Open Source Networks Part II
When I wrote the last column about open-source research networks, I was sure that I would receive a lot of negative feedback—something to the effect that, Bill, you've absolutely lost your mind; no one would share their intermediate results in an open forum.
Now, it is probably true that I have lost my mind, but what is even more surprising is that many of you not only agreed with the concept, but sent me concrete examples where open-source research is already alive and well.
Of course, when I wrote the column, I already had a couple of aces up my sleeve, in case I needed them. Open-source research has actually been in use for many decades. The high-energy physics community has been conducting its research entirely in the sunshine (well, more explicitly in lead- or concrete-lined corridors) since the 1950s and 60s. Teams of scientists from around the world would collaborate at the site of one of the powerful particle accelerators in order to conduct state-of-the-art research. Results were immediately known.
Some say the Web was developed to serve the need for these teams to work across large geographic areas when they weren't together at the accelerator.
In the life sciences, there is an exquisite example: The human genome database has demonstrated the power of thousands of investigators sequencing different loci on the genome and placing that sequence data on a shared database for broad access. Perhaps the fact that the genome sequencing was completed on budget and ahead of schedule is a tribute to the open access to data and results.
Many e-mails I received have documented a number of other open-source projects, from clinical trials databases to software for analyzing gene sequences to research on autism. While there are likely to be thousands of different examples of open-source biomedical research, it seems to me that certain projects are particularly amenable to this approach:
1. Projects that generate large amounts of data, such as the human genome database, that can and should be accessed by multiple investigators. The larger the scale of the project, the more open-source collaborative networks will be the norm.
2. Projects that study rare diseases. While lung cancer has tens of thousands of affected patients and therefore many thousands of investigators studying the disease, autism, amyotrophic lateral sclerosis and other diseases with lower prevalence are not going to undergo rapid scientific advances unless the relatively small global community of scientists can team up, via open-source networking, to collaborate.
3. Software-intensive projects. For example, the Bioconductor project (http://www.bioconductor.org/) develops open-source software for the analysis and comprehension of genomic data.
Scientific progress depends, in no small measure, on the number of people working in a particular area of research. The more people the greater the interaction, and this interactivity drives productivity. Networked research is simply another way of enlarging the size of the research community. So, I leave you with this manifest Research workers of the world, don't rise up—connect up.
Open Source Networks Part II
When I wrote the last column about open-source research networks, I was sure that I would receive a lot of negative feedback—something to the effect that, Bill, you've absolutely lost your mind; no one would share their intermediate results in an open forum.
Now, it is probably true that I have lost my mind, but what is even more surprising is that many of you not only agreed with the concept, but sent me concrete examples where open-source research is already alive and well.
Of course, when I wrote the column, I already had a couple of aces up my sleeve, in case I needed them. Open-source research has actually been in use for many decades. The high-energy physics community has been conducting its research entirely in the sunshine (well, more explicitly in lead- or concrete-lined corridors) since the 1950s and 60s. Teams of scientists from around the world would collaborate at the site of one of the powerful particle accelerators in order to conduct state-of-the-art research. Results were immediately known.
Some say the Web was developed to serve the need for these teams to work across large geographic areas when they weren't together at the accelerator.
In the life sciences, there is an exquisite example: The human genome database has demonstrated the power of thousands of investigators sequencing different loci on the genome and placing that sequence data on a shared database for broad access. Perhaps the fact that the genome sequencing was completed on budget and ahead of schedule is a tribute to the open access to data and results.
Many e-mails I received have documented a number of other open-source projects, from clinical trials databases to software for analyzing gene sequences to research on autism. While there are likely to be thousands of different examples of open-source biomedical research, it seems to me that certain projects are particularly amenable to this approach:
1. Projects that generate large amounts of data, such as the human genome database, that can and should be accessed by multiple investigators. The larger the scale of the project, the more open-source collaborative networks will be the norm.
2. Projects that study rare diseases. While lung cancer has tens of thousands of affected patients and therefore many thousands of investigators studying the disease, autism, amyotrophic lateral sclerosis and other diseases with lower prevalence are not going to undergo rapid scientific advances unless the relatively small global community of scientists can team up, via open-source networking, to collaborate.
3. Software-intensive projects. For example, the Bioconductor project (http://www.bioconductor.org/) develops open-source software for the analysis and comprehension of genomic data.
Scientific progress depends, in no small measure, on the number of people working in a particular area of research. The more people the greater the interaction, and this interactivity drives productivity. Networked research is simply another way of enlarging the size of the research community. So, I leave you with this manifest Research workers of the world, don't rise up—connect up.
