3.2.x Bug Report / Feedback Thread

[mikelove]

Parent categories can contain cards themselves that aren't in any subcategory, so if we supported that it would be impossible for people to select only cards in those subcategories and not in the parent. (not everyone uses categories strictly hierarchically - lots of different ways to organize flashcards)

But we could probably make it more obvious that the parent is not selected, yes.

 

[wibr]

Or maybe auto-select the parent category if there are no cards and all subcategories are selected? Anyway, now that I know the trick it's not an issue anymore for me.

 

[ACardiganAndAFrown]

 

[mikelove]

@wibr - the problem there would be inconsistent behavior; it's not obvious to users why that's happening sometimes and not other times.

@ACardiganAndAFrown - thanks.

 

[rizen suha]

feature request: in "components" tab for hanzi, option to show (always, for any hanzi, accessed fr anywhere) complete definition for each component (without need to click on the component to go to dictionary list view of definition/s). since only one (complete) defintion could be shown, let this be the first entry of said list view (contains multiple readings in multiple dictionaries).

 

[mikelove]

Sorry, what do you mean by 'complete definition' here? (as compared to what's there now)

 

[rizen suha]

...that the defs for components are often one-liners, seemingly belonging to no specific (available/explicitly published within pleco) dict. also, the "componentization" of hanzi is sometimes faulty, maybe originating from that same "unpublished" dict?

 

[pdwalker]

...that the defs for components are often one-liners, seemingly belonging to no specific (available/explicitly published within pleco) dict. also, the "componentization" of hanzi is sometimes faulty, maybe originating from that same "unpublished" dict?

Can you post some screenshots showing some examples that makes it clear which entries have these issues?

 

[mikelove]

They actually come from UNI; it tends to have short definitions that cover every pronunciation for a character, whereas with any other dictionary we'd need to awkwardly lump multiple entries together for a lot of characters.

 

[rizen suha]

one example
1) does not list meaningful components in tab
2) and the components have very short defs (here shown the long def, by clicking short def)
3) a much more meaningful component, by explicit dict lookup

 

[rizen suha]

@mikelove, UNI is the worst, when it is at its best it seems to be copying KEY
i find that thinking in components is absolutely vital (to me at least) for building a hanzi and word vocab
therefore, more meaningful componentization and (option for) eliminating a click to access "full" defs, would be very cool
...associating everything to meaning and understanding is the only glue to make it "stick"

 

[mikelove]

On 1, yeah, it seems like we should include 㥯 as a component there as well. I believe in this case KEY actually got its definition from UNI rather than the other way around (though I don't know where UNI got it from).

I'll defend the short definitions for components in general because the point of that screen is trying to give you a hint about how the original character came together rather than teaching you a bunch of other characters - if you want to drill into a character in more detail you can tap on the one you want to get a longer definition, but I think by choosing definitions that are too long / difficult to parse we actually harm that process
of understanding. Your own example with 工 testifies to that - does knowing that 工 is the third note in the gongchepu scale help you at all in understanding the meaning of a character containing 工?

Actually my ideal for that screen would be to come up with even shorter one- or two-word definitions that give you the most general 'sense' of a character, but I don't know of a good resource for those. For learning-oriented character breakdowns you could buy the Outlier dictionary add-on, which not only provides more accurate / carefully-checked breakdowns than our free character breakdown feature but also highlights their etymologically-relevant meanings.

 

[rizen suha]

outlier dictionary... let me defer comment on that to in a year or so...
on the issue of usability
i know like 500 common components by heart,
with long defs in a single tab overview, it will cost me no effort to swipe past them
but CLICKING (forth and back) on the occasional dingy i do not know/remember, is PAINFUL
when you are totally new, ultra short defs may be good
thereafter, options for "expert view", please
on a slightly different note (somewhat related to outlier):
consider user contributed componentization (and other stuff) "voted" via user authority and coincidence of contributions
as always, thanks for your great tool and fast replies

 

[mikelove]

It doesn't cost effort to swipe past them, it does cost effort to read through them if you don't know them. I appreciate that this is something that you'd personally find useful, but I hate adding off-by-default options like your proposed 'expert view' unless a lot of people would use them and I'm not seeing a broad use case for this yet.

 

[rizen suha]

ok, thanks
just put it into the box of considerations then
ps: yes, that is what i meant, an "expert" (which i am certainly not) knows what he is looking for and is able to ignore what not, without effort (or "consciouness")
somewhat like being able to pick up your own name in a boisterous crowd

 

[chemis3phy6]

Why some of the word does not contain tones like this one ?
and there is another entry with tones?

 

[mikelove]

Sorry, are you referring to the fact that 'liang' here has a neutral tone? A lot of words are pronounced with neutral tones in the PRC but not in Taiwan.

 

[Daniel C]

I just noticed that the examples in the PLC entry for 高雅 consistently render 雅 as 压.

 

[rizen suha]

please let pop up be bigger. why should i have to scroll inside the popup here? samo for popups in reader where it could at least be +50% bigger (on iphone x, on xs max +75%) and so on... thanks

 

[mikelove]

@Daniel C - thanks!

@rizen suha - In the works, thanks.