My view is that sentences like “The cat drinks milk” or “The kid ate a hamburger” are far more useful than sentences that rely heavily on pronouns like “he/she” and vague references like “this/that.” As I mentioned, almost any verb can be combined with these generic words to produce endless sentences. As a result, examples such as “He eats that,” “This can be combined with that,” “This can be considered that way,” or “He refutes that” do little to enrich the diversity of the corpus.
I understand your point that not every sentence needs to be translated. However, I am concerned about having a large number of sentences that cannot be translated into my language. For instance, when a learner searches for a verb, if many of the results are sentences “generated” using “this/that,” they may either find no translation in their language or encounter translations that sound unnatural. Moreover, such examples do not help learners understand what kinds of words the verb is typically used with.
Here is an example I particularly dislike [#6760017]: “I wasn’t surprised when Tom told me I didn’t need to do that.” I have no idea in what context this could be used. It would be much clearer and more helpful to make the sentence more specific, for example: “I wasn’t surprised when my manager told me I didn’t need to rewrite the entire document.”

What I’m discussing is really not an issue of ambiguity. I’m just somewhat tired of the practice of generating large numbers of sentences by relying on generic pronouns. What I dislike is when users repeatedly combine “this/that” and “he/she” with only one (or maybe zero) word that actually carries meaning.

For example, if there are currently no sentences containing “eat” on Tatoeba, I wouldn’t want anyone to just keep adding sentences like “He eats this.”, “He eats that.”, “She eats this.”, “She eats that.”, “We eat these.”, “We eat those.” and so on.

> if language B requires greater specificity than language A in some situations so a sentence using context-dependent words like pronouns cannot be translated 'literally', you may like to add multiple possible translations, each meant for different contexts.

This is not very practical in many cases. If the language A is English, then there are already way too many sentences. Providing every possible translation for the simplistic sentences that are lacking context would be a monumental effort. I originally came to Tatoeba from the Clozemaster language learning website in 2023, because I noticed that it was frequently offering somewhat odd Belarusian sentences, that were implying that "Tom" or "he" had the feminine grammatical gender. Since they had been imported from Tatoeba, I tried to challenge this on Tatoeba as rather misleading for language learners, but was accused of being a "homophobe". Yet I still think that the unusual grammatical gender should be preferably only used in the sentences, where the other words clearly confirm the homosexual/transsexual context, rather than something generic like "Tom is a teacher." or "He visited a dentist yesterday."

Thank you for your reply. However, that is not what I am asking.

I read the “How to Search for Text” section in Tatoeba’s Wiki, and also the “Field search operator” section of the Manticore documentation, where it gives an example “@title hello @body world”.

The @transcription field does not find matches in translations, it only works in languages where the sentence can be written in different scripts, such as Chinese or Japanese, as far as I can tell.

If I use “@text papa @transcription potato” to try to find Spanish sentences with English translations, there are no results (because Spanish doesn’t have a transcription field), even though there are many sentences with “papa” in the source and “potato” in the translation.

If I just search for “papa”, the results include sentences where that word means “dad,” where it means “pope,” and where it means “potato.” I’m wondering if it’s possible to filter those results based on the specific translation that I’m looking for.

I wasn’t there to talk about it, but from what I heard, this screenshot comes from an early version of Tatoeba, and the code has been completely rewritten from scratch since. That’s why our current code repository is named "tatoeba2" [1], that screenshot being "tatoeba 1", or simply "tatoeba" at that time.

In that early version, sentences are not organized as a graph but as a table: only one sentence per language in each sentence group. I guess this model, while having constraints, helped having the feature you want.

[1] https://github.com/Tatoeba/tatoeba2

I think for such sentences that has poor word to word matching, we should simply add a button to mark them as "implied" or something.

For the example of [#3378275] > [#4919050], we can not match "Chinese" to anything, so we can mark it as "implied". However, we can still create matching like [Chinese characters] > [字] and [characters] > [字]

One of the reasons why I give this idea is that, some people may not be able to understand both languages well. So a matching of how words are translated could help them understand the sentence structures easier.

​I don’t frequently contribute to the English Wiktionary, which seems more streamlined due to its large community of maintainers. However, as a contributor to the Chinese Wiktionary, I find the infrastructure much more challenging. Many templates and modules are incomplete or overly complex; editing a single entry often takes hours and is prone to errors.

​Furthermore, Chinese grammar is often a subject of intense debate, frequently leading to edit wars. I’ve also noticed inaccuracies in Korean and Chinese entries, particularly when they are translated from English by contributors who may not be proficient in the target language.

​My point is that we could effectively build a translation dictionary within Tatoeba. For instance, when looking up a word, seeing both the keyword and its translation in bold—similar to how some dictionaries highlight equivalents—would greatly benefit learners in understanding sentence structures.

​Regarding the preservation of Chinese languages, Tatoeba handles this exceptionally well. By treating Mandarin, Cantonese, Wu, and others as distinct entities, it helps maintain the linguistic integrity and "purity" of each language.