docs: fix codecs table data

[mr-bronson]

Documentation

ISO-8859-4 is NOT for Baltic languages. (see Wikipedia and spec preview)

Also, use new English spelling of Belarusian. (see Wikipedia and Google Ngram Viewer)

Linked PRs

• gh-123439

[mr-bronson]

Even though your instructions clearly said trivial changes do not require an issue number, your bot insisted that it was "Required". I've obliged that request, even though it should not have been necessary.

But what the heck does signing a CLA have to do with giving your bot unknown permissions to access my GitHub account? Not gonna happen. Your other system is equally unacceptable because it requires reading and agreeing to Adobe terms and privacy policy. I'm making an agreement with PSF, not Adobe! Are you not competent to make a regular web form your own selves?

Not acceptable. You can take my correction (I offer it without strings attached, not that that should even need to be said) or leave it, but your system stinks, and I'm not gonna play ball as is.

I've contributed to numerous other software projects without this sort of crap. Your Developer Guide says "We encourage everyone to contribute to Python". I'll believe it when I see it.

[hugovk]

Please be civil and be aware we have a code of conduct and expect everyone to follow it:

https://github.com/python/.github/blob/d043932e7db760a58b6005436e2eb969077631f5/CODE_OF_CONDUCT.md

The bot says 'No issue # in title or "skip issue" label found'. You may ask for the "skip issue" for trivial issues and someone will apply it if so.

Are you able to download and print a PDF? You then have the option of posting, faxing, or emailing the CLA:

https://www.python.org/psf/contrib/contrib-form/

[mr-bronson]

Please be civil and be aware we have a code of conduct and expect everyone to follow it:

I'm not trying to be offensive toward any particular person. But if I express my opinions about a system or communicate flaws with it, I might expect people to be "receptive to constructive comments and criticism", "Gracefully accepting constructive criticism", and "Showing empathy" toward the issues I had with the system. Plus, your CoC applies to members. I'm not a member, nor do I expect to be made a member. I'm just some guy on the internet offering a suggestion of how to improve your project.

Thank you for pointing out that the form could be filled out via PDF. I saw that. However there are still issues with this, in addition to the hassle factor. The purpose of the form is to get me to agree to license my contributions to PSF. I don't want to "license" any such thing; I don't believe in that IP malarkey. I'm not going to claim any sort of "copyright" to my contributions. As far as I'm concerned, they are "public domain" or "cc0" or whatever silly moniker you want to put on it.

Even if I agreed to "license" my contributions to PSF under one of the two acceptable initial licenses (which I don't, because that perpetuates malarkey I don't believe in), PSF says it can change it to "any other open source license". Well, how broad is that? Suppose someone writes a license that doesn't have provisions disclaiming liability. Do I then become liable? Who knows? Ask a lawyer. Or, rather, ask the judge who ends up adjudicating any particular case (because that's how the system really works; it's not like the law is king - Lex Rex.) As far as I'm concerned, what you get from me is bits and bytes, and what you choose to do with the bits and bytes is your problem.

So, while PSF claims that it encourages everyone to contribute, it is effectively discouraging contributions from:

1. People who don't appreciate, don't understand or don't have the patience for all the legal malarkey.
2. People who object to perpetuating legal frameworks they don't agree with.
3. People who don't want to give their personal data to PSF and Adobe and whoever else these entities choose to share such details with.
4. People who don't want to give closed-source bots access to their GitHub accounts to perform unknown actions.

And, as I'll point out again, these are people who other open source projects are simply happy to have involved without the hassles and requirements PSF is imposing.

[hugovk]

Plus, your CoC applies to members. I'm not a member, nor do I expect to be made a member. I'm just some guy on the internet offering a suggestion of how to improve your project.

It applies to everyone contributing to this repo:

This Code of Conduct applies to the following online spaces:

• ...
• Code repositories, issue trackers, and pull requests made against any Python Software Foundation controlled GitHub organization

...

This Code of Conduct applies to the following people in official Python Software Foundation online spaces:

• ...
• contributors

https://policies.python.org/python.org/code-of-conduct/#psf-online-spaces

4. People who don't want to give closed-source bots access to their GitHub accounts to perform unknown actions.

It's open source: https://github.com/ambv/cla-bot

[serhiy-storchaka]

It is called "North European" in Wikipedia, but Wikipedia is not a primary source. I have not found "North European" in the spec preview. In the Linux manpage it has note "Scandinavian/Baltic languages". On the other page it is described as for "North European languages", although it misses some Greenlandic, Sami and Latvian letters (covered by ISO 8859-10 and ISO 8859-13). At the same page, Belarusian is referred as Byelorussian (the official name was changed after the dissolution of the USSR).

What are other official or semi-official sources about ISO 8859-4? "Scandinavian/Baltic languages" looks as well as "Northern Europe" to me taking into account incomplete support of Greenlandic and Sami languages. In any case it is worth to add that it was superseded by iso8859_10 and iso8859_13.

Change Byelorussian -> Belarusian is valid.

[mr-bronson]

Thanks @serhiy-storchaka, you're right - I should have said it's not "just" Baltic.

I have not found "North European" in the spec preview.

Just as you won't find "Nordic" in the ISO-8859-10 spec.

The spec preview says under Scope:

The set contains graphic characters used for general purpose applications in typical office environments in at least the following languages: Danish, English, Estonian, Finnish, German, Greenlandic, Latin, Latvian, Lithuanian, Norwegian, Sámi (but see Annex A.1, Notes), Slovene and Swedish.

It does list outliers like Slovene, which is not in the North. But Slovene was not originally in mind. ECMA-94 describes "LATIN ALPHABET No 4" as being

for general purpose applications in typical office environments in the following languages: Danish, English, Estonian, Finnish, German, Greenlandic, Lappish, Latvian, Lithuanian, Norwegian and Swedish.

The designation "North European" goes back at least as far as Roman Czyborra's page created in 1995, which became a major source of information for people looking for specs and information on the topic (e.g. copied in verbatim by aspell in 2004, cited by the MIME::WordDecoder module in Jan 2001, etc.) Czyborra had his own list of places that were linking to his work, but that seems to be lost to history now. Whether he came up with "North European" or copied it from something else, I don't know. "North European" was in the very first version of the Wikipedia article, from Jan 2004 to the present.

Rather than "North European", I put "Northern Europe" to match the other labels saying Western/Central/Eastern/South-Eastern Europe. But there's certainly a case for the now traditional designation.

In the Linux manpage it has note "Scandinavian/Baltic languages". On the other page it is described as for "North European languages"

I'm fine with either of those.

[mr-bronson]

It applies to everyone contributing to this repo:

Thanks for explaining. So apparently there's a difference between being a PSF Member or "a member of the python organization", as the GitHub hover text puts it, and being what the CoC terms "Members of the Python community". That's not immediately clear to someone skimming such a document. Where I come from, being a member (i.e. body part) of something is a lot more formal than simply a drive-by suggestion from an internet rando.

It's open source: https://github.com/ambv/cla-bot

Thanks for sharing. But as an outsider, I have no way of verifying whether there's an unbroken correspondence between that repo and the CPython CLA Bot. (And even if I did, who's going to browse through 200+ files to see what the bot does?) I'm not trying to be argumentative, but just hopefully that you can begin to see where I'm coming from.