Chapter 5 in the computers (Long & Long) text.
Fig. 1-4 in your Computers text shows the Four Primary Hardware Components of Computers - Input, Output, Processing, and Storage. These components are connected by the bus. We have already covered the processing part (CPU and Main Memory) and the storage (Auxiliary Storage or Secondary Storage) part. This lecture covers the last topics on hardware components - Input devices and Output devices.
Input devices transform signals from the world into bits that computers can work with. Output devices display the bits in ways that are understandable by humans.
The BIOS. BIOS stands for Basic Input/Output system. It is stored in ROM chips on the motherboard. The BIOS ROM is accessed by the boot program when your computer is turned on. The BIOS is specifically designed for the components connected to your computer to make them work together. We have compared the characteristics of the BIOS with other memory components and storage components.
Input devices transform information from the world into the bits that computers store. The most common input devices are keyboards and mice. A variety of other input devices exists and are described below as well as in the text.
The process of communicating with computers involves converting signals that are understandable to the outside world into signals that can be read and "understood" by the computer. One important set of symbols used by the computer is the ASCII code set. There are many ways that ASCII characters can be entered into a computer. The most common in use today is the keyboard. We are used to using a keyboard that has the name "QWERTY" for the characters that are on the upper left positions of the alphabetic keys. The QWERTY keyboard was originally designed for teletypes (used for telegrams) over a century ago. Since those were mechanical devices, they had a number of restrictions that have carried over to today's keyboards. Keys had to be linked by direct bars to a printing mechanism; hence the keys on a QWERTY keyboard are offset slightly from each other in vertical alignment to permit such bars to be attached. The keys were laid out on a horizontal plane to make manufacturing easier. Because original teletypes were given to jamming if keys were pressed too fast, the position of letters on the keys was deliberately designed to slow down typists; this layout persists today. Of course, the overall size of the keyboard was set so that hands had to be placed together to type; and a single "space bar" was used to allow typists to enter a space with either thumb (but most typists use only one thumb).
This design has persisted to the present day despite its many defects. "Touch typists" for many decades have been able to type text without looking at keys, and not had any major physical problems. However, typing data into computers is a bit different. There is more repetitive movement without changing position, and there is more need to use control keys, which makes it difficult to "touch type." The result has been that a new set of work-related injuries have arisen. They are referred to as "Repetitive Stress Injuries (RSI)," and they are beginning to become a very important part of Workmen's Compensation problems for U.S. Industry.
There are several alternate designs for keyboards that reduce or eliminate some of the design problems of the QWERTY keyboard. One, the Dvorak keyboard, repositions placement of letters on the keyboard to make typing English text easier. This makes it easier to type faster, but it does not solve other ergonomic problems of the QWERTY design, since the same physical layout of the keys is preserved. A second design, the "Tony Keyboard," named for inventor Tony Hodges, allows the keyboard to be split down the center, and then each half can be rotated and tilted to suite the desires of individual users.
The third alternative design for keyboards is called the Maltron Keyboard. The distinguishing features are: keys are separated and placed in a pod-shaped configuration to allow fingers to reach all keys without moving wrist or arm; multiple keys typed by each thumb; alternate layout of keys (QWERTY or optimized); separated left and right hand pods; vertical alignment of keys. This keyboard, like the Tony, has been shown to reduce Repetitive stress injuries (RSIs), and a test of its ability to solve problems of this type is now being designed here at Davis. Another input device that you will become familiar with is a mouse. A mouse is a device that allows you to move a cursor around the screen. In addition, it has one or more buttons on the top that you can use to "click" (or double click) in a given area to indicate that you wish the computer to take some action. Macintosh mice (mouses?) have one button; many IBM PC mice have two, and most associated with large workstations have three buttons. Each button causes a different action to take place.
The mouse is good for input when no characters (letters or numbers) need to be typed. When there is a mixture of typing and moving the cursor, it become significantly less useful (someone once said it's great for people with three hands). It also tends to lead to "closed end" interface designs: programs in which the user can only select from a pre-defined set of options, and there is little room for creative choices. More and more programs on PCs now have mouse options, though most of them also allow you to use keys (arrow and other) instead. For instance, using Word Perfect under DOS version 5.1 (the version available at 1131 Meyer) makes it possible to use a mouse for cursor control.
Although the most common forms of input to personal computers are keyboard and mouse, there are many other ways to convert data to machine readable form. We will only touch on some of the ones that are available. However, you should be aware that many options exist, and some of them may make input a great deal easier for certain types of applications.
A widespread form of input device is based on using bar codes. There have been several different forms of bar codes developed, but the one used for Universal Product code is probably the most widespread. As a result of international standardization, there are now product codes assigned to most goods sold in stores of all kinds. An international registry of these codes is maintained, and a manufacturer wishing to introduce a new product must apply for a code number before it is printed on the product package.
Bar codes can be used to code alphabetic information as well as numeric. I received a request for name, address and other update information on which there was a bar code for each data element (text as well as numeric); I was asked to write in corrections and the returned form would then be scanned in, a form displayed, and a clerk would update the fields that needed changing. Since there are now hand-held as well as table-mounted bar code readers, they can be used in many areas to improve input speed. Checking out library books is another good example, with a much shorter time required to check out books than was previously the case.
Another widely standardized form of data entry is used in the banking industry. The checks that you write are pre-coded on the bottom left of each check to provide the bank number, your account number, and the last three digits of your check number. The bottom right of the check is used to input (manually) the amount of the check. The ink used to print this part of the check contains little specks of ferric oxide (much like a magnetic tape), and these specks are magnetized just before the check goes through the reader (which can accept many different sizes of checks). The numbers and codes are of a peculiar form designed to minimize the possibility of error in reading. Sometimes, if the data are hard to read, a new stripe is attached to the bottom of the check and new codes printed; then the check is passed through the reader again.
The interesting point about this method is that its success is due to the international standardization within the banking industry. That form of standardization makes it possible to write a check in Europe or the Far East and have it eventually find its way back to your bank here in California. The standardization includes the shape of the characters and an agreed on code for each member bank or related institution.
A fairly large class of input devices are based on various techniques for recognizing marks on paper through optical reading techniques. One type is called mark sense forms, in which the user fills in small boxes (usually multiple choice) on a pre-printed form, and the position of the filled in boxes is translated into some code. Mark sense forms are often used for multiple choice exams, for surveys filled in by untrained users (health forms, demographic data surveys, etc.). The technology allows for users to erase choices (useful in multiple choice exams) by using what is called dark mark logic, meaning that the computer selects the darkest mark within a given set as the most likely correct one. Mark sense forms can be used to collect alphabetic information by having 26 choices instead of the 10 used for numerics, but the rate of user error in alphabetic mark sense forms is usually higher. Mark Sense forms are useful for collecting and processing large volumes of data where the application has been well planned out in advance and the responses are limited to very restricted choices.
A second technique for data input is to use optical scanners to create a bit map image of a graphic image (picture, drawing, or even mixed text and graphics). This technique digitizes the image. The resolution of the image varies, but often it is approximately 200-300 points per inch. Each point is referred to as a pixel. Each pixel is scanned for the darkness of that point on the page, and the a pixel value is assigned (usually between 1 and 256, so that each pixel can be represented by one byte). The aggregate set of points is then assembled into an image and stored in digital form. Note that the accuracy of representation depends first on how many individual points are analyzed per inch, and second, on how many shades of color can be represented. Once digitized, however, the data will be constant from then on.
Optically scanned images are not recognized by the computer as being anything other than a set of dots with various shades of grey. Even if there is text, the computer does not know what the text means, and it would not be possible to edit that text in an alphabetic format. It is, however, possible to edit graphic images, cleaning up unwanted smudges and adding new text to the image. This technique can be used to make computers generate fax images which are then transmitted automatically by the computer just like regular fax machines.
Because each image must be stored with some code representing each pixel on the page, optically scanned images require a large volume of storage. The two simple drawings in Lecture 5, for example, require 28K bytes and 38K bytes respectively, much more than the rest of the text in that set of notes. A more sophisticated form of optical scanning involves optical character recognition. In this method, the computer takes a scanned image and looks for alphabetic information, then converts the scanned image to ASCII format. Because the resulting data is ASCII, the total storage of optical character input data is the same as that of regular text files. The only difference is that Optical Character Recognition (OCR) programs will usually flag characters it has trouble reading with an asterisk or some other marker, so the unedited file is slightly larger than the number of characters in the original text.
OCR input is very useful when a significant volume of data must be input to a computer and there is only a printed page available. If the page is reasonably clean, the current OCR programs do a pretty good job of scanning in the characters with minimal errors. Incidentally, this technique is now being used with some success (though not by any means perfect) in reading Chinese and Japanese texts. We are working with some researchers at Cornell on a project involving input of a large volume of Chinese Buddhist literature.
A few years ago, voice input was regarded as a technology that was too imprecise to use in commercial applications. That perception has changed, thanks to the development of increasingly sophisticated voice recognition systems, some of which run on personal computers. Some languages lend themselves more readily to voice input. Japanese, for example, has fewer different sounds and less rise-and-fall in speech, so it tends to be a little easier to recognize than, say, English. Chinese, on the other hand, has four "tones" that can change the meaning of a given syllable, making voice recognition harder. In general, voice input can play an important roles in situations where other alternatives are harder to use. Jobs requiring constant observation of a production line, for example, would be disrupted if the user had to turn away to a keyboard or other input device. If the user can speak notes to a system (using a very limited vocabulary), then the user would be able to concentrate on the job rather than the input.
A new family of input devices is based on using some form of writing instrument. Pens, light pens, or similar hand-held instruments are the most common. A whole new industry of pen- based personal computers is just now coming on the market. They are especially popular in Japan, where many people do not know how to type using western characters and inputing Japanese characters is an awkward process. You should be on the lookout for new pen-based systems, since they are appearing with increasing frequency in newspaper articles today. Check them to see if they describe different types of operating systems to support this input method. This new approach is described a great deal in newspapers today, and some folks believe that it may spawn a new revolution in personal computing. What do you think?
There are a number of alternate forms of input that can be devised to help people with different types of physical handicaps. Devices held in the mouth and used to point to keys are often used by quadraplegics. Devices have been devised to conform more conveniently to the foot (for people whose arms are unusable). Helmets that recognize what the user is looking at can be used in a way similar to a mouse (some way to "click" is still required).
The field of input continues to be a fascinating one. You will almost certainly see new references in journals and newspapers to still other input devices. Watch for them, and the next time you have an unusual input problem, see if there isn't a "better" way to solve it than the obvious answer of using a QWERTY keyboard.
Monitors contain "dots" called pixels (short for picture elements) and are described by their resolution (number of dots vertically and number of dots horizontally), the variety of colors each pixel can represent, and the dot pitch (closeness of the dots on the screen). The early IBM PCs with color monitors had a 320 by 200 pixel resolution monitor that was called CGA (color graphics adapter). Typical resolutions seen today on PCs are VGA and SVGA. VGA, which stands for video graphics array has a resolution of 640 by 480 pixels with 256 colors (8 bits per pixel to specify the color from a larger palette). SVGA (super VGA) has a resolution of 800 by 600 pixels or 1024 by 768 pixels and 16.7 million colors (24 bit). For dot pitch, smaller values will produce images that are more clear and crisp (.26 mm is common). Larger distances between pixels can make the images appear fuzzy and fatigue your eyes quickly.
Non-impact printers include laser printers and ink-jet printers. Laser printers have very good resolution (300 or 600 or more dpi or dots per inch). Color laser printers are still quite expensive. In contrast, color ink-jet printers of reasonable quality can now be obtained for a few hundred dollars.
A videotape lecture on input devices by Prof. Walters is available for viewing in the media center in 1101 Hart Hall. It is: ECS015 - Video 8 - Input Devices.
Go to the index of
lectures for ECS15 - Fall 1997 .
Go to the homepage for ECS15 - Fall 1997 .