Tatsuya Amano
banner
tatsuya-amano.bsky.social
Tatsuya Amano
@tatsuya-amano.bsky.social
Conservation scientist @uq-cbcs.bsky.social committed to overcoming biases and barriers in conservation and science. See
http://translatesciences.com
http://kaizenconservation.com
he/him
When considering non-English-language papers:
-25% for male non-native English speakers
-1% for male non-native English speakers from low-income countries
compared to male native English speakers from a high-income country.
September 24, 2025 at 12:21 PM
Thanks and a great point! It's somewhere in the middle but for accurate estimates I need to do some calculations...
September 24, 2025 at 7:45 AM
このような格差の数値化は、あくまで問題解決への第一歩に過ぎません。構造的な障壁を取り除くには、科学のあり方そのものを根本的に見直す必要があるでしょう。例えば、AI利用によって誰もが自分の言語で論文を発表し、読めるような未来を実現することができるかもしれません。このアイデアについては下記論文で検討しています:
doi.org/10.1371/jour...
5/5
AI-mediated translation presents two possible futures for academic publishing in a multilingual world
As the availability and performance of AI for language editing and translation continues to improve, we can imagine a future in which everyone can use their own language to write, assess and read scie...
doi.org
September 19, 2025 at 3:15 AM
一つの対策として、研究者の業績を評価する際に、英語論文のみに基づいた指標の利用をやめることが挙げられるでしょう。DORA(https://sfdora.org/)が提唱しているように、研究がどこで発表されたかではなく、その内容自体に注目した研究者の評価が重要だと考えられます。4/5
September 19, 2025 at 3:15 AM
実際、英語と非英語の両方の論文を含めると、生産性の格差は大きく減少しました。むしろ、英語を第一言語としない研究者や低所得国出身の研究者は、英語を第一言語とする高所得国の研究者よりも、全体として多くの論文を発表していました。
3/5
September 19, 2025 at 3:15 AM
我々が以前下記の論文でも一例を示したように、女性や英語非ネイティブ、低所得国出身の研究者は科学活動を行う上で様々な障壁に直面しているため、ここで示された生産性の格差は、これらの研究者の真の生産性を反映している訳ではないと考えられます。
doi.org/10.1371/jour...
September 19, 2025 at 3:15 AM
Quantifying these disparities is just the first step. Breaking down systemic barriers will likely require a fundamental shift in how we conduct science. One future vision? AI enabling everyone to publish and access research in our own languages. We explore this idea here:
doi.org/10.1371/jour...
5/5
AI-mediated translation presents two possible futures for academic publishing in a multilingual world
As the availability and performance of AI for language editing and translation continues to improve, we can imagine a future in which everyone can use their own language to write, assess and read scie...
doi.org
September 19, 2025 at 2:46 AM
We should stop using metrics based solely on publications in English to evaluate the performance of researchers. Instead we should focus on *what* is published, not just *where*, as advocated by DORA: sfdora.org
4/5
Home | DORA
The Declaration on Research Assessment recognizes the need to improve the ways in which the outputs of scholarly research are evaluated.
sfdora.org
September 19, 2025 at 2:46 AM
Indeed when we included both English and non-English publications, the productivity gap narrowed significantly. Non-native English speakers and researchers from lower-income countries often publish *more* papers overall than their native English-speaking, high-income counterparts.
3/5
September 19, 2025 at 2:46 AM
Obviously this does not reflect the true productivity of researchers facing gender, economic and #languagebarriers, as they face tremendous hurdles when conducting various scientific activities, as we previously showed in this paper for non-native English speakers:
doi.org/10.1371/jour...
2/5
September 19, 2025 at 2:46 AM
Reposted by Tatsuya Amano
Much pest and disease data is in grey literature or on ephemeral web sites. Collected by government departments and agencies but they don’t publish. In the global south trade magazines and newspapers may contain earlier reports than official sources. There is a place for AI web crawlers.
August 15, 2025 at 6:59 AM