Toki Pona has an ISO 639-3 code as of this month! Source: https://iso639-3.sil.org/code/tok

I and many others worked hard on the code request. We're very glad SIL (the organization in charge of ISO 639-3) is willing to acknowledge that Toki Pona and its community and literature have become notable enough for inclusion in the standard. :D

I opened an issue on GitHub: https://github.com/Tatoeba/tatoeba2/issues/2898

Thanks!

How can I change the license of my original (non-translated) sentences? I would like to set them to CC0 if possible.

That IS more recognizable as a bird, but much less regocnizable as the bird on the Láadan book. The shape and colors are different, and the branch is missing. Keeping the same silhouette and similar colors is crucial, if we want Láadan speakers and enthusiasts to recognize it as the bird from the book cover.

I think it would be a better idea to augment the original image to have bolder lines and clearer shapes. For example like this: http://i.imgur.com/2n6wrnZ.png

It's also worth pointing out that Tatoeba by default sorts search results with the least amount of words first. It may be a good idea to change the sort order in the advanced search options.

After some googling I found out that the first (and possibly only?) official publication describing Láadan looks like this: http://i.imgur.com/HlJayUG.jpg

I think that bird on the cover might be fitting to use as Láadan's icon. Here's an example I threw together of what it could look like as a 30x20 pixel Tatoeba flag icon: http://i.imgur.com/Qp8njCs.png
It's a little small, but it still looks recognizable.

I recall GrizaLeono mentioning in a previous comment that he mainly uses this page to translate:
https://tatoeba.org/eng/activit...late_sentences

I agree, but I fear people might abuse this to get rid of bad ratings. Especially when there's a disagreement about a grammatical issue or something similar.
(#4940946 is an example of a case where this might be an issue. Three users in the comments agree the sentence should be changed, but the sentence owner refuses.)

However:
- this problem can for the most part be solved by notifying people when a rated sentence has been changed. People can simply renew their previous rating when the changes are minimal.
- bad ratings that remain after a sentence has been corrected is a bigger and much more common problem. So I would say having ratings deleted from a changed sentence still has a higher priority than notifications.

English translation for our staff:

Something seems to be wrong in the program:
Using the button "random sentences" to search for Dutch sentences, I found that the button "next" would go back to previous addresses after a few sentences.

By the way, in case it helps to know:
If you link or unlink a sentence (be it by clicking the (un)link icon next to the sentence, or by manually entering the sentence number) the list resets, and it correctly displays the Languages preferences.

If I can weigh in my own opinion...

Marking a sentence as "unsure" or "not OK" is like openly declaring there is (likely) something wrong with the sentence. Doing this is useful for learners using the site, as they will know to avoid using these sentences in their studies.

Because of this I'd say there is nothing wrong with marking unnatural and incorrect sentences as "unsure" or "not OK". In fact, I would encourage this.
However, I would advise against marking near duplicates as "unsure" or inappropriate sentences as "not OK", because this would communicate to other users that there is probably something objectively wrong with the sentence, when there is not.

If you adopt orphan sentences, you can correct them. You can even unadopt them afterwards, if you want to.

And yes, welcome to Tatoeba! :)

I spent a little time fixing furigana mistakes. Since I am in most cases not allowed to remove the "furigana mistake" tag, I marked the corrected sentences with "furigana fixed", so you can easily find them here: https://tatoeba.org/eng/tags/sh..._with_tag/6998



I found a few small bugs with the new furigana system, though.

Apparently the full width question mark (？) is seen as a character that can receive furigana, which means it shares the furigana with any character following it, such as in the sentence #2929493. The same problem exists with other full width characters, as can be seen here on the test server: https://dev.tatoeba.org/eng/sentences/show/3837896
I had to remove the｛｝after ＝ and ｔｅｓｔ , because got errors that those strings lacked furigana.
Note that full width characters don't "share" furigana with the following characters in the automatically generated example: https://dev.tatoeba.org/eng/sentences/show/3837895 (click the black かな buttton to view)

Since this behavior doesn't happen with the Japanese comma, full stop, and quotation brackets (、 and 。 and 「」), I assume they are in one way or another categorized as exceptions. I would in that case add several other reading signs, such as （ ） ？ ！ to the same list.

Most other signs, such as ％ and letters and numbers should probably not be added to that list and be able to take furigana, for sentences such as #117779 and #161874. However, for sentences such as #408303 it would be useful if they are allowed to have an empty furigana value. At the moment the furigana system does not accept either of the following:

そうです。日本語｛にほんご｝ではウェートレスは英語｛えいご｝の"waitress｛｝"と"weightless｛｝"にも該当｛がいとう｝する。でも"waitress｛｝"という意味｛いみ｝が普通｛ふつう｝だね。

そうです。日本語｛にほんご｝ではウェートレスは英語｛えいご｝の"waitress"と"weightless"にも該当｛がいとう｝する。でも"waitress"という意味｛いみ｝が普通｛ふつう｝だね。

due to this error:

The following characters lack furigana: w, a, i, t, r, e, s, s, w, e, i, g, h, t, l, e, s, s, w, a, i, t, r, e, s, s.

There's also a related problem with numbers. The current automatic furigana system does not assign furigana to numbers, so the string ８才 gets assigned ８才｛さい｝. This displays correctly in the preview before accepting the automatic furigana as correct, but when accepted it reads it as [８才](さい) (furigana shared by both characters), instead of ８[才](さい) (furigana only on the kanji)
(I made an example of this on the test server: https://dev.tatoeba.org/eng/sen.../show/3837897)

If numbers are made to accept a null string this could be fixed by automatically adding a null furigana value to numbers, which can optionally be filled manually. So then the system would automatically assign ８｛｝才｛さい｝ instead of ８才｛さい｝, which can then manually be changed to ８｛はっ｝才｛さい｝ or ８才｛はっさい｝ when desired.
However, the disadvantage to this would then be that strings like 10分 would most likely get assigned 10｛｝分｛ふん｝, which doesn't take the in account the consonant change from ふ to ぷ that happens due to the preceding "10". This is easily missed, which would most likely lead to a number of cases of not entirely wrong but certainly misleading furigana getting accepted by well-meaning users.



A bit of a long post, but hopefully it was useful feedback.

It looks like you are right. It seems the Khmer pronunciation is [kʰmaːe], which is probably why Esperanto Wikipedia also uses the term "Kmera lingvo".

I have changed the translation on Transifex. It should appear on the site whenever the language files are updated again.

Yeah, I meant to update my post, but I guess I got distracted adding translations to all of those.

I agree we could use some more good examples. And I'd say preferably one or two with "customer-friendly" and "user-friendly", since those are pretty common. And maybe "family-friendly"

Good idea! I've made a list (https://tatoeba.org/nld/sentences_lists/show/4898) and added a link in the first comment.

For the native English contributers:

A while back I have started keeping track of English words and expressions I found as I used Tatoeba, that have no or very few Tatoeba sentences. I thought it might be useful for our more active English contributers, such as @CK, @Hybrid, @patgfisher, @AlanF_US, etc. Or anyone else who wants to use them of course.

I'll just start with the first ten:

- general populace
- have it black on white
- good grief
- geez / jeez
- if I (do) say so myself
- prime time (like on TV)
- [noun] friendly
( ↑ such as "customer friendly")
- one-on-one
- unsurprising(ly)
- day care / day-care center

You can add your sentences to this list, if you like:
https://tatoeba.org/nld/sentences_lists/show/4898

Do you still have this problem? I'm not a member of the site staff, but if you only have this problem with certain sentences, it may help them if you tell us the ID(s) of the sentence(s) where this happens.

And I think it may also help them if you made a screenshot of it. (You can upload the screenshot to a free image hosting site, such as imgur.com)

I want to report that two times in the past week I have seen a sentence with an English flag that kept changing back to English when the owner tried to change it to the correct language.

The sentences where this happened are #4589968 (owner tried changing at least 4 times) and #4644008 (owner tried at least three times)

It's not the only or even the primary metric by which English sentences are judged. It's just a list of sentences CK thinks are correct, made available as another tool people can use. That's why they're also referred to as CK's whitelists, not Tatoeba's whitelists.

Tommy_san has comparable lists for Japanese sentences, by the way. There are many other users who make similar lists, but it looks like CK and Tommy_san use it the most.