We currently have the data for the radicals hard-coded, however we plan to use the API in the future to automatically update this data, so if WK changes their radicals, it will be reflected in the keyboard too.
As for its use, there was a point in my learning where I had been studying radicals heavily (using the heisig method). I knew the kanji around me by radical but didn’t know how to say them, so could only look it up via a radical search (which is time consuming, needing at least several taps and a search through a list). So I made something that lets me type a radical word, and get a kanji out. I use it a fair amount at the 区役所 when trying to work out what is written on forms (I can recognise and put an English keyword to many radicals). I figured if I found it useful others might find it useful too (or might indeed have requests on how it could be made to be useful for their particular use case). If you’ve heavily studied radicals (particularly via the Heisig Method), then you might find it useful too. If you don’t, then that is valid feedback too.
And yes, you can’t put ads in a system-wide keyboard I’m afraid. Nor is it trivial to implement any kind of iAP in one either, which limits options. So for the moment, it is a paid app. I do appreciate actionable feedback (particularly if it’s polite and useful - and if you look at the Trello board you’ll see that there are cards to follow up on some of that feedback, including how to make a free version) but I’m not particularly interested in getting into any philosophical argument about what it means to pay for something.
As for it being “beta” - we’ve already gone through a process of testing it via hockeyapp with a couple of interested people, and it does what it’s intended to do, autosuggests kanji based on a radical keyword. In that respect its not “beta”. That doesn’t mean there aren’t any problems with it, and more importantly it doesn’t mean that somebody doesn’t look at it and go “this is alright… But I really wish it did this thing”. The original post was meant to represent our eagerness to follow up on that feedback, and no implication of making someone a QA tester.
More crucially maybe the feedback is generally “I just don’t find this useful”, which is also valid, at which point we’ll probably not develop it any more and move onto something else. Meaning those extra weeks we could have put into working out how to implement in app purchases or anything else would have gone to waste. Therefore the decision was to release first, paid, and see how it worked out.
It’s definitely pretty niche (I’d say its most useful to those who have studied lots via Heisig and live in Japan and have to deal with paperwork themselves), but as with most things, it’s hard to know how niche until it hits the real-world.
There is an Android version and it’s linked in in the original post.
Thanks again for those who have given ideas and actionable feedback!
