At the moment the level 6 radical angel has a barb ホ just like the katakana ho. However none of the kanji that it’s listed as being in have barbs. They have a cross (or a stick): 茶 術 述
So why should angel? Now that there are katakana words like ホテル in the mix, wouldn’t it be a good idea to make the angel radical a cross with fins as it appears in the kanji? Might it also be worth giving kana only words a different colour background banner to kanji-based vocabulary?
That’s just the messiness of kanji meeting the messiness of WaniKani radical choices.
In the case of 茶 you’re right that it usually doesn’t have the barb, but the component the kanji uses is 余 and you’ll notice that this one does have the barb for some reason when it’s written as a stand-alone kanji. Also note that Wanikani treats this as a separate radical that’s introduced much later: https://www.wanikani.com/radicals/surplus and you also have the kanji version that uses the “jackhammer” radical: https://www.wanikani.com/kanji/余
For 術 and 述 the component is actually the unrelated 朮 which does indeed not have the barb but does have the little apostrophe-looking stroke on the top-right which WaniKani decided to treat as a separate “drop” component.
As for katakana ホ it comes from 保 which… does not have the barb, although apparently the original form of the kanji once used a form of 子 which does. And of course the modern version is written like 木, not ホ.
It’s one of these situations where I wonder if it’s even worth fixing in WaniKani. This radical is only useful to remember the meaning and reading of the kanji using the WaniKani-provided mnemonics, for any sort of more rigorous study of these kanji it’s effectively inappropriate and inaccurate.
Also I don’t know if there’s any unicode character that would look like what you describe, which would force them to move back to SVG radicals for this one. I could be wrong though.
As interesting as that breakdown was, I’m not sure what most of it meant. Is a component a kanji as part of another kanji? Doing levels six and seven I’m very much the beginner, and I really don’t understand what 余 as a stand alone kanji has to do with the issue. I expect you are looking at this from a highly experienced point of view and are seeing connections I don’t and probably most other newbies won’t. I can spot a tree in your answer somewhere… I get that there may be a question of using unicode vs graphics for a radical, but I admit I have no idea why that might be a big deal.
I’m just looking at where Wanikani is teaching the angel radical appears, (not jackhammer, not surplus) and that’s solely in kanji with no barb.
For me, there’s also a practical question of how many newbies will stare at katakana trying to decipher a kanji (or the opposite). I’m not chuffed about purely kana words being mixed in the middle of vocab lessons, adding to the cognitive load, but more kana is the way the wind is blowing and I think it would be easier if the banner was differentiated by colour. Why not also also think about differentiating manufactured radicals that look like katakana. I expect most of these will be unavoidable, but why would that mean leaving the ones that are avoidable?
I see where you’re coming from, I’m just pointing out that the situation is really messy and rather arbitrary. WaniKani’s radical decisions here are not based on etymology or even established traditions but rather motivated by level progression and availability of Unicode characters representing the radical they want.
I do agree with you that it would make more sense for WaniKani to use ホ without the barb in this situation but to my knowledge there is no such character in Unicode, so they’d have to use a custom radical which is something they’ve been actively moving away from lately, probably because it creates consistency issues in the display.
Alternatively they could use 十 or 朮 instead, but that would require changing the kanji decomposition and all related mnemonics.
So overall I see why they decided to allow this small discrepancy.
