PlecoDict 2.0 Suggestions

[The Duelist]

Hi Mike

:idea: I have a Suggestion for PlecoDict 2.0
In the New PlecoDict 2.0 Database, Would it be Possible to:
Link Dictionary Entries (Existing/User) to User Defined MP3 Sound Clips.
These Could then be Used Within Flashcards to Test Aural Comprehension & Hanzi Dictation.

Eagerly Awaiting v2.0 8)

The Duelist

 

[mikelove]

We've gotten a few requests for this, it's certainly something we might consider (particularly if we had a way that users could share audio samples they'd created - we could build up a nice database of unusual accents / words / etc) but actually implementing it would be rather complicated - we'd have to develop a user-friendly database conversion utility just for sound files, and that would get us into all sorts of hairy tech support issues with audio codecs, malfunctioning sound cards, etc. So this may not happen until a 2.1 or 3.0 release.

 

[caesartg]

I was asked to give a short talk on PDAs and using them to study Chinese to the 1st year students. Although I didn't quite turn into a salesman, Plecodict was the cornerstone of my talk about why they should invest in a PDA and how they could use one to help them on the course. They will all be doing the same course and have the same demands on them by the material and method of studying, so I thought it would be worthwhile to give a small indication of the kinds of issues that come up with their learning each year, in case it gives you some ideas for Plecodict 2.

Every 1st year, most students have huge problems with the tones and their niggling rules. They also have great confusion over stroke orders. I note the open-source Palm Japanese app, PADict, includes a fantastic little stroke order study tool. Such a function is perfect for Plecodict 2 perhaps?

2nd year students have come back from study abroad at a uni/college in either the mainland or Taiwan. It's probably not a big learning issue that PD could help with but students pick up a lot of Beijing/Shanghai/Hong Kong/Taiwan only slang and sometimes it's hard to know what's used in which place.

Another issue with a growing vocabulary is distinguishing what is spoken from what is only used in a written form (due to the vast number of hononyms that exist in the Chinese language making differentiation impossible when spoken and resulting in confusion). I know that the dictionaries in PD go some way to distinguishing but I've always felt that all the dictionaries I've ever seen don't do enough to help students with this.

Also, as I've already mentioned in another post, students begin to study literature and are expected to be able to paraphrase text by substituting their own vocabulary, which is where a thesaurus would be of use.

By the 3rd and final year, students' workload increases drastically and for just one of several modules, they can expect to have to get through 20-40 pages of Chinese a week. Difficult to balance with maintaining some pub life. I'll be entering my final year in the autumn, so I don't have personal experience of this yet.

Cheers

Ben

 

[caesartg]

Concerning the Plecodict DA, when I use Plecodict, I spend probably about two thirds of my time accessing it in other apps using the DA and only one third directly entering the app itself. I use the DA with DocsToGo, or some other text reader to read through some Chinese, and I maybe append to a list of source-specific flashcards as I go along. I spend a little time browsing a Chinese news site using Blazer and again use the DA for that, but apart from that almost all of my DA use is just in a text reader.

When we spoke before about Plecodict's DA and you mentioned the restrictions you work within for coding it on the Palm, I wondered if you might just be better having some simple Text Reader as part of Plecodict 2, able to read in Unicode/Big 5/GB text files. Then you would have greater control and bypass those annoying DA limitations. Also, as I noted another user queried you about the 'sweeping' dictionary lookup, seen in the desktop product Wenlin (I think?). You were receptive to the idea, but I really can't see how you could do that using a DA.

Another advantage of having a simple text reader is that users could also use that nice font-enlargement feature on the Palm that Pocket PC users seem to have built-in! Hope you can consider that one for PD2!

Oh yeah, and I use a Lifedrive and its hard drive access speeds really slow everything down, so if you had a text reader which could pre-load dictionary entries, my problems with the Lifedrive would possibly magically disappear (or maybe not).

All the best!!!

Ben Caesar

 

[mikelove]

Thanks for all the feedback. (and for spreading the word about PlecoDict!)

A lot of your feature suggestions are already on our 2.0 feature list (see http://www.pleco.com/pdpreview.html). Stroke order diagrams will almost certainly be included (we're still waiting on one contract but I'm pretty confident it'll be done soon) - I'm actually not convinced of their usefulness myself, but every other Chinese software package seems to have them and people keep asking for them so they're definitely something we want to include. (it's pretty much the same situation with audio)

We haven't yet gotten into the business of producing our own dictionaries (the audio recordings for PD 2.0 are the first original content we've developed) but you make a very good point about regional dialects and written versus spoken Chinese - I haven't yet seen a title that does a good job with this, and I'm hoping that post-2.0 we'll have enough money coming in that we can actually start to develop some original, Pleco-only Chinese learning content. (if we come up with a particularly unique/useful reference and only make it available with our software, it could pay for its development cost many times over)

On the reading front, the new document reader (we are indeed developing one) should help a lot with that, and for documents that aren't readily available in electronic form I think we already have a pretty great system, at least until it's possible to conveniently capture text on a PDA with OCR (which I'd estimate is about 5 years away - Chinese OCR is still far behind English, sadly). There will likely still be a DA feature in 2.0, and in fact it may even be possible to expand it a bit (there are some tricks we can play with the new ARM-accelerated database engine that might allow us to squeeze more code into a pop-up application), but certainly the user experience with a Pleco-created document reader would be considerably better.

Re the Lifedrive access speed, unfortunately I don't think we'll be able to spend a lot of time on anything Lifedrive-specific, since the Lifedrive didn't sell all that well and a 4 GB SD card can now be had for as little as $100 (making it very unlikely we'll see any more hard-drive-based PDAs in the future).

 

[koreth]

Stroke order

Stroke order features are really useful for beginning students. Until fairly recently I was constantly having to go back to my books to double-check my stroke order. Now I've learned enough characters that I get it right almost all the time without any conscious effort, but if PlecoDict can double-check my strokes when I'm drawing characters into the flashcard UI, I certainly won't complain. When you get it right, presumably you won't even notice that the software is checking it.

One thing I'll be curious to see is which stroke order PlecoDict enforces. I've found occasional minor differences in stroke order in different books, and once or twice my girlfriend (who's Taiwanese) has seen me copying a character out of a book and commented that she learned the strokes in a slightly different order. I wonder if it'll be possible to have more than one valid order for a character.

 

[mikelove]

I don't know that the stroke order checker will work quite like you describe - more likely (if it's included at all) it'll give you an outline of a character and ask you to tap on (or draw your stylus in, to indicate direction) the strokes in order. That lets us give much nicer/clearer feedback and makes it much less likely that it would be scored incorrectly.

Our system will likely only supply one order for any particular character, but as we're still finalizing our source for the data (it's pretty well settled, but I never consider a deal done until there's a signed contract in the company safe) I can't provide a definite answer on that yet.

 

[caesartg]

That sounds like an effective way to go about it. Personally, I don't need it but I really needed something just like that a year ago (which was why I found PADict's little Japanese stroke order doodle to be a god-send).

 

[mikelove]

Yes, I do like the way they implemented that - good way to get some extra functionality out of a handwriting recognition engine with a minimum of extra work, and the color feedback is nice and clear.