We have thought a bit about the trend leading from the 20th century "science cities" to the 21st century "city science." This is the turn from the "Big" science and technology toward more distributed research activities. What would be the necessary infrastructure for this transition?

Jeff Johannes, medicinal chemist at a large pharmaceutical company, wrote an interesting posting on the blog Sceptical Chymist (http://blogs.nature.com/thescepticalchymist/features/10_miles_from_academia/):

Independence is a really good thing. Some amazing discoveries have come from qualified people or groups that were allowed to truly explore their own ideas, free of external bias or constraints. One clear example of the power of this concept exists in the context of popular music. During the 20th century there was an explosion of diverse musical genres that continues today. Many factors contributed to this process, but one of the most important was the fact that musical instruments and recording equipment gradually became cheaper while at the same time becoming more widely available. This made music accessible to anyone who had a desire to pick up an instrument and create music. Moreover, they could use their own recording equipment to communicate their ideas to interested parties. Today, with the advent of computers and digital recording, musicians can make home recordings of a rather high quality and easily share their songs on the internet. It is truly an exciting time to be a musician.

In terms of accessibility and expense, chemistry, and most modern sciences in general, are way behind music. A budding rock star can buy a $200 guitar at a local retailer and record songs at home, but when I think of chemistry, I think of $600,000 NMRs and $100,000 LCMS stacks installed in the hallowed halls of the worlds great schools. I consider myself extremely lucky to have access to such amazing equipment. But many scientists don’t.

While modern science is more technologically complex than music, I see no fundamental limitation to increasing the accessibility and reducing the cost of doing research. I think this is one of the great challenges facing science. Inexpensive scientific instruments would empower new scientists, give more independence to existing researchers, and lead to an increase in creativity in scientific research.

The indicator to watch to see whether scientific research is indeed following the trajectory of popular music is the availability of the scientific analogue of the "$200 guitar at a local retailer." Are there efforts to lower the cost of spectrometers, DNA analyzers, and NMR machines? Why don't scientific instruments follow the path of PCs, in a way that exponentially improves size and performance?

This has implications for the Brazilian case as well, about which Alex Pang has posted a few signals. In order to have a more diffuse model of scientific activity, one would need more readily available instruments. Perhaps science version of "One Laptop Per Child"?