Show newer

Little headsup, Mastodon developers seem to be testing a language selector in the posting interface (see attached image).

The idea is you set the language you're posting in which will make it easier for other people to filter your post out if they don't speak it.

You can already set your default language in the Preferences/Settings section, but the idea of this new feature is to let you optionally change the setting for each particular message. This is handy if you toot in several languages.

I have no idea if/when this will be rolled out, the developers' instances often get test features like this.



Hi!👋 I'm a translation student from the Netherlands interested in books, languages, nature, gaming, queer issues and more. Feel free to recommend me people to follow, as I really know no one here :)

At the moment I unfortunately do not have much time to dedicate to language learning, but the languages that I mainly study are Swedish, Japanese and Faroese.

A translation project of volunteers is crowdfunding some money to make central works of Critical Psychology available via Open Access:

Excited that my book with Andrew Nevins "Contemporary research in minority and diaspora languages of Europe" (Language Science Press) is nearly complete! Download it here for free!

The seventh conference on Machine Translation, ,

has a "Shared Task": Large-Scale Machine Translation Evaluation for [24] African Languages

"We do so by introducing a high quality benchmark, paired with a fair and rigorous evaluation procedure."

We would like to search for translated articles on scientific repositories like .

Searching for "translation" or "translated" gives many false alarms. Adding a language would likely work much better. Does anyone have an idea how to get a list with the names of "all" languages?

Similarly how to find a list with the translation of "translation" and "translated" in many languages?

Wikidata has a list of 100+ languages and the word translation (but not "translated").

Being in a #public institute in #india, the government has directed us to use #hindi along with #English in all official mails.

With members from all over the country in the community, this was exclusionary.

At the same time, a subset of Hindi speakers used a highly #sanskrit -ized version of the #language in debate mailing threads, presumably to show intellectual prowess?

A student, and a friend of mine, sent out this mail today in response to this practice.

#linguistics #colonization

Just back from our #OER22 session.
We're an ad hoc group bringing together from Switzerland, Cameroon, and Canada. Given our multilingual contexts and core values in #OpenEducationalPractices, we've recently started a multilingual wiki through which we wish to develop shared understandings from the ground up.
Feel free to join us!

Btw, if any of you latest arrivals from twitter are interested in languages, linguistics, translating etc etc, has room for new users. Tooting in any language is not only allowed but even encouraged!

Just make sure to read the instance rules before applying for an account.

Hi, ich bin Uli, bin bei den Grünen in #Karlsruhe aktiv und interessiere mich für #birding, etwas für #fotografie und mache leidenschaftlich gerne Seife. Ich habe zwar vor 20 Jahren in einer AG an der Uni mal HTML zu Fuß programmieren gelernt, aber da endet mein Verständnis für Code auch schon. Ich bin Diplom-Übersetzerin für Medizin, habe aber die letzten Jahre als Trainerin für Selbstorganisation und Improved Reading gearbeitet.
Ich freue mich, hier neue Menschen kennenzulernen!

About half of the CrossRef database entries of article that are or have translations do not have information on the language.

Also in a few percent of cases the language in the database does not match the language of the title.

There are many tools to estimate the language of a text. Does anyone here have any advice on picking one (that could work with titles)?

Would be great to have a method that indicates how accurate the estimate is.

Show thread

Machine Translation Benchmark Dataset for Languages in the Horn of Africa

Lesan AI has just published their first dataset on GitHub with data for , , , and .

Looking at the data (they manage the |s for scholarly publishing) has on translated articles I found some Easter eggs.

They have a publisher, the "Society of Psychoceramics", which has several journals.

My favorite, working on historical climate data, is the "Journal of Metadata Perfection". Would love to be able to publish there.

They published the articles:
- When your best metadata isn't good enough: working with an imperfect specification
- but what if i tried again

The Deep Learning Indaba 2022, a gathering of the African Machine Learning and AI Community, is in Tunis: Applications Open

We will be at the tomorrow (Monday).

We are asking for feedback on what we can do to promote the translation of the scholarly literature.

Our session will be in the time slot 14:45 – 15:30 CET (Berlin). 8:45 New York, 10:45 Buenos Aires, 17:45 Tehran, 19:45 Mumbai, 20:45 Jakarta, 22:45 Tokyo.

A Call to Diversify the Lingua Franca of Academic STEM Communities. "Create infrastructure that standardizes and facilitates the language translation process and hosting of multilingual publications."

Strengthening research capacity: a systematic review of manuscript writing and publishing interventions for researchers in low-income and middle-income countries

"Across interventions, leaders expressed the importance of a high ratio of mentors to participants, the need to accommodate time demands of busy researchers, and the necessity of a budget for open access fees and high-quality internet connectivity."

Show older

Fediscience is the social network for scientists.