this post was submitted on 15 Jan 2022
33 points (100.0% liked)
Open Source
31135 readers
335 users here now
All about open source! Feel free to ask questions, and share news, and interesting stuff!
Useful Links
- Open Source Initiative
- Free Software Foundation
- Electronic Frontier Foundation
- Software Freedom Conservancy
- It's FOSS
- Android FOSS Apps Megathread
Rules
- Posts must be relevant to the open source ideology
- No NSFW content
- No hate speech, bigotry, etc
Related Communities
- !libre_culture@lemmy.ml
- !libre_software@lemmy.ml
- !libre_hardware@lemmy.ml
- !linux@lemmy.ml
- !technology@lemmy.ml
Community icon from opensource.org, but we are not affiliated with them.
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
There isn't any good FOSS translator app, there are all very limitated or use also the Google Translate API. The only you can use a Google/ Bing/Yandex front-end Opensource, like Lingva Translate or Simply Translate. In any case the best in these type of apps, read the TOS and PP and select the one which fits the best for you.
If the backend just sends the data to google there's no point in using an alternative frontend imo.
In long term, I think the fediverse can solve this issue with federated and interconnected tools. Each software putting emphasis on a different aspect of analysis of language and each instance aiming to map the undefined and discuss contraversies.
Dictionaries can be built up from a subsection of instances that focus on definitions. Perhaps have some kind of functional approach to definitions.
Translation instances could use these dictionaries actively to quicker find good translations.
Lmao, the fediverse does not seem helpful or desirable here. I don't want to share my texts-to-be-translated with more 3rd parties, I want to share them with less!
If I must transmit to a 3rd party, I would rather it be a european company like DeepL (who is most likely following GDPR and is probably actually only using texts from their free service to improve their service rather than using it for building profiles on people like Google is...) than to use some thing hosted by some random person.
Fortunately there are several different free software translation engines which you can run locally, so, you don't need to rely on 3rd parties anymore. Via CHEF-KOCH's comment in this thread, I see that there are at least three different free software engines that can be run locally:
Each of those does also have free instances on the web, but, you don't need to rely on those... you can install the software on your own computer!
Im not suggesting that the text to be translated should be posted to instances. That would break the right to privacy.
However, I think a fediverse could improve such offline tools that you are referencing as well improve translation tools that has access to the internet.
This could for example be done by having instances specifically designed to create dictionaries through collaboration, perhaps with a very specific theme. An instance could have an aim for creating definitions in mathematics, for children or for carpenters. A dictionary could then be generated from a set of instances.
These definitions could be designed to allow for automation of translation. Perhaps through use of functional definitions.
The problem belongs to natural language in the machine learning domain and is way more complicated than simply dictionary building. Take a look for nltk and you will see how complex for machine to understand a simple English sentence.
No doubt its difficult, but we need to envision how to create translation tools that ensures the peoples independence.
I dont know how this could be achieved, but I do think we will see amazing ideas as the fediverse cultivates.
I am NOT against to have a FOSS translate app, actually I wish we have one. I just want to assure people understand the difficulties before they waste their time. We might need crowd funding. Some companies like Mozilla step up to take the lead in the tts and voice recognition which I appreciate.
I'm all for that!
I think the best thing we can do, is to organize. Not only internationally, but locally as well. To engage all kinds of people in digital issues.
I was thinking about a website/app that uses Google Translate API, but saves each result to its own database, so that every time anyone uses it, we slowly build an alternative together.
But I guess to have a properly functioning alternative you would also want to have the code that generates the results?
It's very complicated to build a good translator, it's not only that you need a rich diccionary for every language, also the rules, sintaxis and ortography of each of them. Because of this it needs many years of development. That is the reason because a lot of FOSS projects they just threw in the towel or they use Google or Bing APIs, and that only acceptable translators are found among proprietary apps, even paid ones (Deepl, which in it's free version is pretty limitated in vocabulary)