I don't normally crosspost but I'm making an exception.

I've notice that both Thunderbird and Seamonkey have been quoting
articles in Usenet followups and have been substituting non-breaking
space characters for ASCII space characters in quoted text. This is
highly undesireable behavior. Not every followup does this. What setting
controls this?

Followups to text newsgroups must always be in plain text. A
non-breaking space is a typographical character intended to be used in very
limited circumstances and is never used in plain text communication.
Generally it doesn't belong on the Web either. It is absolutely not
intended to be substituted for the ASCII space character.

Firefox newsgroup
non-breaking space characters substitution for ASCII space characters in
quote

root article
Message-ID: <uhu2m6$1n5cs$6@dont-email.me>

followup
Seamonkey newsreader
Message-ID: <uhul27$1qojn$1@dont-email.me>

Thunderbird newsgroup
double spaces between sentences changed from two ASCII spaces to
non-breaking space with ASCII space

precursor article
Message-ID: <ui01gp$25phj$2@dont-email.me>

followup
Thunderbird newsreader
Message-ID: <kqhgqgFdm1dU4@mid.individual.net>

Adam H. Kerman wrote:

> followup
> Thunderbird newsreader
> Message-ID:<kqhgqgFdm1dU4@mid.individual.net>

I copied the body of that one using al.howardknight and pasted it into a
hex editor, I don't see any 0xA0 bytes ...

Andy Burns <usenet@andyburns.uk> wrote:
>Adam H. Kerman wrote:

>>followup
>>Thunderbird newsreader
>>Message-ID:<kqhgqgFdm1dU4@mid.individual.net>

>I copied the body of that one using al.howardknight and pasted it into a
>hex editor, I don't see any 0xA0 bytes ...

Why would you see them on the Web if UTF-8 is being displayed? A
nonbreaking space doesn't display when UTF-8 is displayed. Look at
the article on Usenet and force a character set mismatch by displaying
ASCII or ISO-8859-1. Then you'll see them.

Adam H. Kerman wrote:

> Andy Burns wrote:
>
>> I copied the body of that one using al.howardknight and pasted it into a
>> hex editor, I don't see any 0xA0 bytes ...
>
> Why would you see them on the Web if UTF-8 is being displayed?

I'd expect to see them, maybe "see" is the wrong word for a space or an
NBSP, but unless howardknight strips them, I'd expect to be able to tell
they were there

> A nonbreaking space doesn't display when UTF-8 is displayed. Look at
> the article on Usenet and force a character set mismatch by
> displaying ASCII or ISO-8859-1. Then you'll see them.

If I reply to the same article from Big Al that Carlos did then my
thunderbird also seems to somehow convert spaces following a period into
0xA0 characters

"Adam H. Kerman" <ahk@chinet.com> wrote:

> I don't normally crosspost but I'm making an exception.
>
> I've notice that both Thunderbird and Seamonkey have been quoting
> articles in Usenet followups and have been substituting non-breaking
> space characters for ASCII space characters in quoted text. This is
> highly undesireable behavior. Not every followup does this. What setting
> controls this?
>
> Followups to text newsgroups must always be in plain text. A
> non-breaking space is a typographical character intended to be used in very
> limited circumstances and is never used in plain text communication.
> Generally it doesn't belong on the Web either. It is absolutely not
> intended to be substituted for the ASCII space character.
>
> Firefox newsgroup
> non-breaking space characters substitution for ASCII space characters in
> quote
>
> root article
> Message-ID: <uhu2m6$1n5cs$6@dont-email.me>
>
> followup
> Seamonkey newsreader
> Message-ID: <uhul27$1qojn$1@dont-email.me>
>
> Thunderbird newsgroup
> double spaces between sentences changed from two ASCII spaces to
> non-breaking space with ASCII space
>
> precursor article
> Message-ID: <ui01gp$25phj$2@dont-email.me>
>
> followup
> Thunderbird newsreader
> Message-ID: <kqhgqgFdm1dU4@mid.individual.net>

I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
thing, not a plain text thing.

https://bugzilla.mozilla.org/show_bug.cgi?id=532712
(the discussion petered out 2 years ago)

Seems that discussion is about how to insert the HTML nbsp character,
and also about nbsp in e-mail composition, and not for Usenet messsages.
Since Usenet messages should not be HTML formatted, seems plausible and
even correct that nbsp get changed to sp.

The bug ticket mentions some languages where nbsp is required for proper
syntax, like in French, but again the discussion is how to use nbsp in
HTML (and in e-mail). In HTML parsing, multiple spaces are seen as 1
space, so nbsp gets used to retain the spacing of multiple spaces. But
that's in HTML, not in plain text where nbsp isn't defined despite using
UTF-8, and gets converted to sp.

The ticket (comment 20) indicates there was a change, but not until
Tbird 91, and the articles you mention show the users posted using Tbird
91. The code fix might've been in the next Tbird version. However,
that ticket kept getting closed, and then reopened. A similar bug, but
again discussing HTML where nbsp can be defined:

https://bugzilla.mozilla.org/show_bug.cgi?id=1457613

If the Tbird editor is configured to use plain text (not HTML), I don't
see Tbird should ever insert a nbsp character in a plain text message.
Does cause a problem with some language syntax, like French, but I was
not aware that plain text format was ever language dependent.

Andy Burns <usenet@andyburns.uk> wrote:
>Adam H. Kerman wrote:
>>Andy Burns wrote:

>>>I copied the body of that one using al.howardknight and pasted it into a
>>>hex editor, I don't see any 0xA0 bytes ...

>>Why would you see them on the Web if UTF-8 is being displayed?

>I'd expect to see them, maybe "see" is the wrong word for a space or an
>NBSP, but unless howardknight strips them, I'd expect to be able to tell
>they were there

Not with a Web browser displaying UTF-8 characters. Firefox has taken
away the ability to change character set display options.

To see the bad characters, you'd have to display ASCII or ISO--8859-1
character set.

>>A nonbreaking space doesn't display when UTF-8 is displayed. Look at
>>the article on Usenet and force a character set mismatch by
>>displaying ASCII or ISO-8859-1. Then you'll see them.

>If I reply to the same article from Big Al that Carlos did then my
>thunderbird also seems to somehow convert spaces following a period into
>0xA0 characters

Ok

What setting turns that undesireable behavior off?

VanguardLH wrote:

> I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
> thing, not a plain text thing.

0xFF isn't an nbsp, it's a y with umlaut

0xA0 in extended ascii, is a nbsp, or in html the entity &nbsp; or
&#160; can be used

On Thu, 2 Nov 2023 14:37:58 -0000 (UTC), Adam H. Kerman wrote:

> Followups to text newsgroups must always be in plain text. A
> non-breaking space is a typographical character intended to be used in very
> limited circumstances and is never used in plain text communication.
> Generally it doesn't belong on the Web either. It is absolutely not
> intended to be substituted for the ASCII space character.

When I post plain text diagrams (like the example below), I _always_
replace ASCII spaces in the diagram with non-breaking spaces so that
word-wrap will not damage the diagram. This is a legitimate use of
the non-breaking space in plain text posts.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[Diagram best viewed in a fixed-pitch font]

   |===================== YOUR COMPUTER =====================|        |============ THE INTERNET ============|

    +---------+  OUTBOUND MAIL SERVER         +---------+  FORWARD TO                  +--------------------+
    ¦         ¦------------------------------>¦         ¦----------------------------->¦  SMTP mail server  ¦
    ¦  Email  ¦  localhost port 8025, no SSL  ¦         ¦  smtp.xyz.com port 465, SSL  +--------------------+
    ¦  client ¦                               ¦ STunnel ¦
    ¦         ¦  INBOUND MAIL SERVER          ¦         ¦  FORWARD TO                  +--------------------+
    ¦         ¦------------------------------>¦         ¦----------------------------->¦  POP3 mail server  ¦
    +---------+  localhost port 8110, no SSL  +---------+  pop3.xyz.com port 995, SSL  +--------------------+

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--
Kind regards
Ralph Fox

“Wyt ant wysdom is god warysoun,” quoþ Hendyng.

Ralph Fox wrote:

> When I post plain text diagrams (like the example below), I always
> replace ASCII spaces in the diagram with non-breaking spaces so that
> word-wrap will not damage the diagram. This is a legitimate use of
> the non-breaking space in plain text posts.

providing it's 8 bit extended ascii, which your
Content-Transfer-Encoding: header says it is, some may claim you
shouldn't use that on usenet.

VanguardLH <V@nguard.LH> wrote:
>"Adam H. Kerman" <ahk@chinet.com> wrote:

>>I don't normally crosspost but I'm making an exception.

>>I've notice that both Thunderbird and Seamonkey have been quoting
>>articles in Usenet followups and have been substituting non-breaking
>>space characters for ASCII space characters in quoted text. This is
>>highly undesireable behavior. Not every followup does this. What setting
>>controls this?

>>Followups to text newsgroups must always be in plain text. A
>>non-breaking space is a typographical character intended to be used in very
>>limited circumstances and is never used in plain text communication.
>>Generally it doesn't belong on the Web either. It is absolutely not
>>intended to be substituted for the ASCII space character.

>>Firefox newsgroup
>>non-breaking space characters substitution for ASCII space characters in
>>quote

>>root article
>>Message-ID: <uhu2m6$1n5cs$6@dont-email.me>

>>followup
>>Seamonkey newsreader
>>Message-ID: <uhul27$1qojn$1@dont-email.me>

>>Thunderbird newsgroup
>>double spaces between sentences changed from two ASCII spaces to
>>non-breaking space with ASCII space

>>precursor article
>>Message-ID: <ui01gp$25phj$2@dont-email.me>

>>followup
>>Thunderbird newsreader
>>Message-ID: <kqhgqgFdm1dU4@mid.individual.net>

>I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>thing, not a plain text thing.

They are not plain text. Yes, I know it's in ISO-8859-1. Hm. I'm not
sure why it was visible to me when I set the terminal to ISO-8859-1.

They're a typesetting thing for very special circumstances that, for
the most part, shouldn't be used in HTML. It's an instruction to word
wrap at the previous ASCII whitespace character. It HTML, it's used to
force display of consecutive whitespace not to collapse into a single
whitespace.

If used in HTML, it's best to use the entity &nbsp;

If you want to avoid a line break within a two-word phrase you might use
it if you are setting type on a page.

WhiteNBSPHouse prevents a line from breaking between White and House.

>https://bugzilla.mozilla.org/show_bug.cgi?id=532712
>(the discussion petered out 2 years ago)

>Seems that discussion is about how to insert the HTML nbsp character,
>and also about nbsp in e-mail composition, and not for Usenet messsages.
>Since Usenet messages should not be HTML formatted, seems plausible and
>even correct that nbsp get changed to sp.

Can the composer work in HTML and does it accept entities? I don't do
things like that.

The bug report you summarized was interesting.

The French thing goes back to an error made in ignorance by the delegate
from France (who was not a typographer) at the first committee meeting
defining ISO-8859-1, falsely claiming that certain French characters
weren't characters at all but ligatures. Because they didn't get their
own code point, they need to stay together. There is another entity
called "don't break here" that performs that function. I'm not sure what
an example of the use of no-break space in French typography as that
wouldn't create a ligature.

It's on the Wikipedia page.

Andy Burns <usenet@andyburns.uk> wrote:
>VanguardLH wrote:

>>I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>>thing, not a plain text thing.

>0xFF isn't an nbsp, it's a y with umlaut

>0xA0 in extended ascii, is a nbsp, or in html the entity &nbsp; or
>&#160; can be used

Yes, it's in ISO-8859-1 but it displays for me when I've set the
terminal to ISO-8859-1 and now I don't understand why.

This portion of UTF-8 is still displayed with a two-byte sequence? I can
never remember.

Andy Burns <usenet@andyburns.uk> wrote:

> VanguardLH wrote:
>
>> I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>> thing, not a plain text thing.
>
> 0xFF isn't an nbsp, it's a y with umlaut

I thought Alt+numpad255 on the keyboard gave the nbsp character. It
does in Notepad.

Ralph Fox <-rf-nz-@-.invalid> wrote:

> On Thu, 2 Nov 2023 14:37:58 -0000 (UTC), Adam H. Kerman wrote:
>
>> Followups to text newsgroups must always be in plain text. A
>> non-breaking space is a typographical character intended to be used in very
>> limited circumstances and is never used in plain text communication.
>> Generally it doesn't belong on the Web either. It is absolutely not
>> intended to be substituted for the ASCII space character.
>
> When I post plain text diagrams (like the example below), I _always_
> replace ASCII spaces in the diagram with non-breaking spaces so that
> word-wrap will not damage the diagram. This is a legitimate use of
> the non-breaking space in plain text posts.
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> [Diagram best viewed in a fixed-pitch font]
>
>    |===================== YOUR COMPUTER =====================|        |============ THE INTERNET ============|
>
>     +---------+  OUTBOUND MAIL SERVER         +---------+  FORWARD TO                  +--------------------+
>     ¦         ¦------------------------------>¦         ¦----------------------------->¦  SMTP mail server  ¦
>     ¦  Email  ¦  localhost port 8025, no SSL  ¦         ¦  smtp.xyz.com port 465, SSL  +--------------------+
>     ¦  client ¦                               ¦ STunnel ¦
>     ¦         ¦  INBOUND MAIL SERVER          ¦         ¦  FORWARD TO                  +--------------------+
>     ¦         ¦------------------------------>¦         ¦----------------------------->¦  POP3 mail server  ¦
>     +---------+  localhost port 8110, no SSL  +---------+  pop3.xyz.com port 995, SSL  +--------------------+
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

But you cannot guarantee that the recipient's/reader's client will not
force line wrap. In my NNTP client, I can select the line(s) where I do
not want line-wrap. A placeholder character is added in compose mode,
but not included in the sent message. Those lines won't get wrapped by
my client, but someone else may wrap lines before the fixed line break
at the end of the line.

In the examples Adam mentioned by MID, they have:

Content-Type: text/plain; charset=UTF-8; format=flowed

I thought format=flowed was used with quoted-printable which adds those
=20 soft breaks at the ends of lines.

Adam's posts don't have the Content-Type header. Mine do, but no
format=flowed attribute. Andy does have format=flowed, but I don't see
a formatting issue in my ancient NNTP client (40tude Dialog), and don't
have Tbird to check. You (Ralph) do not have format=flowed. The
examples Adam mentioned do have format=flowed.

If Content-Type is missing, or its format attribute is missing, or
format=fixed, the line lengths are fixed. No soft breaks. Despite
that, some clients will wrap text based on the window size instead of
showing the text running off the page and having to scroll to see it.

I used fixed length lines, and also have them under 72 chars long to
account to indentation with quoting chars for replies. That's in my
compose window. When reading posts by others, quite often, especially
after many replies, the lines go off the page, because their fixed line
breaks, so I have to scroll. Clients that wrap those often end up with
jerky results, like Outlook Express did. RFC 2646 4.1, says when
format=flowed to have lines physically less than 80 chars in length.

In truly plain-text messages, I don't see nbsp is valid. It is a
typesetting character, just like in your example, to compensate for
clients that will wrap fixed lines instead of retaining them as one
physical long line. Too much of what can be done with e-mail has leaked
over into NNTP messages. Or, conversely, too much of Usenet has been
molded to become e-mail-like messages. Even RFC 2646 is e-mail
oriented.

3. The Problem
The Text/Plain media type is the lowest common denominator of
Internet email, ...

RFC 3646 obsoletes RFC 2646, but does not alter comments about
format=flowed, but 3646 merely adds inclusion of DelSp.

VanguardLH <V@nguard.LH> wrote:
>Andy Burns <usenet@andyburns.uk> wrote:
>
>> VanguardLH wrote:
>>
>>> I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>>> thing, not a plain text thing.
>>
>> 0xFF isn't an nbsp, it's a y with umlaut
>
>I thought Alt+numpad255 on the keyboard gave the nbsp character. It
>does in Notepad.

Decimal for 00a0 is 160.

VanguardLH <V@nguard.LH> wrote:

>. . .

>In the examples Adam mentioned by MID, they have:

>Content-Type: text/plain; charset=UTF-8; format=flowed

>I thought format=flowed was used with quoted-printable which adds those
>=20 soft breaks at the ends of lines.

Two different standards. In QP, a soft line break is "= ", note the
trailing space. In format=flowed, it's just a trailing space.

>Adam's posts don't have the Content-Type header.

I'm rebelling against MIME. Actually, I declare the character set when
quoting something legitimate that uses non-ASCII characters. Almost all
the time, I post in ASCII and don't add MIME headers.

>. . .

"Adam H. Kerman" <ahk@chinet.com> wrote:

> VanguardLH <V@nguard.LH> wrote:
>
>>Andy Burns <usenet@andyburns.uk> wrote:
>>
>>> VanguardLH wrote:
>>>
>>>> I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>>>> thing, not a plain text thing.
>>>
>>> 0xFF isn't an nbsp, it's a y with umlaut
>>
>> I thought Alt+numpad255 on the keyboard gave the nbsp character. It
>> does in Notepad.
>
> Decimal for 00a0 is 160.

255 is an extended ASCII (ASCII8) character from the old DOS code pages.
Listed on the wikipedia page you mentioned, and also at
https://www.lookuptables.com/text/extended-ascii-table.

On 2023-11-02, Ralph Fox <-rf-nz-@-.invalid> wrote:
>
> [Diagram best viewed in a fixed-pitch font]
>
> ???|=====================?YOUR?COMPUTER?=====================|????????|============?THE?INTERNET?============|
>
> ????+---------+??OUTBOUND?MAIL?SERVER?????????+---------+??FORWARD?TO??????????????????+--------------------+
> ???????????????------------------------------>???????????----------------------------->???SMTP?mail?server???
> ???????Email?????localhost?port?8025,?no?SSL???????????????smtp.xyz.com?port?465,?SSL??+--------------------+
> ???????client???????????????????????????????????STunnel??
> ?????????????????INBOUND?MAIL?SERVER???????????????????????FORWARD?TO??????????????????+--------------------+
> ???????????????------------------------------>???????????----------------------------->???POP3?mail?server???
> ????+---------+??localhost?port?8110,?no?SSL??+---------+??pop3.xyz.com?port?995,?SSL??+--------------------+
>
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>

Viewed in a fixed-pitch font and it's garbage!!!!

On 11/2/2023 7:37 AM, Adam H. Kerman wrote:
> I don't normally crosspost but I'm making an exception.
>
> I've notice that both Thunderbird and Seamonkey have been quoting
> articles in Usenet followups and have been substituting non-breaking
> space characters for ASCII space characters in quoted text. This is
> highly undesireable behavior. Not every followup does this. What setting
> controls this?
>
> Followups to text newsgroups must always be in plain text. A
> non-breaking space is a typographical character intended to be used in very
> limited circumstances and is never used in plain text communication.
> Generally it doesn't belong on the Web either. It is absolutely not
> intended to be substituted for the ASCII space character.

Windows 7
Thunderbird 52.9.1
Free Hex Editor Neo 7.37.0.8578

I always post in plain-text. I started to reply to your message, copied
the result to a Notepad window, and saved the file. I then opened the
file in Free Hex Editor Neo. At the end of each quoted line I saw 0a,
which is line-feed and not non-breaking space. Non-breaking sapce is
a0, which did not appear at all.

--
David E. Ross
<http://www.rosde.com/>

X (previously known as Twitter) is the place where truth goes to die.

On 11/2/2023 3:33 PM, David E. Ross wrote:
> On 11/2/2023 7:37 AM, Adam H. Kerman wrote:
>> I don't normally crosspost but I'm making an exception.
>>
>> I've notice that both Thunderbird and Seamonkey have been quoting
>> articles in Usenet followups and have been substituting non-breaking
>> space characters for ASCII space characters in quoted text. This is
>> highly undesireable behavior. Not every followup does this. What setting
>> controls this?
>>
>> Followups to text newsgroups must always be in plain text. A
>> non-breaking space is a typographical character intended to be used in very
>> limited circumstances and is never used in plain text communication.
>> Generally it doesn't belong on the Web either. It is absolutely not
>> intended to be substituted for the ASCII space character.
>
> Windows 7
> Thunderbird 52.9.1
> Free Hex Editor Neo 7.37.0.8578
>
> I always post in plain-text. I started to reply to your message, copied
> the result to a Notepad window, and saved the file. I then opened the
> file in Free Hex Editor Neo. At the end of each quoted line I saw 0a,
> which is line-feed and not non-breaking space. Non-breaking sapce is
> a0, which did not appear at all.
>

I tried the same test with an E-mail message. Again, there was no 0a
(non-breaking space, NBSP). Instead, there was 0d0a, which is
carriage-return and line-feed (CR LF). CR LF or CR CR LF was
conventional decades ago for Telex and Teletype equipment and was
carried forth into Internet messages.

The CR CR LF was common because the mechanical printers were very much
like mechanical typewriters. A single CR might not bring the mechanism
back to the beginning of a line, which meant that after LF text might
start typing backwards as the carriage continued to return. Such
printers were designed so that the second CR would not execute until the
first CR completed.

--
David E. Ross
<http://www.rosde.com/>

X (previously known as Twitter) is the place where truth goes to die.

VanguardLH <V@nguard.LH> wrote:
>Adam H. Kerman <ahk@chinet.com> wrote:
>>VanguardLH <V@nguard.LH> wrote:
>>>Andy Burns <usenet@andyburns.uk> wrote:
>>>>VanguardLH wrote:

>>>>>I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>>>>>thing, not a plain text thing.

>>>>0xFF isn't an nbsp, it's a y with umlaut

>>>I thought Alt+numpad255 on the keyboard gave the nbsp character. It
>>>does in Notepad.

>>Decimal for 00a0 is 160.

>255 is an extended ASCII (ASCII8) character from the old DOS code pages.
>Listed on the wikipedia page you mentioned, and also at
>https://www.lookuptables.com/text/extended-ascii-table.

I don't recall these things any more. CP437 (I'd forgotten it was from
the old Wang word processing terminal) largely encodes most of the same
glyphs as ISO-8859-1 (from DEC!) but there are differences.

Yes, you've reminded me that A+NNN code is from this Code page, but to
get ISO-8859-1 extended ASCII codes, it's Alt+NNNN which I think always
begin with "0" and the rest are decimal 128-255.

In your example, it would be Alt+0160, if I recall correctly.

VanguardLH wrote:

> I thought Alt+numpad255 on the keyboard gave the nbsp character. It
> does in Notepad.

My day-to-day machines no longer have numpads (laptops or TKL spacesavers)

I use emoji panel (winkey+fullstop) to type most oddball characters, but
can't see nbsp there

osk.exe is rubbish, even with the numpad enabled, it won't type alt+0nnn
codes

charmap.exe has got about a dozen different spaces

I think there's a model M within reach in the junk room ...

VanguardLH wrote:

> 255 is an extended ASCII (ASCII8) character from the old DOS code pages.
> Listed on the wikipedia page you mentioned, and also at
> https://www.lookuptables.com/text/extended-ascii-table.

Sure, back in DOS days, we'd have "country=uk,44,437" in config.sys and
"keyb uk" in autoexec.bat 0xFF was an nbsp

but now it's iso latin1 or 8859-1 and 0xFF is no longer nbsp, it's now ÿ

Jim Jackson wrote:

> On 2023-11-02, Ralph Fox <-rf-nz-@-.invalid> wrote:
>>
>> [Diagram best viewed in a fixed-pitch font]
>>
>> ???|=====================?YOUR?COMPUTER?=====================|????????|============?THE?INTERNET?============|
>>
>> ????+---------+??OUTBOUND?MAIL?SERVER?????????+---------+??FORWARD?TO??????????????????+--------------------+
>> ???????????????------------------------------>???????????----------------------------->???SMTP?mail?server???
>> ???????Email?????localhost?port?8025,?no?SSL???????????????smtp.xyz.com?port?465,?SSL??+--------------------+
>> ???????client???????????????????????????????????STunnel??
>> ?????????????????INBOUND?MAIL?SERVER???????????????????????FORWARD?TO??????????????????+--------------------+
>> ???????????????------------------------------>???????????----------------------------->???POP3?mail?server???
>> ????+---------+??localhost?port?8110,?no?SSL??+---------+??pop3.xyz.com?port?995,?SSL??+--------------------+
>>
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /EXAMPLE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Viewed in a fixed-pitch font and it's garbage!!!!

Ralph's original looked ok here in TB, consoleas monospace.

On Thu, 2 Nov 2023 18:37:46 +0000, Andy Burns wrote:
> Ralph Fox wrote:
>
>> When I post plain text diagrams (like the example below), I always
>> replace ASCII spaces in the diagram with non-breaking spaces so that
>> word-wrap will not damage the diagram. This is a legitimate use of
>> the non-breaking space in plain text posts.
>
> providing it's 8 bit extended ascii, which your
> Content-Transfer-Encoding: header says it is,

Equally or more important, the Content-Type header has a charset
declaration which tell clients how to interpret the 8-bit bytes in
the message. And the charset is an IANA-registered charset (which
was registered last century).

Usenet has been 8-bit safe for a long time. Email still officially
requires negotiating 8-bit compatibility on each hop of the email
to the next server [ref: RFC1426 -> RFC1652 -> RFC6152].

> some may claim you
> shouldn't use that on usenet.

Those who claim this are windward micturators. Many European
languages with Latin scripts use accented characters which are 8-bit
not 7-bit. And they have their Usenet groups. Other languages have
their own scripts which will use 8-bit bytes on Usenet.

--
Kind regards
Ralph

“Fer from eȝe, fer from herte,” quoþ Hendyng.

"Adam H. Kerman" <ahk@chinet.com> wrote:

> VanguardLH <V@nguard.LH> wrote:
>>Adam H. Kerman <ahk@chinet.com> wrote:
>>>VanguardLH <V@nguard.LH> wrote:
>>>>Andy Burns <usenet@andyburns.uk> wrote:
>>>>>VanguardLH wrote:
>
>>>>>>I though nbsp (non-breaking space) as Alt+255 or U+00A0 was an HTML
>>>>>>thing, not a plain text thing.
>
>>>>>0xFF isn't an nbsp, it's a y with umlaut
>
>>>>I thought Alt+numpad255 on the keyboard gave the nbsp character. It
>>>>does in Notepad.
>
>>>Decimal for 00a0 is 160.
>
>>255 is an extended ASCII (ASCII8) character from the old DOS code pages.
>>Listed on the wikipedia page you mentioned, and also at
>>https://www.lookuptables.com/text/extended-ascii-table.
>
> I don't recall these things any more. CP437 (I'd forgotten it was from
> the old Wang word processing terminal) largely encodes most of the same
> glyphs as ISO-8859-1 (from DEC!) but there are differences.
>
> Yes, you've reminded me that A+NNN code is from this Code page, but to
> get ISO-8859-1 extended ASCII codes, it's Alt+NNNN which I think always
> begin with "0" and the rest are decimal 128-255.
>
> In your example, it would be Alt+0160, if I recall correctly.

Yeah, I remember having to enter a 4-digit numeric code with leading
zero as the alternate input method for extended characters.

I think some programs use Ctrl+Shift+spacebar to enter a nbsp char (not
the HTML &nbsp, but the actual char code for nbsp).

https://support.microsoft.com/en-us/office/keep-text-together-af94e5b8-3a5a-4cb0-9c53-dea56b43d96d

The dialog shown is for Word, or maybe all MS Office components. I've
read Ctrl+Shift+spacebar may have been replaced with
Ctrl+Shift+Alt+spacebar (and Ctrl+Shift+space got repurposed to use Read
Aloud function).

I know about the Alt+255 in many editors or input programs. You
reminded me the leading zero can also be used, as in Alt+0160. I'm not
sure anymore what MS Office, and other software uses, to insert a nbsp.