character
A letter, number, punctuation mark, or other symbol or control code that is represented, to a computer, by one unit (1byte) of information. A character is not necessarily visible, either on the screen or on paper; a space, for example, is as much a character as is the letter "a" or any of the digits 0 through 9.
character set
A collection of characters, e.g. A to Z, 0 to 9, punctuation marks, mathematical symbols, graphics and others. Each character in a set is represented by a binary value (sometimes called a bit pattern). Different character sets have different binary values for characters. Examples of character sets are ASCII, EBCDIC And Kangi.
client/server
A system in which a number of 'client' devices (usually PCs or workstations) are connected via a network to one or more data storage and retrieval machines known as 'servers'. When a user runs a program, their machine (the client) passes any requests for data (along with selection criteria and sorting instructions) to servers. Servers generally hold information in a database format, and should be capable of filtering out the required elements, putting them in order as instructed by their clients, and thus sending just the necessary amount of information back across the network. When the client machine receives the requested information, it processes it according to software held in its own memory. Another useful but more technical description of client/server is 'running software' in parallel on multiple nodes (many PCs operating at the same time on one Network), whereas mainframes and minis run software on one node only (the host machine). Client/server architecture is being increasingly adopted as a means of maximizing the effectiveness of multi-user systems where a number of different applications need to be running simultaneously. In control and telecommunications systems, the term used instead of 'client/server' is 'distributed processing'.
compression
Often called Zipping: Making a reduced-size copy of a disk file using compression programs. Most data or programs include a significant proportion of blank spaces or other repeated bytes - each one contributing to the amount of space occupied by that file on disk. A compression program copies non-repeated bytes in the normal way. However, when it detects repetitions it counts the number of repeats. Then instead of storing that number of bytes, it writes a code indicating the repeated byte and a code for the number of repetitions. These codes normally take only a tiny fraction of the original space, so that the compressed copy is often only 25 percent the size of the original. Compression is very useful for transferring files between machines, either via telephone lines or diskette. Unfortunately, software cannot normally use data or program files in the compressed state. When a compressed file needs to be restored to normal size, it is decompressed or 'unzipped' using another program which reverses the compression procedure. Examples of compression software are PKZIP, Compress, GZIP and WINZIP (for Windows).
Configuration Management of Client/Server
In Client/Server computing (as opposed to Terminal/Host computing), processing can happen anywhere - on the clients, on the application servers (Three-Tiered Systems), or on the Data Servers (stored procedures). The balance and control of the distribution of processing is a major factor in the success of a Client/Server Network. Two tools are necessary for effective Configuration Management: 1. Central Inventory - Identifies all software items, their type, where used and the upward (to servers) or downward (to clients) dependency. 2. Central Repository - Holds all application software programs and their revisions. The repository also holds the software tools for version control, configuration mapping and change control for all new software releases.