using Plecodict with one hand?
- Thread starter mxyzfrank
- Start date
If you put your phone on a table or some other stand / rest, yes. But one-handed keyboard operation as on a mobile phone isn't really possible with our software, you need to be able to use the stylus to tap on buttons / draw characters / etc, so you'd need either a second hand or something else to hold up the phone for you.
Really hope you'll still be implementing setting up the buttons for non-character/pinyin-writing functionality, Mike, especially one-handed navigational look-up in the text reader - you know, use the navigational buttons to flip to the next character/compound or jump up/down a line. It's just that I use Plecodict to read docs for 2-4 hours a day most days and try to fit all the work in every chance I get; I've dropped the stylus into some awkward places while standing on moving buses/trains reading something in Documents To Go using the Pleco DA; when I'm eating, have to shovel food in and pause while I use the stylus to select another compound/character I don't know or add it to a flashcard list; Um, there are other places too, but hopefully that's enough for now. 
Character lookup 2-handed? Yeah, that makes sense, can't do too much about that one. However, one-handed reading would make me happy as a pig in poo.
Character lookup 2-handed? Yeah, that makes sense, can't do too much about that one. However, one-handed reading would make me happy as a pig in poo.
We weren't planning on doing so, but I'm thinking about how this might work... the big problem really is that our text segmentation algorithm in the initial release of 2.0 is going to be passable at best (they're incredibly difficult to get working well, stacks of CS papers have been written on the issue) so you could frequently find yourself in a situation where it's looking up the second character of one word along with the first character of another; easy to fix with a stylus, but considerably harder with a one-handed keypad. I suppose we could have some sort of mode-switch button that would switch from word- to character-based navigation, but that might not do that much to help with the frustration. Anyway, we'll think about it and see if we can figure out a way to make it work well, but no promises.
Thanks Mike
I absolutely understand the problem with text segmentation and I know that you're interested in eventually doing something clever on that front. Also, I'm just talking here about the built-in text reader in it's simplest form without split-screens or anything fancy, not a Pleco-wide one-handed routine.
As to text segmentation, I have seen and used this in a number of other apps. However, when I think about it, I get by quite happily just now without text segmentation, working out which characters to select myself. In fact, feasibly a text segmentation algorithm would be something intermediate and advanced students would prefer to switch off so that they could better develop their own reading ability, or simply because the user would go through a more convoluted process to select the text that they want. I say that, but I have no doubt that it has its uses: it's a great thing for beginners and those with weak reading skills and of course you will likely be using it for the automated hyper-linking of dictionary entries, which would be very nice.
As to simply reading a Chinese document in a text reader though, on the whole I can read quite well and I'm pretty accurate with how to segment the text. Using Word2Go/Pleco DA to read, I occasionally quickly check if a 2 character word I recognise is part of a larger word by using the stylus to select an extra character or 2. Also, when I find, say, a 4-character idiom, I usually try dividing it up into 2 + 2 characters and then individual characters to see what constituent meaning I can derive. I imagine that most intermediate and advanced students probably wouldn't need to depend so much on an automated text segmentation algorithm. Even in the case of looking up some of the long names for organisations (E.g. The Chinese government's "State Environmental Protection Agency"'s "Policy Research Centre for Environment and Economy", if indeed it ever appeared in any C-E dictionary would be over 20 characters long), I would probably highlight it for later googling rather than rely entirely on Plecodict.
However, when I really think about how I actually use the current system (DocsToGo with Pleco DA), I think a nice one-handed implementation would simply >replace what the user does with the stylus in the current 2-handed way with the use of the 4 navigation buttons, select button and a couple of other buttons to control the width of the selecting cursor<.
So, let me revise the original 1-handed request to something a lot simpler/workable. What would be nice for those who don't need segmentation in the text reader (intermediate to advanced students) is to be able to turn it off and use the following navigation and lookup controls:
First of all, have the cursor always default to 1 character width rather than 0 characters in Word2Go, so something is always selected, whether it be a Chinese character, multiple Chinese characters or something else, like white space or punctuation.
Then have the following buttons:
* Basic 4 navigation buttons: up (one line), down (one line), left (one character), right (one character)
* Select button: lookup currently selected character(s). If only one character is selected, either (depending on previous user selection or another MODE button like you suggested) lookup that one character or lookup longest compound that can be made together with the following characters in any of the dictionaries. Use a lookup algorithm similar to the current Plecodict lookup (except maybe with the prioritising of dictionaries that you talked about implementing before - e.g. Set up Oxford to always be the default and if it fails to return a perfect result, fall back on ABC or whichever dictionary is next in the ranking).
* Other user-assigned button: Augment cursor width by 1 (while holding the same first character position).
* Other user-assigned button: Shrink cursor width by 1 (while holding the same first character position), unless it's already of width 1, in which case, do nothing.
Also, obviously, if the cursor is set to a width of more than one and one of the navigation buttons is pressed moving the cursor's starting character, then the cursor width should reset to one again.
----
I hope you know what I'm now suggesting. To illustrate, If I saw 3 characters I'd like to check in the middle of a text, I would just press and hold the nav buttons to get the cursor on the first character, press the augment cursor button twice to select all 3 characters and then press the select button to lookup. Bingo! No stylus needed. Also, because few words in the dictionaries are over 4 characters long, the user wouldn't be pressing the augment/shrink buttons a ridiculous number of times.
I honestly think this would be an excellent one-handed method for reading documents using PlecoDict, without getting tangled up in text segmentation issues. Really just a simple replacement of the current stylus selection process.
I think it would be particularly worth doing simply because once the student of Chinese really gets into his/her studies, he/she will likely be reading a lot and using the PlecoDict text reader to do so. Therefore they could be using the Palm one-handedly without special need of a stylus to write a character for a while. I already spend at least a couple of hours a day reading documents in Word2Go using the Pleco DA. During that reading process, I only occasionally flip back to the main program to examine a character in more depth. So if there was a one-handed operation, I could probably go quite a long time without needing to use the stylus to access the more complicated HCI stuff, like entering a character or selecting a dictionary hyper-link.
I'll be honest, I'm not a huge fan of using the stylus unless it's absolutely necessary (E.g. Practising characters). It's a big HCI hassle; it's easily droppable and if I get so much as an itchy nose, I have to stop my reading and find a safe place to put the stylus where it won't roll away. I fell down a few times on the bus in Taiwan stupidly using it when standing (and once dropped it in someone's shopping basket). When I got my first Palm 5 years ago, I ended up going through 2 three packs of styluses because I just kept losing them (or absent-mindedly chewing the ends). Besides, sometimes I like to lean my chair right back, have a beer or coffee reachable with one hand and study Chinese on my Palm in the other. Of course, every time I need to look up a word, I have to put stuff down, get into a safer sitting position and use the stylus.
I think old Steve Jobs feels the same way, hence his almost emotional stressing of the doing away with any stylus in his iPhone.
All the best Mike
Ben
I absolutely understand the problem with text segmentation and I know that you're interested in eventually doing something clever on that front. Also, I'm just talking here about the built-in text reader in it's simplest form without split-screens or anything fancy, not a Pleco-wide one-handed routine.
As to text segmentation, I have seen and used this in a number of other apps. However, when I think about it, I get by quite happily just now without text segmentation, working out which characters to select myself. In fact, feasibly a text segmentation algorithm would be something intermediate and advanced students would prefer to switch off so that they could better develop their own reading ability, or simply because the user would go through a more convoluted process to select the text that they want. I say that, but I have no doubt that it has its uses: it's a great thing for beginners and those with weak reading skills and of course you will likely be using it for the automated hyper-linking of dictionary entries, which would be very nice.
As to simply reading a Chinese document in a text reader though, on the whole I can read quite well and I'm pretty accurate with how to segment the text. Using Word2Go/Pleco DA to read, I occasionally quickly check if a 2 character word I recognise is part of a larger word by using the stylus to select an extra character or 2. Also, when I find, say, a 4-character idiom, I usually try dividing it up into 2 + 2 characters and then individual characters to see what constituent meaning I can derive. I imagine that most intermediate and advanced students probably wouldn't need to depend so much on an automated text segmentation algorithm. Even in the case of looking up some of the long names for organisations (E.g. The Chinese government's "State Environmental Protection Agency"'s "Policy Research Centre for Environment and Economy", if indeed it ever appeared in any C-E dictionary would be over 20 characters long), I would probably highlight it for later googling rather than rely entirely on Plecodict.
However, when I really think about how I actually use the current system (DocsToGo with Pleco DA), I think a nice one-handed implementation would simply >replace what the user does with the stylus in the current 2-handed way with the use of the 4 navigation buttons, select button and a couple of other buttons to control the width of the selecting cursor<.
So, let me revise the original 1-handed request to something a lot simpler/workable. What would be nice for those who don't need segmentation in the text reader (intermediate to advanced students) is to be able to turn it off and use the following navigation and lookup controls:
First of all, have the cursor always default to 1 character width rather than 0 characters in Word2Go, so something is always selected, whether it be a Chinese character, multiple Chinese characters or something else, like white space or punctuation.
Then have the following buttons:
* Basic 4 navigation buttons: up (one line), down (one line), left (one character), right (one character)
* Select button: lookup currently selected character(s). If only one character is selected, either (depending on previous user selection or another MODE button like you suggested) lookup that one character or lookup longest compound that can be made together with the following characters in any of the dictionaries. Use a lookup algorithm similar to the current Plecodict lookup (except maybe with the prioritising of dictionaries that you talked about implementing before - e.g. Set up Oxford to always be the default and if it fails to return a perfect result, fall back on ABC or whichever dictionary is next in the ranking).
* Other user-assigned button: Augment cursor width by 1 (while holding the same first character position).
* Other user-assigned button: Shrink cursor width by 1 (while holding the same first character position), unless it's already of width 1, in which case, do nothing.
Also, obviously, if the cursor is set to a width of more than one and one of the navigation buttons is pressed moving the cursor's starting character, then the cursor width should reset to one again.
----
I hope you know what I'm now suggesting. To illustrate, If I saw 3 characters I'd like to check in the middle of a text, I would just press and hold the nav buttons to get the cursor on the first character, press the augment cursor button twice to select all 3 characters and then press the select button to lookup. Bingo! No stylus needed. Also, because few words in the dictionaries are over 4 characters long, the user wouldn't be pressing the augment/shrink buttons a ridiculous number of times.
I honestly think this would be an excellent one-handed method for reading documents using PlecoDict, without getting tangled up in text segmentation issues. Really just a simple replacement of the current stylus selection process.
I think it would be particularly worth doing simply because once the student of Chinese really gets into his/her studies, he/she will likely be reading a lot and using the PlecoDict text reader to do so. Therefore they could be using the Palm one-handedly without special need of a stylus to write a character for a while. I already spend at least a couple of hours a day reading documents in Word2Go using the Pleco DA. During that reading process, I only occasionally flip back to the main program to examine a character in more depth. So if there was a one-handed operation, I could probably go quite a long time without needing to use the stylus to access the more complicated HCI stuff, like entering a character or selecting a dictionary hyper-link.
I'll be honest, I'm not a huge fan of using the stylus unless it's absolutely necessary (E.g. Practising characters). It's a big HCI hassle; it's easily droppable and if I get so much as an itchy nose, I have to stop my reading and find a safe place to put the stylus where it won't roll away. I fell down a few times on the bus in Taiwan stupidly using it when standing (and once dropped it in someone's shopping basket). When I got my first Palm 5 years ago, I ended up going through 2 three packs of styluses because I just kept losing them (or absent-mindedly chewing the ends). Besides, sometimes I like to lean my chair right back, have a beer or coffee reachable with one hand and study Chinese on my Palm in the other. Of course, every time I need to look up a word, I have to put stuff down, get into a safer sitting position and use the stylus.
I think old Steve Jobs feels the same way, hence his almost emotional stressing of the doing away with any stylus in his iPhone.
All the best Mike
Ben
1-character cursor width makes a lot of sense in a document reader, yes - if you're not inserting text then there's no point to a narrower one. But we could actually streamline this further if we automatically looked for the longest possible word beginning at the current cursor position and searched for that - there'd be an easy way to shorten it, but in general looking up 1 character doesn't do you a whole lot of good so there's no need to make that the default behavior.
But it is good to be thinking about this, I guess, particularly since Microsoft is supposedly going to make yet another boneheaded decision with the release of Windows Mobile 6 - the current "Smartphone" will now be called "Windows Mobile Standard," while the Pocket PC versions will be known as "Classic" and "Professional," thus making it virtually impossible for us to explain in one sentence which Windows Mobile devices our software will and won't work on. Which will be one more reason for us to make a version for touchscreen-less Windows Mobile devices. (but that's still #2 on our priority list after desktop Windows)
But it is good to be thinking about this, I guess, particularly since Microsoft is supposedly going to make yet another boneheaded decision with the release of Windows Mobile 6 - the current "Smartphone" will now be called "Windows Mobile Standard," while the Pocket PC versions will be known as "Classic" and "Professional," thus making it virtually impossible for us to explain in one sentence which Windows Mobile devices our software will and won't work on. Which will be one more reason for us to make a version for touchscreen-less Windows Mobile devices. (but that's still #2 on our priority list after desktop Windows)
Yeah, I was basing it on how I currently use Word2Go with the Pleco DA - I press a button to lookup the dictionary after I have selected text. I think you are looking instead to have the doc reader looking up the dictionary at all times, so I absolutely agree with you on the default behaviour. I'm curious to know how you'll allow the user to break entries down (Eg. she4hui4 zhu3yi4 into 2 entries) but I have to say that Pleco's UI has to be the finest I've seen in any Palm app and you'll no doubt work it out.
By the way, will the Pleco DA still be extant in the 2 release?
Cheers!
Ben
By the way, will the Pleco DA still be extant in the 2 release?
Cheers!
Ben
For the breakdown, we'll probably have some sort of a user-assignable shrink / expand button command (accompanied by an onscreen arrow control) - with she4hui4 zhu3yi4, you'd press this once and it would drop to she4hui4, and if you then pressed the "go to next word" button it would give you the definition for "zhu3yi4."
The Pleco DA will most likely still be in the 2.0 release, yes; really it's about 20 lines of code, all it does is locates the PlecoDict application file and launches it with a code telling it to start up in Instant Access mode. And since we can't offer the array of launch/invocation methods that you'd get with a dedicated application/DA launcher program it makes sense to continue offering an interface to that.
The Pleco DA will most likely still be in the 2.0 release, yes; really it's about 20 lines of code, all it does is locates the PlecoDict application file and launches it with a code telling it to start up in Instant Access mode. And since we can't offer the array of launch/invocation methods that you'd get with a dedicated application/DA launcher program it makes sense to continue offering an interface to that.