current opinion.

He also made a Chart of History in 1769 and dedicated to Benjamin Franklin.

Free Infuencer: I love the net!

Another interesting app built on the Genealogy of Influence data at Freebase. (Will and Robert are showing me how to make the data part of a common domain.)

albertine meunier, net artist,
jérôme alexandre, developper,
cornelius reed, smart monkey

…have built Free Influencer which has a nice interface and unique URLs for viewing the influence data. Here’s the Nietzsche page, for instance:

They also have a concept of heaviness (which I guess counts centrality of the node?):