Karibu kwenye WikiBabel

Ujumbe wetu

Katika Wikibabel, lengo letu ni kusawazisha upatikanaji wa habari kwa kuleta maudhui ya thamani ya Wikipedia kwa lugha zisizostahili za ulimwengu. Lengo letu ni juu ya Kiswahili, lugha inayotumiwa na watu milioni 50-100, lakini kwa habari ndogo sana zinazopatikana kwa Kiswahili kuliko Kiingereza. Kwa mfano, Wikipedia ya Kiswahili inarasa 38,000 ikilinganishwa na milioni 5.5 kwa Kiingereza; na maendeleo katika tafsiri ya mashine, tunaweza kusaidia kufunga pengo hili la habari. Maudhui hapa ni tafsiri ya Google ya maudhui ya Wikipedia iliyochaguliwa kulingana na kanuni hizi.

Jinsi Tunayotafuta?

Mafanikio ya hivi karibuni na ya kuendelea katika tafsiri ya mashine yameboresha ubora wa kutafsiri kwa kiasi kikubwa, na kuruhusu kutafsiri kiasi kikubwa cha habari. Wakati tafsiri sio ya ubora wa watafsiri wa wanadamu, wale ambao tunaona kuwa hundi walikuwa rahisi kueleweka, ikiwa ni kidogo ya kusikia kwa sauti wakati mwingine. Tulichagua seti ya kuchaguliwa ya makala za Wikipedia hapa kufanya tathmini ya kwanza, na zaidi katika bomba.

Kwa nini tunatafsiri: kupatikana

Je, si msemaji wa Kiswahili anayeenda kwenye Google kutafsiri na kutafsiri ukurasa wa Wikipedia wa Kiingereza wenyewe? Hii itawahitaji kujua kwamba Wikipedia ipo na ni chanzo cha ubora, kwamba ni bora kwa lugha tofauti, na kwamba Tafsiri ya Google ipo na kutafsiri kurasa. Hii ni bar ya juu kwa watumiaji wavuti mpya. Aidha, wangehitaji kulipa gharama mbili za data. Wikibabel ni kweli kutatua tatizo la kugundua kwa kuzuia matokeo na kuifanya inapatikana kwa injini za utafutaji. Tunatakiwa kufuata mfano wa Wikipedia na kushiriki katika programu ya Msingi ya Msingi, ambapo watumiaji wanaweza kufikia maudhui haya bila kulipa kwa gharama za data.

Je, hii ni ya thamani?

Dhana yetu ni kwamba ubora na kiasi cha tafsiri na urahisi wa kugundua hutoa wasemaji wa Kiswahili ambao hawazungumzi Kiingereza kwa habari muhimu. Tunatathmini hypothesis hii kwa kutumia bidhaa za uchambuzi ili kupima jinsi watumiaji ambao hupangia kurasa zetu kuingiliana nao.

Kwa habari zaidi wasiliana na close_the_gap@wikibabel.com

Welcome to WikiBabel

Our Mission

At Wikibabel, our mission is to equalize information access by bringing the most valuable Wikipedia content to the world’s underserved languages. Our focus is on Swahili, a language spoken by 50-100 million people, but with far less information available in Swahili than English on the internet. For example, the Swahili Wikipedia contains 38,000 pages compared to 5.5 Million in English; with advances in machine translation, we can help close this information gap. The content here is a Google translation of selected Wikipedia content based on these principles.

How We Translate?

Recent and ongoing advances in machine translation have improved translation quality significantly, and allow for translating massive amounts of information. While the translation is not of the quality of human translators, the ones we spot checked were quite understandable, if a bit weird sounding at times. We translated a select set of Wikipedia articles here to make a first evaluation, with more in the pipeline.

Why do we translate: Discoverability

Can a Swahili speaker not go to Google translate and translate the relevant English Wikipedia page themselves? This would require them to know that Wikipedia exists, that it’s better in some other language than in theirs, and that Google translate exists and translates pages. We are solving a discoverability problem by caching the results and making them available to search engines. We aim to be fairly high in search engine results for Swahili technical keywords (where we found far less information than on the English internet for example). We intend to follow Wikipedia’s example and participate in the Free Basics program, where users can access this content without paying for bandwidth.

Is this valuable?

Our hypothesis is that the quality and amount of the translation and the ease of discoverability provides Swahili speakers who don't speak English with useful information. We are evaluating this hypothesis by using an analytics product to measure how users who land our pages interact with them.

For more information contact close_the_gap@wikibabel.com