Computing's role in science makes the New York Times

Image courtesy of Nicolas Raymond.

Regular readers of iSGTW are probably well aware of how essential advanced computing is to scientific research. The same is not true, however, for the average member of the public.

That's why we must applaud the New York Times' recent article, “Digging Deeper, Seeing Farther: Supercomputers Alter Science,” by John Markoff.

Markoff writes:

The physical technology of scientific research is still here — the new electron microscopes, the telescopes, the particle colliders — but they are now inseparable from computing power, and it is the computers that let scientists find order and patterns in the raw information that the physical tools gather.

Computer power not only aids research, it defines the nature of that research: what can be studied, what new questions can be asked, and answered.

Of course, Markoff's article was not universally well-received. One reader complains that the article failed to mention BOINC, the widely used volunteer computing platform. Another, going by the handle “Greenpa,” disagrees with the pro-computing stance of the article, writing: One would never know, from reading this article, that there are many very serious scientists out here who do not share this enthusiasm for the ever-increasing prevalence of massive computing power.

Greenpa argues that, “The entire construct is so complex as to defy proof-reading of all the processes. So- in fact- much of it is never proofed, beyond seeing if when you ask it "what is the cube root of 64?" it answers "4".”

He goes on to describe supercomputing as seductive and tempting to scientists – as though that was somehow a comment on its validity.

We have our own quibbles with the article, although they are far less fundamental than Greenpa's complaints. Markoff's liberal use of the word “supercomputing” could give the impression that high performance computing is the only form of advanced computing that is crucial to scientific progress. It is unclear whether that is his intention; he may be using the term as a catch-all phrase for all forms of computing that go beyond a single workstation.

Markoff also suggests – based on quotes from Larry Smarr – that the big bottleneck that faces scientists today is the resolution we are able to display when creating visualizations.

That hardly seems the case to us. The most-discussed issues at conferences we've attended have related to the massive amounts of energy exascale machines will need, the struggle between maintaining security and agility while increasing ease of use for the less computer-savvy scientists, the dearth of developers who can properly optimize code for use on a complex architecture, and the difficulty of parallelizing existing software so that it can take advantage of today's ubiquitous multicore machines.

Regardless of our concerns, we're glad to see the New York Times writing about the value of computing to science. Too often, computing is the unsung hero of science, and that needs to change.