But How Do It Know? - the Basic Principles of Computers for Everyone (29 page)

When a computer with a virus is running, it does all of the things it is supposed to do, but whenever it runs the program that contains the virus, the inserted jump instruction causes the virus program to be run instead. Now the virus usually will do something simple, like check for a predetermined date, and if it is not a match, then the virus program will jump back to the beginning of the file where the operating system program still exists.

Thus, your computer will appear totally normal, there are just a few extra instructions being executed during its regular operations. The virus is considered dormant at this point. But when that date arrives, and the virus ‘decides’ to do whatever is in the rest of its program, it can be anything. When the virus program is running, it can do whatever mischief the person who wrote it could think of. It can erase files on your disk, or send them somewhere else via the internet. One humorous virus would, every once in a while, make the letters on the screen appear to come loose and fall into a pile at the bottom of the screen.

Here’s an example of how to catch a virus. Let’s say that you have a friend who finds a funny movie on the Internet. It makes him laugh, and he thinks that you will enjoy it too, so he emails the movie file to you. You receive the movie file and play it, and you do enjoy it.

There are two different things that could have occurred here. If your friend sent you a file named “funny.mov,” and your OS includes a program that plays ‘.mov’ files, then the OS will load that program into RAM, and that program will read the pictures in the “funny.mov” file and display them on your screen. This is fine, the program that ran was something that was already on your computer. The “funny.mov” file just provided a series of pictures that were displayed on your screen.

But if your friend sent you a file named “funny.exe,” then when you ask the OS to play the movie, it will load “funny.exe” into RAM and jump to its first instruction. Now you have a program running in your computer that came from somewhere else. If it is a virus program, it will probably play the movie for you so that you don’t suspect anything, but it can do anything else that it wants, to the files on your disk while you are watching the movie. It will probably install itself and go into a dormant state for days or weeks, and you won’t even know that your computer is ‘infected.’ But sooner or later it will come alive and do whatever damage it was designed to do.

This sort of malicious program is called a virus because the way it works is similar to the way that real viruses infect living things. A real virus is a thing that is smaller than a one celled animal. It doesn’t quite qualify as being alive because the virus by itself cannot reproduce. They do reproduce, however, by invading a cell of something that is alive. Once in the cell, the virus uses the mechanisms of that cell to make copies of itself, which can then go on and infect other cells.

The computer virus also cannot reproduce or do anything else by itself. It needs to get into a computer, and somehow get itself executed one time by that CPU. When it runs that first time, it inserts itself somewhere into the operating system so that it will thereafter get executed on a regular basis. Those instructions will do whatever damage they are designed to do to the computer on which they are running, and they will also usually do something that is designed to spread the virus to other computers.

Firmware

Of course, RAM is an essential part of any computer. The ability to write bytes into RAM, and read them back out again is an integral part of how the machine works.

But in some computers, there are sections of the RAM that only get written to when the computer starts up, and thereafter these sections remain unchanged as the computer operates. This could be true in any computer that always runs the same program. Perhaps half of the RAM is used to contain the program, and the other half of the RAM is used to contain the data that the program is working on. The half with the program has to be loaded at some point, but after that, the CPU only has to read the bytes of the program in order to fetch and execute them.

When you have this sort of situation, you can build half of your computer’s RAM the normal way, and with the other half, you skip the NAND gates, and just wire each bit directly to an on or an off in the pattern of your program.

Of course, you can’t write into the pre-wired RAM, but you can read from it just fine. This type of RAM was given the name Read Only Memory, or ROM for short. You use it the same way you use RAM, but you only read from it.

There are two advantages to ROM. In the early days of computers, when RAM was very expensive, ROM was a lot less expensive than RAM.

The other advantage is that you no longer have to load the program into RAM when you first turn the computer on. It is already there in ROM, ready to be executed by the CPU.

The point here is a new word. Since software was named ‘soft’ because it is changeable, when it comes to ROM, you still have a pattern in the bits, but they’re not so soft anymore. You can’t write into a ROM, you can’t change the bits. And so this type of memory came to be known as ‘firmware.’ It is software that is permanently written into hardware.

But that isn’t the end of the story. The ROM described above had to be built that way at the factory. Over the years, this idea was improved and made easier to use.

The next advance was when someone had the bright idea of making ROM where every bit was set on at the factory, but there was a way of writing to it with a lot of power that could burn out individual connections, changing individual bits to an off. Thus this ROM could be programmed after leaving the factory. This was called ‘Programmable ROM’ or ‘PROM’ for short.

Then someone figured out how to make a PROM that would repair all of those broken connections if it were exposed to ultraviolet light for a half an hour. This was called an ‘Erasable PROM’, or ‘EPROM’ for short.

Then someone figured out how to build an EPROM that could be erased by using extra power on a special wire built into the EPROM. This was called ‘Electrically Erasable PROM’, or ‘EEPROM’ for short. One particular type of EEPROM has the name ‘Flash memory.’

So there is RAM, ROM, PROM, EPROM, EEPROM and Flash. These are all types of computer memory. The thing they have in common is that they all allow random access. They all work the same way when it comes to addressing bytes and reading out the data that is in them. The big difference is that RAM loses its settings when the power goes off. When the power comes back on, RAM is full of all zeros. The rest of them all still have their data after power off and back on.

You may ask then, “Why don’t computers use EEPROM for their RAM? Then the program would stay in RAM when the computer was off.” The answer is that it takes much longer to write into EEPROM than RAM. It would slow the computer down tremendously. If someone figures out how to make an EEPROM that is as fast and as cheap and uses the same or less power as RAM, I’m sure it will be done.

By the way, the word ROM has also come to be used to mean any type of storage that is permanently set, such as a pre recorded disk, as in ‘CD ROM,’ but its original definition only applied to something that worked just like RAM.

Boots

What do boots have to do with computers? Well, there is an old phrase that goes “pull yourself up by your own bootstraps.” It is kind of a joke, it literally refers to the straps that are sewn into many boots that are used to help pull the boots onto your feet. The joke is that if you are wearing such a pair of boots, and want to get up off the ground, instead of getting a ladder or climbing a rope, you can get yourself off the ground by simply pulling hard enough on those bootstraps. Of course this would only work in a cartoon, but the phrase has come to mean doing something when there is no apparent way to do it, or doing something without the tools that would normally be used, or accomplishing something by yourself without help from anyone else.

In a computer, there is a problem that is similar to needing to get off the ground and having no tools available to accomplish it. When a computer is operating, the memory is full of programs that are doing something, and when the operator of the computer enters a command to start another program, the operating system locates the program on disk, loads it into memory, and jumps to the first instruction of the program. Now that program is running.

But when you first turn on a computer, how do you get the operating system into memory? It takes a program running in memory to tell the disk drive to send over some instruction code, and the program needs to write that code into memory at an appropriate place, and then jump to its first instruction to get the new program running. But when you turn the computer on, every byte in memory is all zeros. There are no instructions in memory at all. This is the impossible situation, you need a program in memory to get a program in memory, but there is nothing there. So in order for the computer to get going in the first place, the computer has to do something impossible. It has to pull itself up by its bootstraps!

A long time ago, in the early days of computers, the machine had switches and push buttons on the front panel that allowed the operator to enter bytes of data directly into the registers, and from there, into RAM. You could manually enter a short program this way, and start it running. This program, called a “bootstrap loader,” would be the smallest possible program you could write that would instruct the computer to read bytes from a peripheral, store them in RAM, and then jump to the first instruction. When the bootstrap loader executes, it loads a much larger program into memory, such as the beginnings of an operating system, and then the computer will become usable.

Nowadays, there are much easier ways of loading the first program into the computer, in fact it happens automatically immediately after the computer gets turned on. But this process still happens, and the first step is called “booting” or “booting up” and it only means getting the first program into memory and beginning to execute it.

The most common solution to this problem has three parts. First, the IAR is designed so that when the power is first turned on, instead of all of its bits being zero, its last bit will be zero, but the rest of its bits will be ones. Thus for our little computer, the first instruction to be fetched will be at address 1111 1110. Second, something like the last 32 bytes of the RAM (235-256) will be ROM instead, hardwired with a simple program that accesses the disk drive, selects head 0, track 0, sector 0, reads this sector into RAM, and then jumps to the first byte of it. The third part then, had better be that there is a program written on that first sector of the disk. This sector, by the way, is called the ‘boot record.’

This word ‘boot’ has become a verb in computer talk. It means to load a program into RAM where there are no programs. Sometimes people use it to mean loading any program into RAM, but its original meaning only applied to loading the first program into an otherwise blank RAM.

Digital vs. Analog

You’ve no doubt heard these terms bandied about. It seems that anything associated with computers is digital, and everything else is not. But that’s not quite close enough to the truth.

What they mean is quite simple, but where they came from and how they ended up in their current usage is not so straightforward.

The word ‘digital’ comes from digit, which means fingers and toes in some ancient language, and since fingers and toes have been used for counting, digital means having to do with numbers. Today, the individual symbols that we use to write numbers (0, 1, 2, 3, etc.) are called digits. In the computer, we represent numbers with bits and bytes. One of the qualities of bits and bytes is their unambiguous nature. A bit is either on or off; there is no gray area in between. A byte is always in one of its 256 states; there is no state between two numbers like 123 and 124. The fact that these states change in steps is what we are referring to when we say digital.

The word ‘analog’ comes from the same place as ‘analogy’ and ‘analogous,’ thus it has to do with the similarity between two things. In the real world, most things change gradually and continuously, not in steps. A voice can be a shout or a whisper or absolutely anywhere in between.  When a telephone converts a voice into an electrical equivalent so that it can travel through a wire to another telephone, that electricity can also vary everywhere between being fully on and fully off. Sound and electricity are two very different things, but the essence of the voice has been duplicated with electricity. Since they are similar in that respect, we can say that the electrical pattern is an ‘analog’ of the voice. Although the meaning of ‘analog’ comes from this ‘similarity’ factor, when you make an analog, you are usually making an analog of something that is continuously variable. This idea of something being continuously variable has come to be the definition of analog when you are comparing digital and analog. Something that is analog can be anywhere within the entirety of some range, there are no steps.

Digital means change by steps and analog means change in a smooth continuous manner. Another way to say it is that digital means that the elements that make up a whole come from a finite number of choices, whereas analog means that a thing is made of parts that can be selected from an unlimited number of choices. A few non-computer examples may help to clarify this.

If you have a platform that is three feet above the floor, you can either build stairs for people to climb up to it, or a ramp. On the ramp, you can climb to any level between the floor and the platform; on the stairs, you only have as many choices as there are steps. The ramp is analog, the stairs are digital. 

Let’s say that you want to build a walkway in your garden. You have a choice of making the walkway out of concrete or out of bricks. If the bricks are three inches wide, then you can make a brick walk that is 30 inches wide, or 33 inches wide, but not 31 or 32. If you make the walk out of concrete, you can pour it to any width you want. The bricks are digital, the concrete is analog.