menu
Tatoeba
language
S'inscrire Se connecter
language Français
menu
Tatoeba

chevron_right S'inscrire

chevron_right Se connecter

Parcourir

chevron_right Montrer une phrase au hasard

chevron_right Parcourir par langue

chevron_right Parcourir par liste

chevron_right Parcourir par étiquette

chevron_right Parcourir les enregistrements sonores

Communauté

chevron_right Mur

chevron_right Liste de tous les membres

chevron_right Langues des membres

chevron_right Langues natales des membres

search
clear
swap_horiz
search
soliloquist soliloquist 17 octobre 2019 17 octobre 2019 à 19:56:16 UTC flag Signaler link Permalien

**Using the Vocabulary Feature for Corpus Maintaining**

Recently, I have listed some common Turkish spelling errors as search links. There are hundreds of incorrect sentences. We have fixed some of them, and are still fixing.

https://tatoeba.org/eng/sentenc...omment-1131503

I was planning to check those links from time to time (once in several months), but then I realized that we could use the vocabulary feature to continuously check and monitor all of them at once, like an alarm system. I created an auxiliary account for this purpose and added those searches as vocabulary items.

https://tatoeba.org/eng/vocabul...shSpellChecker

It might be considered as bending the tools, but since this feature isn't used by many people, I wanted to give it a try. Also, people adding new sentences with wanted vocabulary items are usually native speakers, so they can easily notice that those items are spelling errors and ignore them. I think the risk-benefit ratio is favorable for now. If a feature showing who added which vocabulary item were implemented, this would be even less of a problem.

{{vm.hiddenReplies[33249] ? 'expand_more' : 'expand_less'}} cacher les réponses montrer les réponses
TRANG TRANG 17 octobre 2019 17 octobre 2019 à 21:07:04 UTC flag Signaler link Permalien

It's a creative way to do corpus maintenance :)

While it is not the intended use for the vocabulary feature and does mess up a bit the "sentences wanted" list, your use case is still a valid one, I think. It only means that we need a way to distinguish between vocabulary that was added to request an example sentence and vocabulary that was added just as part of a list of vocabulary.

That's an idea that I already mentioned in another thread:
https://tatoeba.org/eng/wall/sh...#message_31919