Mike Easter wrote:
> population, you shouldn't assume MS-centric, /n/x-centric, or
> Mac-centric and that -2- you shouldn't even assume that the /n/x-ers are
> going to be using UTF-8, because typically the default is not UTF-8 and
> many newsreaders users don't know how to reconfigure and -3- what you
> UTF-8/ed below shows as total 'garbage' characters in a Windows system

What I "UTF-8/ed" displays correctly for most users of Thunderbird.
Heck, it even works for my computer-illiterate friend in Outlook Express!

> and displays 'incompletely' in my Mepis Tbird which is default
> configured not UTF-8 and also displays incompletely when I reconfigure
> Tbird to use UTF-8 instead of ISO 8859-1.

Thunderbird on Mac, and any _correctly_behaving_ news or email client
on any platform states in a header what encoding method it is using.

Thunderbird on Mac, and any _correctly_behaving_ news or email client
on any platform reads those headers and renders the glyphs appropriately.

So there are three things to do:

1. Dump the broken software and get some that works

2. Find out how to type the characters you want.

3. Once you know you can send and receive properly,
when you receive some garbage, don't let it bother
you--at least you can be glad it happens a lot less
than before (1).

Wes Groleau wrote:
>
>> Yes, but HOW did you do that?
>
> See above. I would like to know the best way to do it on Kubuntu.
>
I don't know about 'best' but the simplest way I have found is to use
the char map application found under Accessories and just paste
characters in from there.

Cheers

Ian

Ian Thompson-Bell wrote:
> Wes Groleau wrote:
>>
>>> Yes, but HOW did you do that?
>>
>> See above. I would like to know the best way to do it on Kubuntu.
>
> I don't know about 'best' but the simplest way I have found is to use
> the char map application found under Accessories and just paste
> characters in from there.

For a few Greek letters, that's probably OK.
But for the 2,000+ glyphs used in Japanese, ....

Wes Groleau wrote:
> Ian Thompson-Bell wrote:
>> Wes Groleau wrote:
>>>
>>>> Yes, but HOW did you do that?
>>>
>>> See above. I would like to know the best way to do it on Kubuntu.
>>
>> I don't know about 'best' but the simplest way I have found is to use
>> the char map application found under Accessories and just paste
>> characters in from there.
>
> For a few Greek letters, that's probably OK.

Which is pretty much all I need.

> But for the 2,000+ glyphs used in Japanese, ....
>

Indeed, a whole new ball game but one I an fortunately not playing.

Cheers

Ian

Florian Diesch wrote:
[..]
> <Multi_key> <g> <S> : "Σ" U03C3 # GRREK CAPITAL LETTER SIGMA
> <Multi_key> <g> <T> : "Τ" U03C4 # GRREK CAPITAL LETTER TAU
> <Multi_key> <g> <U> : "Î¥" U03C5 # GRREK CAPITAL LETTER UPSILON
> <Multi_key> <g> <V> : "Φ" U03C6 # GRREK CAPITAL LETTER PHI
> <Multi_key> <g> <X> : "Χ" U03C7 # GRREK CAPITAL LETTER CHI
> <Multi_key> <g> <Y> : "Ψ" U03C8 # GRREK CAPITAL LETTER PSI
> <Multi_key> <g> <W> : "Ω" U03C9 # GRREK CAPITAL LETTER OMEGA
>
> --8<---------------cut here---------------end--------------->8---
>

Many Thanks for that. Savely saved away now.

>
>> I am surprised noone has created an emulation of
>> the Windows method.
>
> Gnome has a "Shift with numpad keys works as in MS Windows" option in
> its keyboard settings. But IMHO using a compose key is much more
> userfriendly than having to learn all this character codes.
>Plus you can tailor it to your exact needs.

Many thanks for you help.

Cheers

Ian

Hi Ian,

Ian Thompson-Bell wrote:

> When composing an email I sometimes want to include special characters
> like beat and mu in the text. How can I do this?

Try "abcTajpu", a TBird extension that should do what you want.

https://addons.update.mozilla.org/en...bird/addon/459

Good Luck!
Tim

Ian Thompson-Bell wrote:
> When composing an email I sometimes want to include special characters
> like beat and mu in the text. How can I do this?
>
> Cheers
>
> Ian

It doesn't look like many people answered what you want. You should put
it in using the unicode character. First check out the following url:
http://www.unicode.org/charts/ (Or to go to the greek directly
http://www.unicode.org/charts/PDF/U0370.pdf ) It will give you a list of
codes with the unicode equivalent. Then hold CTRL+SHIFT+U, which will
bring up a text block that has an underlined u. Type in the unicode
equivalent (Beta is 03B2 & Mu is 03BC), which will produce the
following: β μ.

Hope that helps.

"Justin E. Miller" wrote:
> Ian Thompson-Bell wrote:
>> It doesn't look like many people answered what you want. You
> should put it in using the unicode character. First check out the
> following url: [..] (Or to go to the
> greek directly [..] ) It
> will give you a list of codes with the unicode equivalent. Then
> hold CTRL+SHIFT+U, which will bring up a text block that has an
> underlined u. Type in the unicode equivalent (Beta is 03B2 & Mu
> is 03BC), which will produce the following: β μ.
>
> Hope that helps.

It won't. There is a header, called Content-Type. Most readers
understand some set of values there, and will adjust. Without
proper values you are just generating garbage. On my machine those
values show up as a capital I covered with a ^ and followed by a
power of two or a small 1/4.

On 2008-03-23, CBFalconer <cbfalconer> wrote:
>> It won't. There is a header, called Content-Type. Most readers
> understand some set of values there, and will adjust. Without
> proper values you are just generating garbage. On my machine those
> values show up as a capital I covered with a ^ and followed by a
> power of two or a small 1/4.
>
Maybe something is wrong with your news server or client.
I saw a proper Content-Type line and the greek characters.

Bill Marcum wrote:
> CBFalconer <cbfalconer> wrote:
>
>> It won't. There is a header, called Content-Type. Most readers
>> understand some set of values there, and will adjust. Without
>> proper values you are just generating garbage. On my machine those
>> values show up as a capital I covered with a ^ and followed by a
>> power of two or a small 1/4.
>
> Maybe something is wrong with your news server or client.
> I saw a proper Content-Type line and the greek characters.
>
>>> Content-Type: text/plain; charset=UTF-8; format=flowed

You didn't quote enough. He was recommending 16 bit character
values. UTF-8 is 8 bit character values.

Hi there


CBFalconer wrote:

> You didn't quote enough. He was recommending 16 bit character
> values. UTF-8 is 8 bit character values.

Actually, UTF-8 are sequences of 1 to 6 bytes (man utf8);
Conversions between 16 or 32 bit wide characters and UTF-8 are done on
the fly. '03B2' is converted to the bytes 'ce' and 'b2', '03BC' is
converted to 'ce' and 'bc', '20AC' (Euro sign) is converted to the bytes
'e2', '82' and 'ac', etc.

Windows uses 16 bit wide characters internally (most GUIs do) and should
not have any problems with UTF-8 texts or file names.
At the club we have UTF-8 as default charset on the Linux servers
(running Samba) and Windows Thunderbird email clients. File names with
non ASCII chars (E.G. Cyrillic) are legible on both systems. The same
applies to mail.

I suspect you told Thunderbird to convert to ISO-8859-1, which of course
won't work.
I suggest you use UTF-8 as your default charset.


Regards,
Rob

Hi there


Justin E. Miller wrote:

> It doesn't look like many people answered what you want. You should put
> it in using the unicode character. First check out the following url:
> [..] (Or to go to the greek directly
> [..] ) It will give you a list of
> codes with the unicode equivalent. Then hold CTRL+SHIFT+U, which will
> bring up a text block that has an underlined u. Type in the unicode
> equivalent (Beta is 03B2 & Mu is 03BC), which will produce the
> following: β μ.

A neat trick

> Hope that helps.


Regards,
Rob

Rob van der Putten wrote:
> CBFalconer wrote:
>
>> You didn't quote enough. He was recommending 16 bit character
>> values. UTF-8 is 8 bit character values.
>
> Actually, UTF-8 are sequences of 1 to 6 bytes (man utf8);
> Conversions between 16 or 32 bit wide characters and UTF-8 are
> done on the fly. '03B2' is converted to the bytes 'ce' and 'b2',
> '03BC' is converted to 'ce' and 'bc', '20AC' (Euro sign) is
> converted to the bytes 'e2', '82' and 'ac', etc.

I guess I went off half-cocked. Thanks for the correction.