Icelandicizing Your Computer

DOS / WINDOWS 3.X, NT, EUDORA, ETC.



by James Wilde, UK

First, some considerations...

I would first like to say that the whole deal of two-language typing on the computer is a good deal simpler under Windows 95. There you can have two languages as default and switch between them with two mouse clicks. I have Swedish and Icelandic as the defaults. English I can cope with on almost any keyboard! Three other considerations are as follows:

1. You can type Icelandic from any version of Windows 3.1 or above without difficulty, - IF - you have an Icelandic keyboard or at least know the layout of one. This is not usually a problem for the Icelandic letters, especially if you touch-type in Icelandic, but it can be a real chore to find the symbols which appear on, for example, the numeric keys on the alpha part of the keyboard, or to the left of 'z' and the right of '.', and beside the Enter key. Of especial significance is to know where the backslash, '\', is to be found.

hint: type a key plan of your keyboard once you've got you system configured.

2. Sending and receiving Icelandic characters will still work for anyone using another computer with Windows 3.1 or above, and for Macs, but not necessarily for any other kind of computer. If computers receiving Icelandic characters are not configured for Icelandic, they may get funny symbols like { [ ] } or letters like v e and d instead of some of the Icelandic letters. One soon gets used to this, especially as each symbol will always translate the same way. Most complicated is when you get =B4 or =F3 or = coupled with other hexadecimal combinations in the middle of a word, especially if you get two on the run. This last method is a way of coding the Icelandic letters so that they can be transmitted in 7-bit code. I'll put an explanation of that at the end, so those who wish can read it, and those who don't can skip it.

3. Some Internet sites through which your messages may pass may not be able to transmit Icelandic characters. Some of them corrupt the characters and pass them through, others refuse to pass them through and (usually) return them to sender. This problem and the previous one are actually linked as I will explain at the end.

Now here is the good news...

Only the writing of Icelandic is difficult in Windows 3.1 and above. Any Windows 3.1 system will present Icelandic characters on your screen properly if the application you are using will accept them. Eudora will accept them, of course. The following notes refer to Eudora lite version 1.5.2. Earlier versions will also handle Icelandic, but the method of setup may differ - in fact, I know it does.

Eudora Lite ver 1.5.2

In order for Eudora to display Icelandic properly you need to check that it is set up as follows:

In Eudora select the menu item Special then Settings. A dialog box will appear with a scrollable window on the left showing different categories of settings. Scroll down to Sending mail, click once on it, and, when the settings for that category appear on the right, make sure that May Use Quoted-Printable is checked.

Now scroll further down the categories until Attachments shows up and click once on that. On the right of the box now, make sure that MIME encoding is selected.
And that is that for reading Icelandic.

Now for typing Icelandic...

For this you need to turn your keyboard into an Icelandic one. This needs two files from Windows 3.1 which may not automatically have been loaded. The files are on disk 2, so have that disk ready.

Start Control Panel and double click on the International icon. Change the keyboard to Icelandic and click OK. Windows will now ask you to insert disk 2 in the diskette drive, and will install these two files for you. At the same time it will change your keyboard to an Icelandic one, so beware, because the '\' key is no longer on the bottom left of the keyboard.

Type your letter or whatever using the Icelandic keyboard settings - I can't tell you unequivocally what these are, because I have a Swedish keyboard which is different again from the US keyboard, and probably from the Canadian keyboard as well. Let's just say that, if your keyboard has

1234567890
QWERTYUIOP
ASDFGHJKL
ZXCVBNM

then Ð is to the right of P on the 2nd row, Æ is to the right of L on the 3rd row and the comma, ´, is to the right of Æ, also on the 3rd row, and Þ is the third key to the right of M on the bottom row, where I think you have , and . in between. Ö is the key to the right of the 0 key (zero) on the very top row. To get the magic '\' symbol you must hold down the AltGr key (right of the space bar) and press the Ö key (right of the 0 key).

To change back to your normal keyboard, you go into Control Panel/International again and reselect the US or Canadian keyboard. You should find that Windows will use the files already loaded from the original installation.

Both these procedures can be automated by using the little program called (I think) Recorder (I use Windows 95 now, so I've forgotten a lot of Windows 3.1 stuff), but how to do that I'll leave you all to figure out for yourselves. :) And that's it.


Now for the technical stuff I mentioned earlier...

To better understand all of these problems, you should know that computers represent letters, numbers and other printable characters as simple codes, the most common of which is called ASCII. This code arranges the characters in a given order and assigns them numbers from 0 to 127. 127 because this is 2 to the power 7 or 2x2x2x2x2x2x2 (that's 128, but with 0 as the first, the sequence stops at 127). Thus all characters in the ASCII series can be represented by this range of numbers. 10 of them go to 0-9, 26 to A-Z and 26 more to a-z. Then there are various symbols such as !"#%&/()=? etc and some so-called control codes.

In binary (that is numbers using only 0 and 1) this ASCII code can be represented by a number having 7 bits, so that 0 is 0000000 and 127 is 1111111, hence the term 7-bit code.

Unfortunately these 128 characters are not enough to include letters and symbols from non-english languages, such as the Icelandic characters, the Danish, Swedish, Norwegian ones and others found in French, German, Dutch, etc. Accordingly these characters have been represented by codes from 128 to 255. There is more than one version of the codes for these characters. One is called extended ASCII and another is called ANSI - there are more. Windows 3.1 uses ANSI codes.

Since there are now 256 characters in these expanded codes, and 256 is 2 to the power 8 or 2x2x2x2x2x2x2x2, these codes are called 8-bit codes and are represented by 8 0's and 1's, thus: from 00000000 to 11111111. So to write Icelandic on a computer, the computer must be able to handle these 8-bit codes in order to be able to represent the extra letters needed.

The problem is that many of the computers used for transmission of Internet messages only operate in 7-bit mode. If an 8-bit code, say for the letter Þ, comes at them, they may handle it in different ways. Let's say that the 8-bit code for Þ is 10110011. (It may be - I don't know offhand). Some computers simply strip off the extra bit, the first one, so that the code becomes 0110011, which might be the code for, say, 'd'. Thus, instead of the word 'Þig' one would see 'dig'. Other computers might refuse to handle the message at all, and may or may not let the sender know about it.

To avoid this, one has to find a way to convert 8-bit codes into something that can be reliably transmitted by a 7-bit system and equally reliably converted back at the other end. This is what the settings which I described in Eudora were designed to do - to use internationally accepted methods of encoding special characters. Sometimes something goes wrong, and it is then that one will see the strange symbols such as =FE in the middle of words. This is a method of converting 8-bit special character codes to 7-bit format but something has gone wrong in the translation.

Generally speaking though, anyone using Windows 3.1 or higher or a Mac will be able to read anything written by anyone else using a PC with Windows 3.1 or higher or a Mac. They may be able to read the output of other types of computers and have their output read by those computers, but this is not certain.

Hope this helps.
James

jwilde@algonet.se
jwilde@cix.compulink.co.uk


Top