Hello, all

I ask this question here simply because on this machine I cannot read or
write Russian-language FidoNet articles (yet). What is the default
encoding of a Russian FidoNet message, the one assmed in default of a
CHRS cludge? Is it koi8-r or 866?

---
* Origin: nntp://news.fidonet.fi (2:221/6.0)

Hi Anton,

16.09.2024 5:19:16, Anton Shepelev
<[email protected]> wrote:

I ask this question here simply because on this machine I cannot read
or write Russian-language FidoNet articles (yet). What is the default encoding of a Russian FidoNet message, the one assmed in default of a
CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software.
If you're using an NNTP client under Linux it should be KOI8-R.

For tin, add/edit the line in tinrc as follows:

mm_network_charset=koi8-r


--
"Imagine if every Thursday your shoes exploded if you tied them
the usual way. This happens to us all the time with computers,
and nobody thinks of complaining." -- Jeff Raskin.
--- InterSquish NNTP Server/FTN Gate
* Origin: www.wfido.ru (2:5023/24.4222)

Hi, Gleb Hlebov!
I read your message from 16.09.2024 11:42

I ask this question here simply because on this machine I cannot
read or write Russian-language FidoNet articles (yet). What is the
default encoding of a Russian FidoNet message, the one assmed in
default of a CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software. If you're using an NNTP
client under Linux it should be KOI8-R.

For tin, add/edit the line in tinrc as follows:

mm_network_charset=koi8-r

Modern and even not very modern NNTP clients can also be tuned for UTF-8. In this case you must use a FIDO gate with UTF-8 support, like this for instance, which I use for many years (Thunderbird mail client):

news.dewy.ru

PS: You must add a Russian keyboard of course in your operating system.

Bye, Gleb!
Alexander Koryagin
english_tutor 2024

---
* Origin: news://news.fidonet.fi (2:221/6.0)

Hi Alexander,

Mon 16 Sep 2024 at 12:18, you wrote to me:

What is the default encoding of a Russian FidoNet message, the one
assmed in default of a CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software. If you're using an NNTP
client under Linux it should be KOI8-R.

Modern and even not very modern NNTP clients can also be tuned for
UTF-8. In this case you must use a FIDO gate with UTF-8 support, like
this for instance, which I use for many years (Thunderbird mail
client): news.dewy.ru

Sure, but utf-8 makes little sense as long as Russian Fidonet is stuck with cp866. And it's fine this way. :-)


... End of message, Alexander. Confused? Me too! :-)
--- A Damned Hobbyist+ 1.1.5
* Origin: Microstuff, Inc. (2:5023/24.4222)

Hi, Gleb Hlebov!
I read your message from 17.09.2024 06:18

What is the default encoding of a Russian FidoNet message, the
one assmed in default of a CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software. If you're using an NNTP
client under Linux it should be KOI8-R.

Modern and even not very modern NNTP clients can also be tuned for
UTF-8. In this case you must use a FIDO gate with UTF-8 support,
like this for instance, which I use for many years (Thunderbird
mail client): news.dewy.ru

Sure, but utf-8 makes little sense as long as Russian Fidonet is
stuck with cp866. And it's fine this way. :-)

Everything is OK with cp866 at news.dewy.ru. If I send my UTF-8 message (using any version Thunderbird) to this gate it gates it to FIDONET in cp866. And vice versa.

Bye, Gleb!
Alexander Koryagin
english_tutor 2024

---
* Origin: news://news.fidonet.fi (2:221/6.0)

Gale Hlebov <[email protected].2> wrote:

<[email protected]> wrote:

I ask this question here simply because on this machine I cannot read
or write Russian-language FidoNet articles (yet). What is the default
encoding of a Russian FidoNet message, the one assmed in default of a
CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software.
If you're using an NNTP client under Linux it should be KOI8-R.

How can it be determined by the OS and protocol? I thought the required encoding depended on the settings of the NNTP gateway...

For tin, add/edit the line in tinrc as follows:

mm_network_charset=koi8-r

Thank you. This options lets me read Russian Fidonet echoes on this
server, but I still cannot post. Although I made sure my editor (Vim)
saves the files in koi8-r (set fenc=koi8-r), Tin warns me that I am
posting in UTF-8 and replaces the Russian characters with question
marks...

---
* Origin: nntp://news.fidonet.fi (2:221/6.0)

Alexander Koryagin <[email protected].2> wrote:

Modern and even not very modern NNTP clients can also be tuned for
UTF-8. In this case you must use a FIDO gate with UTF-8 support, like
this for instance, which I use for many years (Thunderbird mail client):

news.dewy.ru

Thanks. I was asking about the general-purpose NNTP servers, such as paganini.bofh.team, that servere Russian FidoNet echoes without a proper Content-Type: header.

---
* Origin: news://news.fidonet.fi (2:221/6.0)

Hi Anton,

Tue 17 Sep 2024 at 22:29, you wrote to me:

What is the default encoding of a Russian FidoNet message, the one
assmed in default of a CHRS cludge? Is it koi8-r or 866?

It's set to 866 in its "native" software.
If you're using an NNTP client under Linux it should be KOI8-R.

How can it be determined by the OS and protocol? I thought the
required encoding depended on the settings of the NNTP gateway...

You're correct, it doesn't depend on the user's OS. I just saw that you were using tin to post the message.

For tin, add/edit the line in tinrc as follows:
mm_network_charset=koi8-r

Thank you. This options lets me read Russian Fidonet echoes on this server, but I still cannot post. Although I made sure my editor (Vim) saves the files in koi8-r (set fenc=koi8-r), Tin warns me that I am posting in UTF-8 and replaces the Russian characters with question marks...

Weird, I don't remember having those issues with vim/tin and charset (mis)conversion. I didn't modify vim's config for koi8-r either. Btw, I also have this line in tinrc:

post_mime_encoding=base64

Not sure if this is the reason but you could try.


... End of message, Anton. Confused? Me too! :-)
--- A Damned Hobbyist+ 1.1.5
* Origin: Microstuff, Inc. (2:5023/24.4222)

Hi Anton,

Wed 18 Sep 2024 at 09:37, I <2:5023/24.4222> wrote to you:

mm_network_charset=koi8-r
post_mime_encoding=base64

Not sure if this is the reason but you could try.

I checked my tin setup on a linux machine again and posted a msg in FIDOTEST, if you can read it alright then those settings do make sense. :-)

Also, my $LANG is "ru_RU.UTF-8", and I DID NOT modify vim's configs.


... End of message, Anton. Confused? Me too! :-)
--- A Damned Hobbyist+ 1.1.5
* Origin: Microstuff, Inc. (2:5023/24.4222)

Gleb Hlebov <[email protected].2> wrote:

mm_network_charset=koi8-r
post_mime_encoding=base64

Not sure if this is the reason but you could try.

I checked my tin setup on a linux machine again and posted a msg in
FIDOTEST, if you can read it alright then those settings do make sense. :-)

Also, my $LANG is "ru_RU.UTF-8", and I DID NOT modify vim's configs.

No luck with this setup. When I try to post a message, tin (2.6.3)
says:

ERR: Posting failed (Posting failed (Unsupported charset "KOI8-RU"))

which is actually strange, because encoding option is a drop-down
list with predefined values, and of course it should be `koi8-r'
rather than `KOI8-RU'.

# Anton exists tin, fixes his tinrc by hand, runs tin again

Indeed! It is a bug in my version of tin:
the menu has KOI8-RU instead of koi8-r .
Thank you!

---
* Origin: news://news.fidonet.fi (2:221/6.0)

Hi Anton,

Thu 19 Sep 2024 at 19:47, you wrote to me:

# Anton exists tin, fixes his tinrc by hand, runs tin again
Indeed! It is a bug in my version of tin:
the menu has KOI8-RU instead of koi8-r .
Thank you!

It's the Tao of unix: you can never set anything up through "drop-down menus". :-)

You can also check this setting which allows FTN-style quoting in replies (recommended for Fidonet), with leading initials:

======
# Characters used in quoting to followups and replies.
# '_' is replaced by ' ', %I is replaced by author's initials.
quote_chars=_%I >
======


... End of message, Anton. Confused? Me too! :-)
--- A Damned Hobbyist+ 1.1.5
* Origin: Microstuff, Inc. (2:5023/24.4222)

Gleb Hlebov <[email protected].2> wrote:

GH >Hi Anton,
GH >

# Anton exists tin, fixes his tinrc by hand, runs tin again
Indeed! It is a bug in my version of tin:
the menu has KOI8-RU instead of koi8-r .
Thank you!

GH >
GH >It's the Tao of unix: you can never set anything up through "drop-down
GH >menus". :-)

Turned out there /exists/ a KOI8-RU as well!

GH >You can also check this setting which allows FTN-style quoting in
GH >replies (recommended for Fidonet), with leading initials:
GH >
GH >======
GH ># Characters used in quoting to followups and replies.
GH ># '_' is replaced by ' ', %I is replaced by author's initials.
GH >quote_chars=_%I >
GH >======

Yes, and works good, I can try to choose the right refix
via the attributes files, although it works by group rather than
by server...

---
* Origin: news://news.fidonet.fi (2:221/6.0)

Gleb Hlebov <[email protected].2> wrote:

Hi Anton,

Fri 20 Sep 2024, 20:18, you -> me:

GH >======
GH ># Characters used in quoting to followups and replies.
GH ># '_' is replaced by ' ', %I is replaced by author's initials.
GH >quote_chars=_%I >
GH >======
Yes, and works good, I can try to choose the right refix
via the attributes files, although it works by group rather than
by server...

Oops, there seems to be a typo by me, while I tried to make it readable
(I couldn't use actual underscore pairs because newsreaders interpret
them as "color code" modifiers).

The correct prefix is:
[underscore]%I>[underscore]

Why? The spaces work quite well as this post demonstrates.

---
* Origin: news://news.fidonet.fi (2:221/6.0)