unicode locale library

[zandimna]

coi ro do

        Unicode CLDR had recently accepted my request to add the lojban locale into their locale library. What this means is that by the release of CLDR 32 in September, we will have lojban locale in the computers which have the updates. (If you use some version of linux, it will appear in the list in /etc/locale.gen). If any of you are interested in translating locale data like date formats, country names etc into lojban, please request a survey tool account in CLDR and list lojban[jbo] in your request. Then, you will be able to add translations.

The Unicode Common Locale Data Repository

http://cldr.unicode.org/

This is what the tool looks like

co'o

[suke...@gmail.com]

Great job.

But I guess this is only the beginning. Standard OS inputs methods should have proper dictionaries, and I wish that dictionaries spread out too. It's already a pain having a proper spelling / grammar checking for widespread OSes (Linux, Windows, Android, iOS...) in my mother's tongue language (*), so I guess it would be much more difficult for Lojban.

However, do know anything about the current status of those? Do you think there would be a way to make it happen, at least for partially open-sourced (Linux & Android families)?

Thanks.

(*) It's been 27 years that a reform of French orthography has been official. However, Android still uses the old one (AFAIK), and Windows let you choose the one you want in a deep buried option (which default to "old").

[Pierre Abbat]

On Thursday, July 13, 2017 6:58:33 AM EDT suke...@gmail.com wrote:
> Great job.
>
> But I guess this is only the beginning. Standard OS inputs methods should
> have proper dictionaries, and I wish that dictionaries spread out too. It's
> already a pain having a proper spelling / grammar checking for widespread
> OSes (Linux, Windows, Android, iOS...) in my mother's tongue language (*),
> so I guess it would be much more difficult for Lojban.

Grammar checking can be done algorithmically, but since the language is fairly
young and people are still making up lots of words, a spell checker would be
difficult. However, there is no inflection of open classes in Lojban, so there
cannot be arguments like "mosquitoes/mosquitos".

> (*) It's been 27 years that a reform of French orthography has been
> official. However, Android still uses the old one (AFAIK), and Windows let
> you choose the one you want in a deep buried option (which default to
> "old").

I prefer the new way of writing numbers because it makes clear in writing the
distinction between e.g. 52/1100 and 50/2100, and I like "cigüe, aigüe, et
ambigüe" better than "ciguë, aiguë, et ambiguë". But some words just look
wrong without their circumflex. i to la'o fy. cigüe .fy. cu vindu rulsantyspa
.i la'o fy. aigüe .fy. cu kinli .i la'o fy. ambigüe .fy. cu smuvrici

mu'omi'e .piier.
--
lo ponse be lo mruli ku po'o cu ga'ezga roda lo ka dinko

[Sukender]

coi ro do

Well I agree about the fact the language is young, but if one uses this fact as an argument for not pushing spell check dictionaries, then I guess the language will never spread!

Dictionaries evolve. I am convinced that we'd better have incomplete an not fully accurate dicts rather than no dicts at all. It may (should) be updated later on. And better having basic spell check without grammar check, rather than nothing at all.
Don't you think?

I'd love seeing at least partial support of Lojban in Hunspell, LibreOffice, Notepad++, Android default keyboard(s), and many others. The more software bricks have support, the more people will accept the effort of adding or improving support for other software and systems (and I guess starting with open sourced will push closed source to move).

How can we help? Which software library / system should we start with (to get the greatest impact)? Which source of vocabulary should we use?
Any idea?

And what about translation websites? Is it too early to start thinking about including Lojban in popular 3rd-party ones?

By the way, I'm pretty sure a good Lojban keayboard+spell/grammar check (in the future) could include very fine rules about deducing next word to type (as most mobile-like virtual keyboards nowadays). Imagine your mobile device "remembering" you just typed "be", and automatically let two "bei" and "be'o" buttons appear on-screen until some terminator closes the block... :-)

About the French reform, well that's an old debate now. As for coding rules (formatting, naming), and as for laws, nobody 100% agrees the conventions, apart from those who wrote them! But the most important is that communication is facilitated, by ensuring people use the same conventions and rules.

co'o

[gleki.is...@gmail.com]

Em sexta-feira, 21 de julho de 2017 10:51:53 UTC+3, Benoit Neil escreveu:

coi ro do

Well I agree about the fact the language is young, but if one uses this fact as an argument for not pushing spell check dictionaries, then I guess the language will never spread!

Dictionaries evolve. I am convinced that we'd better have incomplete an not fully accurate dicts rather than no dicts at all. It may (should) be updated later on. And better having basic spell check without grammar check, rather than nothing at all.
Don't you think?

I'd love seeing at least partial support of Lojban in Hunspell, LibreOffice, Notepad++,  Android default keyboard(s), and many others. The more software bricks have support, the more people will accept the effort of adding or improving support for other software and systems (and I guess starting with open sourced will push closed source to move).

How can we help? Which software library / system should we start with (to get the greatest impact)? Which source of vocabulary should we use?
Any idea?

And what about translation websites? Is it too early to start thinking about including Lojban in popular 3rd-party ones?

By the way, I'm pretty sure a good Lojban keayboard+spell/grammar check (in the future) could include very fine rules about deducing next word to type (as most mobile-like virtual keyboards nowadays). Imagine your mobile device "remembering" you just typed "be", and automatically let two "bei" and "be'o" buttons appear on-screen until some terminator closes the block... :-)

On Android you can just use Multiling with Lojban and "Next word" plugins.

Yes, it has to learn from your typing.

You can make it learn faster using: 

Settings → word prediction → word prediction → Train next word.

Not that it's an ideal piece of software, it's the only one for Lojban yet.

[suke...@gmail.com]

Thanks for the info. Indeed there are a few interesting hits for "Lojban" in the (mainly used) store. I'll keep that in mind.

By the way, I'm currently preparing a script to generate a very basic (and incomplete, and inaccurate) Hunspell / Myspell / LibreOffice dictionary. It'll be better than nothing! I'll let you know by posting here.

[Pierre Abbat]

I'm looking at http://cldr.unicode.org/index/survey-tool/accounts .

> If there is an official organization for your language that would like to
> have liaison status for contributing data, please file a separate ticket
> under CLDR Change Requests. Other individuals can request accounts via the
> Unicode contact form. In resolving conflicts, they are treated as if they
> all belong to a single organization ("Guest").

Does the LLG have liaison status, should the LLG get liaison status, or should
we sign up as guest?

Pierre
--
ve ka'a ro klaji la .romas. se jmaji

[Pierre Abbat]

I got my login info yesterday and have been looking around. I will have some
questions.

Pierre

--
sei do'anai mi'a djuno puze'e noroi nalselganse srera

[Pierre Abbat]

The system is in vetting mode, which means that existing data can be fixed,
but new data cannot be entered. I'm guessing that we'll be able to enter data
in October. If not, it will be April or May when we can start entering data.
Meanwhile, we can discuss some translations and localization problems.

Pierre
--
.i toljundi do .ibabo mi'afra tu'a do
.ibabo damba do .ibabo do jinga
.icu'u la ma'atman.