On Tue, 15 Jan 2008 00:41:40 -0000, "Bill Wright"
 wrote:

>
>"Paul Martin"  wrote in message 
>news:slrnfon4hb.961.pm@thinkpad.nowster.org.uk...
>> In article ,
>> Bill Wright wrote:
>>
>>> "Paul Martin"  wrote in message
>>> news:slrnfoh3f9.pqo.pm@thinkpad.nowster.org.uk...
>>>> However some people have broken newsreaders which don't say which, of
>>>> the many ways of encoding a pound sign there are, they're using. Yours
>>>> falls in that camp.
>>
>>> What do you mean 'broken'? It isn't broken. Do you mean inherently
>>> defective?
>>
>> I do. Specifically, I mean that the software breaks (or ignores) the
>> standards.
>
>Ah, but whose standards? If anarchy rules, as you seem to be saying, then 
>there aren't any standards.
>
>Bill 
>

Pound signs - £ - seem topresent fine in Agent.

May I humbly suggest you try Agent in preference to outlook Express?

-- 
Cheers

Peter

"PeterT"  wrote in message 
news:udrso3d29fb335mg96op0ahjlqn6v8upuq@4ax.com...
> Pound signs - £ - seem topresent fine in Agent.
>
> May I humbly suggest you try Agent in preference to outlook Express?

I get £ signs with no problem. It's these people with weird non-Microsoft 
software who seem to have problems.

Bill

Bill Wright wrote:

> "PeterT"  wrote
>
>> Pound signs - £ - seem topresent fine in Agent. May I humbly suggest 
>> you try Agent in preference to outlook Express?

> I get £ signs with no problem. It's these people with weird
> non-Microsoft software who seem to have problems.

Outlook Express has fairly extensive character encoding support. It can 
decode and display almost any character encoding in existence and use on 
the present-day Usenet.

However, for this to work, the posts written by other people must 
include a MIME "Content-type" header that will correctly identify the 
character encoding they're using. (This is generally the case with _all_ 
newsreader programs that support variable character encodings - not just 
OE. The character encoding used in a post must be explicitly stated on 
the header lines of that post; the program can't just guess it.)

An OE user can verify that the Content-Type header is present by 
pressing Ctrl+F3. This keyboard shortcut will display the message 
"source" in a separate window, in "raw" format, with all the header 
lines included. Among those lines, you should be able to find something 
like this:

 Content-Type: text/plain;
  charset="iso-8859-1"

In this sample, the character encoding is set to ISO 8859-1, which is 
one of the very commonly used 8-bit character encodings. You should be 
able to find that kind of header line on my messages, for example. (The 
actual character encoding may vary by the poster, of course, but for the 
Westerners, it is most commonly a member of the ISO 8859 character 
encoding family, Windows 1252, or UTF-8.)

* * *

Curiously, if all settings are left at their defaults, Outlook Express 
does not send these "Content-type" headers itself!

For example, your messages don't have them, which will cause problems if 
you try to use any characters that fall outside the ASCII character 
repertoaire. (This includes the pound sign, which may or may not show up 
on other people's screens as the character encoding is not correctly 
advertised in the headers.)

That can be easily fixed, though:

Just go to...

 "Tools"
     "Options..."
         "Send"
             "News sending format" / "Plain text settings"

...from the pull-down menus of the main OE window, and select the "MIME" 
radiobutton in the "Message format" box.

You should also make sure that the option named "Encode text using" is 
set to "None" or "Quoted-Printable", and not to "Base-64".

After having made these changes, your posts, too, should have proper 
MIME "Content-type" headers with the charset parameter on them, 
indicating the character encoding you're using. This allows you to use, 
for example, all the fancy characters in the ISO 8859-1 character set 
with a relative safety - or even the entire Unicode character 
repertoaire, if you set OE to use the UTF-8 character encoding for your 
posts.

The actual character encoding can be set on a per-message basis from the 
pull-down menus of OE's message editor window (See "Format" -> 
"Encoding" there.)

It is also possible to set a default character encoding. This can be 
done by going to OE's main Window and navigating to...

 "Tools"
     "Options..."
         "Send"
             "Sending" / "International Settings..."
                 "Default encoding"

I'd set the default encoding either to "Western Europe (ISO)" or 
"UTF-8".

The former is just another name for the "ISO 8859-1" or "ISO Latin-1" 
character encoding, which is probably the most widely supported 8-bit 
character encoding in existence, and allows typing English - including 
the pound sign - and most of the Western and Northern European 
languages.

The latter, on the other hand, allows typing even Chinese, Arabic, and 
other languages with non-Western scripts.

More information on ISO 8859-1 is available here...

 <http://en.wikipedia.org/wiki/ISO/IEC_8859-1>

(scroll down the page and you'll see a complete table of the allowed 
characters)

...and on Unicode / UTF-8, here:

 <http://en.wikipedia.org/wiki/Unicode>
 <http://en.wikipedia.org/wiki/UTF-8>

* * *

Regardless of the character set that is selected in use in the menus, OE 
allows typing all kinds of characters in its message editor window, 
without any filtering or restrictions. Only when you click the "Send" 
button will it warn you if you've used characters that fall outside of 
the selected encoding's repertoaire. If you get such a warning, you can 
Cancel the window, go back and remove those offending characters, or go 
to the "Format" -> "Encoding" menu and switch to another character 
encoding that includes them.

You're not likely to encounter this problem too often with ISO 8859-1 
(or "Western (ISO)", as OE wants to call it) as it already includes most 
of the characters one would need in Western context, but if you do, 
switching the encoding to "Unicode (UTF-8)" should help, as Unicode is 
the "ultimate" caracter set/encoding whose goal is to support all the 
characters, symbols and scripts in the entire world (within reasonable 
limits.)

-- 
znark

In article <TECjj.282044$fm3.210464@reader1.news.saunalahti.fi>,
	Jukka Aho wrote:

> After having made these changes, your posts, too, should have proper 
> MIME "Content-type" headers with the charset parameter on them, 
> indicating the character encoding you're using. This allows you to use, 
> for example, all the fancy characters in the ISO 8859-1 character set 
> with a relative safety - or even the entire Unicode character 
> repertoaire, if you set OE to use the UTF-8 character encoding for your 
> posts.

It's worth noting that ISO 8859-1 doesn't contain the Euro symbol. It's
ISO 8859-15 that does.

Of course, to turn the whole thing back on itself, you could use the
international standard trigraphs for each currency, viz. GBP, EUR, USD.

-- 
Paul Martin

Paul Martin wrote:

> It's worth noting that ISO 8859-1 doesn't contain the Euro symbol.
> It's ISO 8859-15 that does.

That's true. But I'm a bit divided as to whether to use the ISO 8859-15 
or UTF-8 for representing that symbol.

On one hand, 8859-15 is _almost_ the same as 8859-1, and so won't cause 
too many problems even if someone's newsreader is hardwired for 
8859-1... or Windows 1252, for that matter. (If things won't quite work 
out, the euro sign will then just display as an international currency 
symbol.)

On the other hand, if the program at the receiving end is modern enough 
to include a proper support for 8859-15, it will almost certainly 
support the UTF-8 encoding as well... and UTF-8 might even be the 
better-supported alternative, as the 8859-15 is a somewhat late and 
obscure addition to the ISO Latin family...

> Of course, to turn the whole thing back on itself, you could use the
> international standard trigraphs for each currency, viz. GBP, EUR,
> USD.

Or just type "euro" (which is what I find myself doing most of the 
time.)

-- 
znark