Online economics
Archives: October 2007

The future of operating systems?

(Warning: nerdy blog post that’s not really about economics).

There’s a fierce battle going on right now in the computer operating systems market. Microsoft has the dominant market share by far, but its latest version of Windows, Vista, has received quite a bit of criticism for being slow and having other problems. Apple has just released the latest version of its OS X operating system that is getting pretty good reviews. Then the Linux system, and Ubuntu in particular, has been improving by leaps and bounds.

Recently I read this article which explains how to run the latest version of Ubuntu directly off a USB pen drive. This got me thinking. What if you had a pretty big USB drive, say 32GB, then you could put the operating system and all your applications on there too. Then plug that drive into any computer, and that computer instantly becomes ‘your’ computer, with all your applications and preferences. This could be really useful … For example, I used to work at two different offices with two different desktop computers, plus another computer at home. It was a real pain keeping all these PCs set up with the software that I needed and the preferences that I liked. Things like web bookmarks are also a pain to keep synchronised. Nowadays I use a single laptop for home and work, but this has disadvantages too. I prefer working on a desktop PC with a bigger screen and proper keyboard, and I have to carry the damn thing to and from work every day. With the OS-on-pen-drive solution, most of these problems go away. Just plug in your pen drive to your home computer or work computer and it has all your programs, all your bookmarks, all your settings, etc. And it’s not much trouble to carry around with you.

The only problem would be accessing all the rest of your data that doesn’t fit on the pen drive. I have a hundred gigabytes or so of photos and documents, and those aren’t going to fit on a pen drive, at least not the ones that are available at the moment. But what if I also had access to an internet-based storage service that I could use to save this kind of data. Then I’d have the ultimate portable system — the ability to use all my programs and access all my data from any PC with an internet connection, without carrying a laptop. Sounds good to me.

by aaron. Permalink. Comments (0). Comments RSS.

Chatter up

Economics chatter on blogs has picked up significantly in the past few days. There have been huge spikes in a few keywords, for example “debt”:

debt.png

And “Credit”:

credit.png

Is it the subprime crisis part II?

by aaron. Permalink. Comments (0). Comments RSS.

Music by donation: Some data

I’ve been looking around the Jamendo site that I mentioned in my previous post. It’s a website that facilitates musicians to sell their music by donation, as Radiohead have recently done with much publicity. As well as finding a lot of cool music, I found that they publish the data on the donations that people have made to musicians on the site. Naturally, as a nerdy economist, I was much more excited by the presence of this data than the music itself.

With a little Excel trickery I cleaned up the data and did some basic analysis. The donations through the site are made to artists, and are not necessarily linked to specific albums or songs. On the page with the donations data, the name of the musician or group that each donation was made to is given. However, a few quick searches revealed that almost all musicians received very few donations. Because of this observation and since it would have been a lot more work, I didn’t attempt to cross-tabulate the donations with artists. I just analysed total donations per month. These can be made in US dollars or euro. I converted all euro donations to US dollars using the average euro/USD exchange rate for each month. I used the data from January 2006 to October 2007.

First some high-level statistics: Total donations were small. Jamendo currently claims about 69,000 songs are available for download (some other statistics are available on this page). Over the 22 months there were 1,454 donations made, for a total value of US$21,150. So each artist is receiving very little money, if anything. Here’s the number and value of donations each month:

count.png

Looking at the raw data, people generally make donations of round numbers, mostly multiples of 5 dollars or 5 euros. There were a few odd donations though, like 5.99 or 6.49. The largest donation was about $204. The smallest was $5, which is the default minimum donation that the website suggests. Across all donations the average was $14.55. This graph shows the average and standard deviation of monthly donations across time:

mean1.png

There’s a slight trend upwards over time in the average monthly donation, but a linear trend is not statistically significant. As you might expect, the distribution of donations (across all months) is skewed. Most donations cluster around relatively low values, but there are a number of higher donations. There were 12 donations of $100 or more. Here’s the histogram of all donations:

hist.png

Finally, the raw data is reported in US dollars and euro by donation, so assuming that Americans would choose to donate in US dollars and Europeans would donate in euros, we can test whether there’s a significant difference in altruism between these two groups. One thing that might bias this result is that Jamendo appears to be a Europe-based website, and many of the artists are European. There were also no US dollar donations before December 2006 (I don’t know the reason for that). And there are many fewer US dollar donations than euro donations (167 versus 1287). With these things in mind, the average for donations originally made in US dollars is $10.70, while the average for donations originally made in euro is $15.04. This difference is highly statistically significant. So it would seem that Europeans are more altruistic than Americans (at least towards the artists on this website).

Click here to download my Excel file of the data, if you are interested.

by aaron. Permalink. Comments (12). Comments RSS.

I want my free MP3

Hot on the heels of Radiohead’s experiment, Trent Reznor and Saul Williams are doing the same — offering an album for download for your choice of $5 or $0. I’m a little bit suspicious that they want to collect your email address whichever option you choose, but it seems to be legitimate. I wonder if they’ll have as much success with it as Radiohead apparently did though.

Here’s a few more sites where you can get legitimate free MP3 downloads, mostly of lesser-known artists: Jamendo, Podshow, SoundClick and ArtistServer.

by aaron. Permalink. Comments (0). Comments RSS.

In case you missed it

Some videos of interesting economics-related talks given at Google. Most are about an hour long.

Dimensions of reputation in electronic markets: Arun Sundararajan of NYU Stern School of Business talks about modelling and quantifying the information generated by online feedback mechanisms.

Laboratory experiments with reputation mechanisms for electronic commerce: Tad Hogg of HP Labs talks about using lab experiments to test the effectiveness of different reputation mechanisms.

An economic response to unsolicited communication: Boston University professor Marshall Van Alstyne talks about how to create an ‘attention market’ to solve the spam problem.

Market failure and market design: Harvard professor Al Roth talks about making matching markets work.

Algorithmic mechanism design: Noam Nisan of Hebrew University talks about how to apply mechanism design in computational settings.

Joseph Stiglitz discusses his book “Making Globalization Work”.

Internet advertising and the generalized second-price auction: NBER researcher Michael Schwarz talks about the “generalized second price” auction (GSP) used by search engines to sell online advertising.

Generating trading agent strategies: University of Michigan lecturer Daniel Reeves talks about how to use numerical techniques to approximate Nash equilibria for infinite games, and for intractably large finite games.

by aaron. Permalink. Comments (0). Comments RSS.
© Copyright 26econ.com 2008