EtikettenAlle Etiketten anzeigen
LizenzCC BY 2.0 FR
Dieser Satz ist ein Original und nicht als Übersetzung entstanden.
hinzugefügt von CK, am 29. Februar 2020 um 16:07
Lizenz gewählt von: CK, am 29. Februar 2020 um 16:07
bearbeitet von CK, am 2. März 2020 um 01:21
Satz Nr. 8585408
@CK, what was the purpose of adding these two near-duplicate sentences?
We already have enough sentences with this pattern:
A lot of them have been linked to their "everyone" counterpart so I don't really understand why this additional pair was needed.
They are both valid sentences.
They are linked to each other, so people can add their translations to both at the same time.
Eventually, we will likely get both, but they won't be linked, so adding them at the same time and linking them is useful I think, since all linked translations become indirect links.
> They are both valid sentences.
Did Trang say they weren't?
> They are linked to each other, so people can add their translations to both at the same time.
But many people won't want to. By adding the near-duplicates, you're:
(1) increasing the work load of people who feel compelled to add translations to the different variants of a sentence
(2) decreasing the proportion of translated sentences, because many people won't add translations to near-duplicates
(3) saturating the corpus with sentences that don't add value
> Eventually, we will likely get both
That assumption is unfounded and probably wrong. Personally, I avoid adding near-duplicates like this, and I'm sure many others do the same. But by adding the near-duplicates yourself, you're ensuring that we'll get both, aren't you? But then you're also ensuring that your sentence ownership count is as high as possible, so I guess it's worthwhile.
> linking them is useful I think, since all linked translations become indirect links
But indirect links cannot be trusted the way that direct links can. Why is it better to have two near-duplicate sentences with mixed direct and indirect translations when a single sentence would get a higher proportion of direct translations?
Here are just a few random examples where it has happened already.
[#51719] Everybody is equal before the law. (Eldad)
[#5044989] Everyone is equal before the law. (Zurich899)
[#40397] Everybody seeks happiness. (CM)
[#276541] Everyone seeks happiness. (Dejo)
[#2164592] Everybody ran outside. (Hybrid)
[#3425642] Everyone ran outside. (AlanF_US)
[#354852] Everyone makes mistakes. (Hautis)
[#1883089] Everybody makes mistakes. (Spamster)
[#3077789] Everyone likes them. (shake0615)
[#7759870] Everybody likes them. (raggione)
[#430471] Everyone likes her. (FeuDRenais)
[#451284] Everybody likes her. (minshirui)
[#4777240] Everyone has to start somewhere. (devilyoudont)
[#5631616] Everybody has to start somewhere. (Hybrid)
I would like first to remind you that we still have this issue: https://github.com/Tatoeba/tatoeba2/issues/2107. Until we solve it, I would appreciate if contributors could avoid linking sentences in the same language. It is not something people would intuitively do anyway and there might be cases where it's acceptable. But nonetheless, with the current features of Tatoeba, it is absolutely not a good practice.
Secondly, I am concerned that such contributions are setting a bad example to others. You are here creating intentionally near-duplicates with the rationale that "eventually, we will get both". If someone would follow this rationale, they could just go ahead and copy every sentence with "everyone", replacing it by "everybody", and vice versa. I would find this quite useless but also quite toxic.