I was wondering if, in the iOS app, a list could be built of all the words that have been scored incorrect. This list would be purely for studying from, and would not affect the score.

Suggest a few options to 'clear'' the list:

1) Manual Clear (ie. just keep adding to the list until user clears it by pressing a button)
2) Auto clear if words were last reviewed more than X hours ago (ie. 24 hours)
3) Clear word if user gets it right while reviewing the list.

For reviewing the list (kind of like scratchpad) it would be nice to have one or more of the following modes:
1) Loop continuously, but remove correct words. Stop when all words are correct (because they're all removed)
2) Loop continuously until stopped.
3) Run once through the list. User could run through it again if they want.

The reason I request this is for efficiency of study:

Assume, for example, that I have 200 characters to review, and I don't know 20 of them. I spend 10 seconds per word I know, and 5 minutes studying a word I don't know. Reviewing the 180 characters I know, at 10 seconds each, requires half an hour. Studying the characters I don't know would take me 1 hour 40 minutes. Also assume I only have one hour per day to review.

Scenario 1: Now if I pause at a character I don't know while doing a normal Skritter session to spemd time studying it, there's no way I'll get theough the whole list and the list will pile up day after day. This happened to me last week, and I got up to 900 reviews due (I partially ended up clearing it by just marking wrong what I don't know and not studying the wrong ones anymore).

Scenario 2: On the other hand if I could burn through the reviews, knowing I could more thoroughly study the characters I don't know later on, I could clear my queue much easier, properly recalling the other words at the designated SRS interval. Even if I didn't have time to review all the characters I got wrong that day I'm a lot better off than I was in Scenario #1.

Thoughts?

I think that the better solution is to not spend five minutes studying a word you don't know. If you use the SRS as intended, you'll solidify your memory of that word much more time-effectively than if you just drill it for five minutes at once.

Now, one exception to this would be if you're making a mnemonic for the word. Then spending a little more time at first can lead to improved total efficiency. But making mnemonics shouldn't take very long, either.

Well, it's true. Drilling the same thing over and over for 5 minutes would not be very productive :)

When I say 'study' I mean:
1) make a mnemonic
2) look at usage in sample sentences
3) look at grammatical function
4) compare and contrast with similar words/characters
5) look at what words a character is used in
6) find keywords in my heisig book for forgotten components

An alternative I'd be pretty happy with is the feature you mentioned about listing the most leechy characters/words ( http://www.skritter.com/forum/topic?id=207862136&comments=7 ), especially if this could also be shown in the iOS app.

If you wanted to do something like this, why don't you just get another SRS program, such as Anki, and use it solely for this function?

Think of Skritter as your overall goal, but then use the SRS program when you want to spend a little extra time on something specific.

@Seant018, Because Skritter is awesome! :)

Seriously though, I have used Pleco SRS for reinforcing new vocab when I'm first learning it, but there's no easy way to identify/export which words I'm 'bad' at in Skritter.

I do this sort of thing too, so for now I star these tough words as they show up, so that I can subject them to extra study and analysis away from Skritter.

It would be nice if I could sort my whole word list by time spent on a word, % correct, and maybe some calculated aggregate 'score' over all components for that word. Many things in this world would be 'nice to have'!

Each sort like that which we add is very expensive, having to add another index copy of all of the item data, and making it slower every time a review needs to be saved. So we try to keep those sorts to a minimum: only the most useful.