Tuesday, July 10, 2012
Thursday, September 08, 2011
Visible using ordinary binoculars or amateur telescopes, this supernova experiences peak brightness tonight and tomorrow. Look for it in the Pinwheel Galaxy (M101) off the handle of the "Big Dipper" aka Ursa Major in the northwest sky in a clear evening just after sunset from the Northern Hemisphere of Earth. This exploding star is a Type 1a, the type used a "standard candle" for measuring galactic distances and the size and age of the Universe, and is only 21 million light-years away and was first captured on August 22. The last time a supernova of its kind was seen this close to Earth was 1972. It's a terrific opportunity for both scientists and the public to see one of the biggest booms in nature.
Numerous reports have been made including those by The Bad Astronomer, the PBS NewsHour, the International Business Times, the BBC, and Global Rent a Scope. Although it's expected to peak at 10th magnitude in the next couple days, SN 2011fe might be bright enough to see with small telescopes on Earth for a couple weeks.
Saturday, September 03, 2011
One thing we know about time in the universe is that it continues, far longer than any human can experience directly, into a great length, if not infinity. But how would you draw this concept and see an experience of the infinity of time, both in its size and detail?
Roman Opalka drew the concept of time's infinity by painting numbers, starting at 1, for the remainder of his life. He started in 1965. His obituary in the Economist describes this project well, defining the series of "Chronomes" as he completed them over the years. One might "see" time stretched out in detail before the viewer. In 1972 he passed one million. Before he died he passed five and a half million.
Wednesday, August 03, 2011
The Economist article "Parallel bars" points out the need for parallel computing and how the computer industry as a whole is failing to fulfill its potential. They also discuss it in their Babbage podcast. An excerpt of their podcast discusses it beautifully.
- "There has been so little progress in parallel programming, even though multicore chips have been widespread for five years."
- "What used to be quite an obscure programming problem for supercomputers and academics, which is 'how to you reliably and efficiently take advantage of massively parallel computers', is starting to become a real problem on the desktop."
- "If software doesn't take advantage of these cores, then chip makers can go on pushing out chips that have more and more processing cores inside them but the software won't be getting any faster, and we'll have this growing gap between what my computer is theoretically capable of and what it actually does."