Common feature set?

[omängum fra'uti]

Of course, it just needs to be done.  The schema just needs a re-design to support it.

Muzer, probably not really a need for that.  It isn't like it's updated very often, and NeotrekkerZ has been very responsive in the past, I'd can't imagine he would have a problem publishing a mobile friendly version such as HTML along side the regular version.

[Muzer]

It was a fun thing for me to make during a boring lunch break though, I didn't actually expect it to be used at all

[omängum fra'uti]

I got a chance to play with the Android version today, pics to follow once I get them off my phone .  Biggest thing I noticed was the view transitions.  When switching from the main screen to the resources screen, it should be a simple slide from one to the other.  I don't know if you noticed this or not, but the images of the main screen and the resources border each other so that it can look like the camera just slides over in the scene.  Next, when transitioning from the resources screen to either About or Disclaimer pages, the view should flip along the central vertical axis.  The background should show this as well since its the same from one to the other, but flipped vertically.  Next, in the dictionary it should be Na'vi > 'ìnglìsì then English > Na'vi.  Other than that, its looking pretty cool.

Ask, and you shall receive eye candy!  It turns out anything you can do with matrix transformation can be used as an animation, with the right amount of coding.  So a flip is fairly simple.  Sliding doesn't even need a custom animation for, just define the parameters.  Both are now in.  Also adjusted all buttons to 75% transparency.

All that's missing now from the Android version to be comparable is jumping to a specific letter and filtering by part of speech.  Is the part of speech filter a "one at a time" (Or all) thing?  Also, what does the refresh button on the dictionary page do, is it related to search?

(And if anyone else is going to work with the Android source...  I hope your Na'vi is good...  One of the files is, sort of, mostly commented in Na'vi.)

[Seze]

All that's missing now from the Android version to be comparable is jumping to a specific letter and filtering by part of speech.  Is the part of speech filter a "one at a time" (Or all) thing?  Also, what does the refresh button on the dictionary page do, is it related to search?

I'm not exactly sure what you mean by one at a time for the part of speech filter.  All it does is removes any entries that don't match the filter setting.  The part of speech filter is also customizable by the end user.  In the Settings App (holds all the settings for your Apps on iPhone) you can pick what each section in the bottom bar does, except for the "All" section.

The "refresh button" is not a refresh button.  Its the button that changes the dictionary from English -> Na'vi to Na'vi -> 'ìnglìsì and back again.  This transition is also a vertical axis flip.

[Seze]

(And if anyone else is going to work with the Android source...  I hope your Na'vi is good...  One of the files is, sort of, mostly commented in Na'vi.)

Once I get a better grasp of the language, I think I will convert the iPhone comments over as well...

[omängum fra'uti]

It's difficult, there are a LOT of lacking vital concepts.  I only did one of the smaller classes.  (It handles checking for an updated DB version...  Or as the comment says "Piak si tsaheylu URLur" or something like that.)  I tried another but gave up as the result becoming unintelligible.

[Sіr. Ηaxalot]

And people is whining on me because I write Swedish comments. Bad habit they say

[omängum fra'uti]

Seze, do you have anything particular in mind for the learning tools?  I noticed there are buttons for practice and phrase books...  I was thinking about taking a stab at that once I finished the dictionary screen, but if you had something in mind, it would be good to start with that.

[omängum fra'uti]

Probably more of an Android  feature, but Na'vi word of the day!  I'm thinking a widget on Android that updates once a day.

Ideally the word should have full picture, pronunciation, and sample usages to accompany it.

For short term, perhaps just display a random word and meaning, occasionally changing.

[Muzer]

YES, do that!

[Seze]

I haven't really done anything on the Learning Resources front.  I'm open to pretty much anything.

[Sіr. Ηaxalot]

Are there any plans on filling up the main screen with other resources/features?

The way it is now it feels a bit unnecessary to separate the pages when the main page only have a link to the dictionary and to the other resources.

[Seze]

Sran, I was planning on putting more there at some point.  At one point there used to be a "Phrase Book" button and a "Practice" button that I had on that page, but they were nothing more than UI filler until I figured out what I really wanted there, though I will be adding "Na'vi in a Nutshell" to that page once I get that guide rebuilt for mobile devices. 

[omängum fra'uti]

There are so many ways to approach a learning material...  But I'm a HUGE fan of what MemoryLifter does.  It's a spaced repitition method based on a series of "card boxes".  So here's my basic idea...

User selects learning mode...
Selects question & answer type
Available question types:

• Spoken
• Picture
• Na'vi word
• Definition (In language of choice)

They can pick spoken or no, picture or no, and Na'vi word, definition, or neither.  As long as at least ONE thing is selected.  (Obviously this works best with a good pool of spoken and pictures.)

The answer type can then be any ONE visual method not part of the question.  (So no spoken answers...  Unless someone wants to start writing Na'vi speech recognition and formant analysis...)

For simplicity, the form of the question would probably have to be limited to true/false or multiple choice.  (This would probably be a VERY good application of Levenshtein search to find similar words to present as well.)

Finally... the spaced learning.  The way this works in Memory Lifter is as follows...

All flashcards start out in a pool.  One is drawn from the pool at random and asked.  If it is answered correctly it is put into box 2.  An incorrect answer (From anywhere) always goes into box 1.  Each box has a limited number of spaces, and any time a box is overfull, the front card is picked and asked.  Each correct answer promotes a card, each incorrect sends it back to box 1.  If all of the boxes are within their limits, a new card is pulled from the pool.  By this means, the words you have trouble with you see often while the words that are easy you don't spend much time with.  It's a fairly simple algorithm, but does require some tracking.

In memory lifter the box sizes are something like 10, 25, etc and quickly escalating.  I'll see about throwing together the groundwork for this when I get a chance.  Though having a library of pictures and spoken words would greatly help.

[Muzer]

Swinging quite offtopic here, but I wonder if we could make a Na'vi TTS program? The pronunciation is regular save very few exceptions, so it wouldn't be hard to parse - the only issue is getting recordings that sound good and at least vaguely natural