How Computers Work
Part 8 – Files
Computers are generally filled with files. These files are generally written in one of three types:
1. Text based files, which are commonly called ASCII files. These are by and large word processing files. They often have extensions such as TXT which is short for text.
2. Number based files, which can be either binary (using the digits one and zero) or hexadecimal (using a number based system that counts from zero to fifteen using letters A, B, C, D, E, F for numbers 9 to 15)
3. Text and number based files, which are a hybrid used to store structured data in some software applications such as data bases, spreadsheets and formatted word processing files like those with the DOC extension used by Word. Image files are always numeric, even though when you open them in a text based word processor they seem to be gibberish text, that is simply how the numbers are created.
All files have names. This name is generally several letters or numbers ending in a period or dot ( . ) and followed by a few more letters or numbers to the right of the dot. These digits to the right of the dot are generally called a file extension and your computer works with these files based on the type of extension. For example, TXT means text, DOC means document, dbf means it is a dBase compatible data base file, xls means it is a Excel spread sheet, JPG means it is a jpeg format image file, etc.
Large computer systems used in business since the 1960’s were able to deal with more complex file name and more specifically the first networking system, which was AT&T Unix as developed by the various colleges in the world, made extensive use of the dot and multiple dots in their file names.
Home computers were based on a much smaller format, specifically 8 or 16 bits starting in the 1970’s, so the file names had to be reduced in total size to accommodate the way files were found and put into use. The first home computers were based on file names about 8 letters or numbers to the left of the dot and three to the right.
Starting in the late 1980’s with the advent of larger format computers used by Apple (Macintosh), IBM (OS/2) and the PC compatible using Windows 98 longer file names were allowed, however in the case of the Windows operating system these had to be translated back into the older DOS based 8 letter names with only 3 letters of extensions.
Many people don’t fully understand file types, file extensions and cross-platform porting of files – especially those who got into computers in the 1990’s after Macintosh and Windows 98 allowed for point and click operations with file names of any size having automatic extension handlers.
Even if extensions are hidden from you view by the operating system (OS) they are required if you want to move or share those programs with other people on other computers. The MP3 extension is used to exchange audio files between many systems, while JPG and GIF extensions are used to exchange common image files between systems. Without these extensions and the proper file formatting (which is done by the software when you have a file) someone on a PC can’t see or hear a file created by a user on a Macintosh system.
Image files on some older computers generally can’t be used on another system unless you have special translation software. Image files from DOS based games such as Doom, for example, can’t be viewed on a Windows based system using many photo viewing software programs, because no one has written translations filters to work with them. Basically there is not much difference between the associated data in the file of a DOS image and that of a JPEG. Both files are made up of numbers, but the sequence and preparatory information differ a little, which is why you won’t be able to see the file.
Preparatory information in a file is known as a “header” and this information might contain information about the total size of the image, the resolution or bit depth for viewing the file, etc. After all this “header” information comes the actual number data that tells each pixel what color to display.
By making a custom “header” file a given software company is allowed to copyright or in some instances patent their file (or at least no step on those property rights held by other people). By making a special header file given company becomes associated with that file, often in the hopes that other makers will license the technology because their process is superior to another process.
This same problem occurs with data base files. Basically the information contained in a dBase, Access, SQL or Oracle file is essentially the same, but the “header” information is different as are the formatting marks.
Formatting marks are used to help create the visual display or layout of a given page, such as multiple columns for a Word document or a table in a spread sheet.
Many software companies openly provide information about their file headers and then it is up to a company to buy, license or create “filters” to read that header and set up the page. Not every company does this, however, thus in some photo programs or audio software you can’t see or hear every file type ever created.
The header file works much like the way boot up in a computer works. The software reads the front part of the file and then attempts to match the information it finds with known instructions. If the software can’t match the data up to existing parameters a file read error message occurs.
When you select a “file type” from your file selector box the software sends the incoming file information data to a special sub-programs often called a “filter” which translates the header file for the software. This allows a photo viewing program, for example, to see a DOS image file made with GW BASIC or used by the game program Doom, because the header filter knows what all the parameters are and puts them into more common terms the viewing program understands. As far as your photo viewing software is concerned once the DOS image file has been translated it is just another JPG image (but in reality it is not, as the file data information is a little different).
Not all filters work fully and completely, so while you might be able to load an old word processing document file made in Aimee Pro not all the formatting elements may come out looking the same when you bring it into Word for Windows. Embedded pictures may not load, the fonts may not be the same and alignment of the text could be irregular. The sheer fact that Word for Windows was even able to open the file and bring up the text for you to read is usually a major feat!
Many software programs put hidden files on your system in strange places that can only be read by the software. This, as an example, is how date shareware expires after 30 days. A date file is hidden deep inside some secret folder, deep inside your Windows operating system folder. Since Windows contains thousands of little files and folders it’s virtually impossible for you to ever find this date file, but the software knows where it is! When it checks the date against your system clock it won’t let the program run and some software makers put a locking bit inside the system so that once the system clock goes past the date the software will no longer run even if you change the system date back several weeks.
Much of today’s software is made up of many, many little files that are opened by other files. These allow a small program to do lots of functions because it will run a dozen mini-programs in the background to handle all these tasks. Included in those files are “keys” and “locks” designed to disable programs not installed with the original disk, out of date or installed more than once with the same serial number. This is basically how piracy protection works and it’s just another files in the maze of files on today’s average computer!