June 2009 Archives

After having been in Bangalore now for a period of a few days, the first true "deep" observation I have had is that I and those who speak English like me do not necessarily speak the most widely used form of English.

Bruce Willis joked in the movie "The Fifth Element" that he only spoke two languages: English and Bad English. First-World educated people worldwide are likely to see much of the English spoken by those not like us as the latter.

I don't think it's as simple as American/Non-American however. There is a growing divide, based not simply on the way people are being educated but also in the way that they communicate. It seems like (and I have no data to prove this) those who speak bad English communicate with themselves very easily. People naturally customize things to the tasks that they seek to accomplish with them.

Language is no different. With that in mind, I would argue that the version of "bad english" that a taxi driver in Bangalore might speak is no less vital to his existence as the version that I speak is to mine. In fact, I often find that my version is ill suited to gathering information here, since my inflection is vastly different from the local norm. The differences in communication ability that I have noted are no different here than they might be if I wandered into one of many American towns where people speak differently than me.

I wonder if we're not at a time, similar in some respects to the rise of romance languages in Europe, where a common language is beginning to adapt itself into the dialects of many regions. Once adapted, it fuses with the existing lexicon to create new versions of the pervasive tongue.

This is one seed that might potentially sow the rise of peasantdom 2.0. For the moment, The version of English in which many informational texts are written is still in the "English" that is as one's primary school teachers taught. What happens when things begin to be published in the "vulgar" tongues?

After all, one of the great things about English is its universality. The informational content and quality of thought in a manuscript prepared for an Indian audience written in Indian English is likely to be vastly different than one written in world english for a world audience.

I wonder if it's something that either can or should be stopped.

I learned, via cryptogon, of a recent attempt by two Japanese nationals to creep into Switzerland via Italy. This wouldn't have been that strange, except they were carrying $134 billion (with a b) in possibly counterfeit US Government securities.

It gets even crazier. The securities are in denominations that are not available to ordinary human beings in the private market. Some of these bonds are available only to states and carry billion-dollar denominations.

So, if they're real, this is a pretty big deal. Italy gets a finder's fee and the private market becomes aware that the Japanese government or some other nation-state actor (the only people who could get these things) want to do something with their US government debt holdings.

If they're fake, it's even more interesting. Though it's of course fascinating to surmise that foreign governments are dumping US Treasuries, what about the supposition that there might be in circulation a quantity at least as large as $134 Billion in fake Treasury obligations? Obligations so realistic as to be indistinguishable for 7 days to Italian authorities? This story came out there in a press release dated June 4.

I was delighted to learn somebody is trying to create a new, open-source model for producing opinions on creditworthiness. Any attempt to replace broken credit rating agencies as decision makers for who receives capital should be greeted as a wholly welcome step.

Freerisk is not yet - and does not seem to hope to be - a replacement for rating agencies in their entirety. It is instead a way to lever individual analysts by allowing them to query across a wide range of SEC filings at once using a relatively simple language called SPARQL. In addition, it provides a framework for people to share insights from these statements with each other. In time, the creators hope to move to a more user-friendly way of getting data from their database (which is wonderful).

The creators are toying with the ability to include comments and insights from footnotes into standard query data in hopes of creating a fuller picture from a standard query.

A great potential "value add" for this comes in the example of highlighting an interesting footnote. Having read company filings, I can attest that a great deal of their footnotes are pretty boring. There are occasional gems contained therein which make the process worthwhile, but wouldn't it be nice if there was a way to get quickly to the interesting ones?

The difficulty that Freerisk and other projects like it will face in replacing credit rating agencies is that consumers have come accustomed to regarding credit simply: as a letter. There will always be a market for the distillation of knowledge that can be gathered off of a platform like Freerisk into a format that allows for simple, "thumbs up/thumbs down" evaluation.

If anything, I think Freerisk is simply an excellent substitute for certain functions of a Bloomberg terminal for people willing to learn a Query language (not many). It's nice to see so much data opening up for public consumption, especially with the ability to share relevant insights on a public forum.