In the history of astronomy, there have been three distinct phases in how we detect the light gathered by a telescopes. The first phase, of course, we just to use the eye. For the first 200 or so years after the invention of the telescope and its first us by Galileo this was the only way observe. The light was collected by the telescope and delivered to the eye allowing you to see fainter. The eye is a good detector but essentially it reads out about once every 10th or 15th of a second, so we can retain the illusion of continuous motion. That limits the amount of light that's gathered by the eye before it delivers the signals to the brain and that limits the depth you can see. Beginning in the mid 19th century photography emerged as a medium for art and for registering images of people and astronomers quickly began to use it to record astronomical images. Photography continued to develop for over a century and was the prime means by which astronomers recorded images from the late 19th century through to about the middle of the 20th century. Famous astronomers such as Edwin Hubble used photography to record their images of galaxies in their spectra. Photography is an important technique for storing data that we can gather for a long period of time, but it's not noticeably more efficient in its detection capability than the eye, both are chemical detectors. Beginning in the 1950s and 60s the electronics revolution started to produce electronic detectors. For several decades these were things like image tubes that were amplifying light and propelling electrons and to a plate where they could record data. But they were quite finicky detectors and not very reliable, especially when used on a high-altitude mountain top. The true revolution for astronomy came in the mid to late 1970s when CCDs or charged coupled devices were developed. Flash forward 30 years and every major astronomical observatory and many amateur astronomers as well use CCDs to collect their data. CCDs are small detectors made of solid state material such as silicon, they're physically small, smaller than a postage stamp in most cases. To be excellent detectors they're usually cooled to liquid nitrogen temperature because this reduces the level of the background noise, which would, of course, convert into background light in the detector and so the darkness of the night sky can be recovered only by observing a chilled or cryogenically cooled detector. How does a CCD work? It's an array of small picture elements or pixels, square and very small typically 15 or 25 microns each laid out in an array and undressed by a series of charge transfers which lead charge be moved along the device first along the rows and then along the columns. Think of it like a bucket brigade where the tiny detector has a series of tiny buckets gathering light and then the buckets are tilted to move the charge or the water down a line and then down a column. What the CCD actually does is it turns the incoming photons into electrons and stores them in little potential wells. A typical potential well can hold a few 100 thousand or a few million electrons before it fills up. So there are limits to how much light CCDs can gather before their potential wells are essentially full. These are the same detectors that are found in your cell phone and your camera. So first photons are collected into electrons and stored in the solid-state detector, then they are read out and the final step involves turning those electron signals, essentially current in a wire, into a series of ones and zeros a digital signal. So the picture you see is a series of ones and zeros of intensity that originally was retrieved from a set of electrons stored on a physical detector. The CCD's used in your phone or camera are essentially the same devices used by research astronomers, but the research astronomer's tools are exceptionally high-quality, they convert essentially every photon that falls on them into an electrical signal which means they are perfectly efficient, they have essentially no blemishes or bad pixels and their ability to transfer the charge across the device is exceptionally good, usually 99.999 percent efficient. These are as close to perfect detectors as it's possible to make. We've already mentioned one of the large detectors used in astronomy, the CCD for the Large Synoptic Survey Telescope which is larger than a dinner plate so substantially larger than the CCD's you'd find in your phone or camera. This camera will have five gigapixels. So each time you hit the button and take an exposure an inundation of data, number of DVDs, worth will have to flow onto your computer. So the data rate of the largest CCDs is phenomenal. Information technology actually will struggle to keep up with the data rate from these large detectors on large new telescopes. The LSST has a phenomenal charge. It's a staring telescope that will scan the sky and look for things that change, it will essentially survey the entire Northern sky to the level of the Hubble Space Telescope depth every three days do it over and over again and over a course of a year or several years look for anything that varies for any reason whatsoever. This is a fundamentally new type of science and it's called celestial cinematography. We will probably learn completely new things about the universe by looking at it in the time domain as opposed the wavelength domain or just looking deeper and deeper. Let's take a moment to register and be impressed by the ubiquitous CCD. Your life is filled with CCDs, you undoubtedly own several maybe half a dozen or more without giving them much consideration. Astronomers use the same devices. They were first invented by two engineers at Bell Labs in the late 1960s. In a very interesting story, their small research operation was almost shut down because they could see no path to commercializing their devices. Literally, over lunchtime, they came up with a back-illuminated concept for their electronic detector, which became the CCD that we all use today. It's amazing to think that this project which was nearly closed down produced a device of which half billion were sold last year. Indeed as you watch this course you're seeing the results of CCD imaging. Willard Boyle and George Smith, the two engineers, were rewarded for their incredible efforts and their transformation effect on modern culture by the award of a Nobel Prize two years ago. The success of CCD has engendered an interesting problem for astronomers. Historically, the data rate and astronomy was not so large that we couldn't keep up even though computers were not nearly as powerful then as they are today, the detectors were modest and the data rates they produced could be handled by computers and networks of the time. But the detectors are growing in sensitivity and scale to the point where they produce data that will swamp information technology, that's one problem. Then, may indeed be solved by Moore's Law, whereby, the data rate and bandwidth is increasing by a factor of two roughly every 18 months. A second and more profound problem is that we have to understand the data, we have to analyze the data and that is a trick that astronomers have not solved. We don't know what's out there in the universe to be discovered. So it's very important not to throw the baby out with the bath water by getting rid of data that you don't think is interesting to find the objects that you do think it's interesting because that would suppose you know everything in the universe that you wanted to find. So astronomers agonize quite a lot about how to best analyze their data and keep up with the volume of data while getting every bit of important information and not losing potential discoveries. This data problem is shared by all types of astronomers and another version of this problem is shared by high-energy physicists like the people who work at the Large Hadron Collider because their data rates are every bit the equal of the Large Synoptic Survey Telescope. As an example of this issue, the Large Synoptic Survey Telescope that I've talked about will produce roughly 20 perhaps 30 terabytes of data a night. Because it's a publicly funded project by taxpayer dollars, the project has committed to making the data public and available online in real time. You will be able to go to your web browser, a handheld device and look for notices of supernovae or quasars or interesting things happening in the sky. Which means the project has to digest 20 or 30 terabytes a night and somehow compress, condense and deliver it to small handheld devices. They don't actually know how they're going to do that. So in a sense, it's just as well this project will not take first light for another five or six years. So to conclude, astronomy is in a regime where essentially every major observatory has a CCD or charge-coupled device as the electronic detector recording the light collected by the telescope. These are just fancy versions of the CCDs that you find in your cell phone and your camera. They are used ubiquitously and they are almost perfect detectors. Ironically, the fact that all the gains have been made on the detector side means that to do better, to see deeper and further astronomers now build even larger and larger telescopes to collect more light.