Talk:DeepL Translator

This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Articles for creation This article was reviewed by member(s) of WikiProject Articles for creation. The project works to allow users to contribute quality articles and media files to the encyclopedia and track their progress as they are developed. To participate, please visit the project page for more information.Articles for creationWikipedia:WikiProject Articles for creationTemplate:WikiProject Articles for creationAfC articles This article was accepted from this draft on 4 October 2019 by reviewer K.e.coffman (talk · contribs). Linguistics: Applied Linguistics Low‑importance This article is within the scope of WikiProject Linguistics, a collaborative effort to improve the coverage of linguistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.LinguisticsWikipedia:WikiProject LinguisticsTemplate:WikiProject LinguisticsLinguistics articles Low This article has been rated as Low-importance on the project's importance scale. This article is supported by Applied Linguistics Task Force. Websites: Computing Low‑importance This article is part of WikiProject Websites, an attempt to create and link together articles about the major websites on the web. To participate, you can edit the article attached to this page, or visit the project page.WebsitesWikipedia:WikiProject WebsitesTemplate:WikiProject WebsitesWebsites articles Low This article has been rated as Low-importance on the importance scale. This article is supported by WikiProject Computing. Software: Computing Low‑importance This article is within the scope of WikiProject Software, a collaborative effort to improve the coverage of software on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.SoftwareWikipedia:WikiProject SoftwareTemplate:WikiProject Softwaresoftware articles Low This article has been rated as Low-importance on the project's importance scale. This article is supported by WikiProject Computing.

The WP:COIRESPONSE . DeepL Repr (talk · contribs) has been paid by DeepL on their behalf. Their editing has included contributions to this article. COI declared here.

On 24 January 2023, it was proposed that this article be DeepL. The result of the discussion was no consensus.

Why are there two standard versions of Portuguese and English counted as separate languages? I think it is slightly misleading and the actual number of translated languages should be 29 (with an option of selecting a variety in these two languages). (Jan)

The article is so superficial that it really looks like a "lose weight fast" scam ad. Some questions that should be addressed are:

• What group (business? academics? ??) is developping DeepL and where?
• Which scientific papers describe the underlying technology a bit more specifically?

PhS (talk) —Preceding undated comment added 06:25, 25 January 2020 (UTC)[reply]

As of today, I think the article sufficiently explains who is developing it and that the technology is a trade secret, while describing the general strategy. (I'm not an expert in the field.) Fourthark (talk) 01:50, 16 May 2021 (UTC)[reply]

I was trying to improve the wording in the stats section, because it sounds awkwardly German. That section contains the sentence

> In Germany it ranks 276th, with 30% of the site's users concentrated, followed by France, Switzerland, Spain and Italy.

"Concentrated" is very awkward here but I can't fix it because I don't know if it means something different from the later sentence

> As of 12 May 2019, 21.8% of traffic comes from Germany, 17.4% from France, 10% from Spain, 7.9% from Switzerland and 4.1% from Poland.

If the same meaning is intended, the first sentence should be deleted because it is out of date. Fourthark (talk) 01:56, 16 May 2021 (UTC)[reply]

@AngryHarpy: The reception section wording is a remnant from the 2019 article, which was promotional. It quotes positive aspects mentioned by press, while disregarding the articles conclusion. E.g. in the case of Le Monde, concluding that "Google was still a far surperior service", per Connexion translation. More academic sources that thoroughly evaluate the service are needed.

The section also needs restructuring, e.g. the 7 sources used for blind tests contain reviews that can be used throughout the section. Alexa and Similarweb statistics have been deemed un-encyclopedic AFAIK.

Btw the privacy section can probably be removed, original research and primary source. IgelRM (talk) 13:54, 8 July 2021 (UTC)[reply]

WP:ALEXA, even though it's largely useless here and maybe shouldn't be, so it's gonna be a resounding eh on my part. I've removed the Privacy section as suggested by you, seems like a reasonable move. Your issues with the Reception section seem also well justified; I only came upon this by coincidence and wanted to prod a little. The criticism from the LeMonde-piece and anything else from around that time should probably be put into perspective for referring to an early version of the service, and the site's usability seems to have unambiguously improved since then, but yeah, some academia'd be nice. AngryHarpy‍^talk 16:40, 8 July 2021 (UTC)[reply

]

AngryHarpy, this has been the most positive interaction I had on WP, so thanks.

Ok, I have read the Alexa part. Though the additional Similarweb estimate graph still doesn't feel right.

You asked for more examples and SPIEGEL wrote that they had an overall positive impression with good conjugation and sentence structure while there have been some issues with some translations. ("Allerding hakt es auch bei DeepL noch an manchen Stellen, einige Übersetzungen waren sprachlich oder grammatikalisch fehlerhaft. Insgesamt überzeugt der Onlinedienst aber mit der richtigen Konjugation von Verben und stimmigen Satzkonstellationen."). This is all from 2017, but the section notes "reception in 2017" (newer reception would be useful of course). So this section needs the NPOV template in my opinion. IgelRM (talk) 15:59, 16 July 2021 (UTC)[reply]

I am working for DeepL (just to be very transparent regarding COIs). I have first suggested a long edit on December 3, fixing several errors in the articles, but also trying to improve longer paragraphs. Probably this was a mistake. The edit was reverted two weeks later. Today I tried to only correct three short passages that are clearly wrong. I edited only these three parts, and entered very clear information on why the information was wrong before. Unfortunately my edits have been reverted again, without giving any reason based on the content. How can I correct the wrong information, if everything is reverted right away?

Can someone please have a look at the three little changes I made today, and re-implement them? It is the ambition of Wikipedia that there are no false claims in articles. It would be great if someone could check these three edits and the reasons I gave for the edits. It is about wrong information, and nobody wants false claims to be present in the article. DeepL Repr (talk) 13:58, 18 December 2021 (UTC)[reply]

Hi, answering here from my talk page. You are acting in good faith and this article has multiple issues with citations as is. However, conflict of interest editors need to use the

Request edit system instead. I saw your first edit had reliable sources you could use in the request form. I only reverted because I did not have time to look throughly at your claims. IgelRM (talk) 16:48, 18 December 2021 (UTC)[reply

]

Dass Frahling sich eher als Wissenschaftler versteht, der lieber in Ruhe seinen Forschungen nachgeht als sich mit der Hektik des Tagesgeschäfts herumzuschlagen, konnte man im vergangenen Jahr bereits ahnen, als er nicht nur UEPO.de bat, sämtliche Fotos von ihm aus früheren Artikeln zu entfernen. IgelRM (talk) 13:09, 3 August 2022 (UTC)[reply]