Note you might be interested in looking at interactive machine translation systems and the associated research papers - for example Lilt [1, 2] or TransSmart [3, 4] - while those are designed for professional translators rather than for language learners, what they do is very similar to your long-term aims for this project (a specialized MT system that, given a partially written translation to the foreign language and some context written in in the user's native language, suggests the following word).
The names of all countries that are written in any right-to-left languages (e.g. Arabic or Hebrew scripts) are written backwards. They should be right to left, but the letters are shown left to right.
Chinese: 程序错误 (程序=program, 错误=error) is used for the generic sort of software bug. If the bug is a security hole, then it's instead called 漏洞 (leak). Though sometimes the English word "bug" is used, see for example https://zh.wikipedia.org/wiki/%E7%A8%8B%E5%BA%8F%E9%94%99%E8... which uses a mix of the two terms. Sometimes it's mistakenly capitalized as BUG, see for example https://baike.baidu.com/item/%E8%BD%AF%E4%BB%B6BUG/12618010 (this phenomenon in Chinese of capitalizing some English words that aren't actually acronyms happens with some other words too, like APP instead of app/application, etc).
Ah I meant P2P in the "from user to user with no transaction fees charged, doesn't go through Visa/Mastercard/Unionpay" sense (similar to Venmo, except that in China businesses accept it everywhere), not in the "fully decentralized and doesn't need a server" sense. For Zhifubao the servers are run by Ant Financial (it's a spin-off company of Alibaba), for WeChat the servers are run by Tencent. You can deposit/withdraw money to your bank account.
I'm assuming you're probably more interested in the new digital RMB wallet technology which supposedly works even if your phone doesn't have internet access, but I'm not familiar with how that is implemented (I'm guessing it uses blockchain and broadcasts the transactions later when you connect to the internet). There's some info about it at https://en.wikipedia.org/wiki/Renminbi#Digital_Renminbi
I meant distribution as a zip file that you can load in developer mode, like the installation instructions for Bypass Paywall ( https://github.com/iamadamdev/bypass-paywalls-chrome ). It's not a very user friendly installation process but it still works. But yes, CRX-based sideloading no longer works on Chrome.
A Chromium fork is going to be a pain to maintain. My contingency plan if it gets removed from the Chrome store is to try to get it accepted into the Edge and Opera stores, and ask users to switch to either Edge or Opera (and provide sideloading instructions for those who want to stick to Chrome).
It works fine with the current versions of Chrome (and Chromium-based browsers like Edge), you'll just need to sideload it once it gets removed from the Chrome store. Alternatively, if/when I manage to get it accepted into the Edge store, you could switch to Edge.
I tried porting the extension over to Firefox when Firefox switched to WebExtensions, and at the time there were tons of incompatibilities, mostly with Firefox's Shadow DOM implementation (HabitLab is a huge and complicated codebase, porting it is non-trivial - I had an issue tracking it at https://github.com/habitlab/habitlab/issues/137 ). I'm sure it's a valid option for smaller extensions however. At the moment I'm trying to get it accepted on the Edge store, as Edge is much more compatible with Chrome extensions than Firefox.
I definitely feel for the author - the Chrome Extension team has been growing increasingly developer-hostile recently. My own open-source extension HabitLab ( https://habitlab.stanford.edu/ ) that I've been maintaining for the past 3 years is going to be removed in 2 days (got a 14-day removal notice for permissions even though all permissions it requests are used and needed, and every update I try to submit is rejected by their system after about 3-4 days) and I feel utterly helpless. It's only used by about 12,000 users so unlike PushBullet I probably don't have the visibility to get a human to intervene, so will be going the way of Kozmos most likely.
I suspect the reason for few users is because it's relatively simple to build such an app and so you have tons of competitors, such as Pleco. For example, the visualization you have is pretty similar to a learning tool that annotates subtitles I built in 2013 - it's no longer online but you can see screenshots at https://dspace.mit.edu/bitstream/handle/1721.1/90411/Miller_... and a video at https://www.youtube.com/watch?v=3j-eXUB3eaA
I got the same notification yesterday morning for my own open-source extension HabitLab ( https://habitlab.stanford.edu/ ) - same vague request for "you're not using the minimal set of permissions" without mentioning what permissions they want me to stop using (HabitLab is already using the minimal set of permissions for the features it implements - any removal of permissions would have to be done at the expense of reduced functionality). Emailing just results in them sending me a link to the policy. So this is definitely not an isolated case.
This is quite similar to HabitLab https://habitlab.stanford.edu/ which also includes a number of other options (such as pausing videos on youtube before playing them, hiding comments and news feeds on Facebook, etc). [Disclaimer: I built HabitLab]
[1] https://lilt.com/technology/translate [2] https://lilt.com/research [3] https://transmart.qq.com/ [4] https://arxiv.org/pdf/2105.13072.pdf