I have attempted both, but so far failed in either case. I think anki is doable once python-pyqt4 compiles. mnemosyne seems to be very nice as a program but uses some insane compilation setup that does not lend itself to cross-compilation. The author is not willing to use anything else.

For those who are interested, I have released version 0.5.0 of toMOTko.

The two most important features are the font selection in the Preferences dialog and limited copy/paste functionality. The latter only works for words. It's not possible to copy/cut/paste folders or glossaries yet. It should be done in the next release.

For Kurochka: the font selection feature should allow you to see special characters like French and Spanish accents if you have a unicode font installed on your system.

Feedback is welcome.

I found a new FlashCard program today. I haven't tried it yet but it looks interesting.

http://jvlt.sourceforge.net

It's made in Java so it probably works on the Zaurus.

mnemosyne is a python program - what insane compilation does your version do? Mine just copies python scripts to the python directory.

I'll try setting up python on my Z again and see if I can get ti to work - something to do on the train

Stu

I saw this post and realised that it could be the Unicode BOM (http://unicode.org/faq/utf_bom.html).

koan

Thank you, Frederic. I'll check it out and report. Font selection is the most very much appreciated. I also hope that "(un)select all entries" has made it into the new release.

I'm looking forward to hear from you.

I've added two buttons in the Glossary Manager, one to select all the entries and another one to inverse the selection. To uncheck all the words, you check them all and inverse the selection.

Next release (0.6.0) will offer the following :
- Support other languages (Not just English and Japanese)
- Multilingual interface (French, English and maybe Spanish and Japanese). Other languages will be added if users are willing to contribute.
- Optimized data format (xml->binary) => smaller (instead of taking 4.6MB, my glossaries now take 2MB) and faster (instead of 70 secs, loading my glossaries now takes 12 secs)
- And maybe the completion of copy/cut/paste feature but it can probably wait till version 0.7.0.

Thanks for your great work on toMOTko.

A feature I would like to see is, actually having a text field to write the kanji.
The handwritting recognition which comes with the japanese rom would recognise it; and tomotko would compare to the correct answer and give you the result.

It would give the possibility to train at writing kanji.

Thanks

Hi Matthis,

Thank you for using toMOTko.

I have given some thought about your suggestion. I'm not very enthused by it.

Even though I like that we can input kanjis on the Zaurus, handwriting them on the screen (and it's pretty much why I bought a Zaurus in the first place), it's not something that I enjoy doing very much. It may be good to look up for a word in a dictionnary once in a while but I don't think I would like to use it to write lots of words. Of course, this is subjective.

I also think that the default pattern recognition input method is not the proper tool to learn writing kanjis as it's smart enough to recognize kanjis even when they are not written correctly. For example, try to write "right"(migi) and "left"(hidari). The stroke order for these two kanjis is particular but the Zaurus will not complain about it if you miss because it knows that most of the people (including Japanese) may write them wrong. I know, it's probably no big deal for most people but it's something I find odd for a tool like toMOTko that aims to assist the learning of a language.

If other people ask for it, I may implement it later though. Just don't hold your breath...

For now, I would recommend that you write the kanjis with a good old pencil and paper when using toMOTko. That's what I do every day and it feels a lot better that writing on the Z's screen.

Hi

I made a simple flash card file for the Thai alphabet, it covers the 44 consonants. In Thai it is important to learn the class of a consonant so you can work out it's tone through a set of rules. When I first learnt Thai I used flash cards to go through the classes.

You can download the file here

Here is a screenshot

If you don't have a font with Thai glyphs, you can find one here, personally I prefer "Loma"

When I edited the XML file I changed the target language to "th" but after importing into toMOTko I couldn't see my text; so I had to change it back to "ja" even though this is wrong.

@Frederic: Maybe toMOTko could read the language from the XML and change the GUI to show which language it is ? "Japanese" should then become "Thai" and "Furigana" could become "meta explanation" or something. Also, choosing a suitable font for the language is good for us people learning more than one language.

I might do some Thai phrases/expressions later.

thanks!

P.S. Any random person listening, if you have any Kansai-ben or Osaka-ben; I'm quite interested in Japanese dialect. I can extract words and explanations from EDICT but it's no substitute for good examples.

Hi Koan,

Next release of toMOTko will support other languages. I hope to make it for the end of August or mid-September. I include some screenshots of the development version. You can see that you can specify a font for a particular language like you mentioned.

Screenshot 1 : From English to German
Screenshot 2 : List of study languages
Screenshot 3 : Font overrides by language

Thanks for your Thai file. When I release version 0.6.0, I will modify it to adjust to the new format of toMOTko and put it on the website (if you don't mind, of course).

Fred

P.S.: We don't see it in the screenshot but Thai language is also supported.

Hi Frederic,

Nice to see the screenshots of the new version. It's looking good.
I don't mind you updating my file. I will probably update it myself too.

Very glad to see Thai is supported in a language app

Hi Koan,

I tried to update your glossary file today but there is something I don't understand.

For each word, you declare 4 elements of data. Let's take the first one as an example :
1) gor gai
2) n (or something that looks like an 'n')
3) n ln (or something that looks like so)
4) chicken, middle class

What are these data? I guess that 4 is the English meaning. But for the rest, I'm not sure. The problem I have is : I don't know where to put this info in the new format.

In the new format (that is very similar to the previous one), I have these fields :

a) English word
b) Thai word
c) Thai pronunciation (or alternate format)
d) Comments/Examples (that can be expressed in English and in Thai)

So I feel that :

4) should go with a).
2) should go with b)
1) should go with c)
3) should go with d)

Am I right?

Or maybe 2) and 3) should be combined together into the same field b) and the comment field d) should be left empty?

I was flexible in the use of the available fields

This is a flash card set to help learn the Thai "alphabet" so it's not a normal flash card set with word -> word.

2. is the letter (there are only consonants in this file, I'm still trying to think of a sensible way to do vowels - it's not obvious). This letter is a little bit like a "g"
1. is a "romaji" version of "g is for gai" like "c is for chicken"; In English we make up any word for "A is for xxxxx" but Thai people learn specific objects relating to specific letters.
3. is the same as 1. but spelt in Thai.
4. are notes, gai = chicken, the consonant is middle class.



I see. It's not straightforward to find the best way to fit this to learning the Thai alphabet. Actually, I should add pronunciation: in addition to 1, Thai consonants change sound depending on if they are at the start of a word or the end.

The other thing is that there isn't a 1-1 mapping between 26 consonants of Roman and 44 consonants of Thai.



Um, I guess it should be

1.) a) English word
2.) Thai word
3.) c) Thai pronunciation (or alternate format)
4.) d) Comments/Examples (that can be expressed in English and in Thai)

Perhaps I should have made

1. chicken (g/k)
2. ก (ก ไก่)
3. gor (gor gai)
4. middle class

The g/k is the sound as an initial/final consonant.
Then for the quiz, the best order of revealing would be 2, 3, 1, 4.

Having ก (ก ไก่) combined in one item is not really ideal though because you want to learn that ก is related to ไก่...

Does it make sense ? I'm not sure this helps you define your new format. Any other questions, please ask!

I think it makes sense. Unless an unsuspecting problem occurs, I should release version 0.6.0 of toMOTko this week-end. I will also publish your glossary according to your explanations.

Thanks.

All these programs are GUI oriented. Does anybody of you know of a curses based flash program
(using the Leitner-methodology) ?

Detlef

OK, thanks. I will install the new version and check it when it is available.

The only one I know is latrine.

toMOTko 0.6.0 has been released. Take a look at the release note before installing.

Things I plan to do in the next release are :

- Finish implementing cut, copy and paste.
- Add key accelerators.
- Save/restore the state of the application (expanded tree nodes, quiz, etc.).
- Improved appearance.
- Optimize loading/saving data and tree updating.
- Begin porting to Windows/Linux.
- Etc.

Koan: I will let you update your vocabulary file yourself. After installing the latest version, you can export your glossary and then, edit it (replace 'ja' by 'th') and relocate the data in the appropriate fields (even though the format is not perfect for such a case).

@Frederic

I installed the latest toMOTko, thanks for your hard work.

I must admit the filter buttons at the top confused me; I didn't realise that I had imported my file so I imported a few times before I realised that I had to add the language and then set the filter to show it.

So I went to edit my tree.xml but as the imported file is number.xml it was difficult to know which files to remove.

When I switch the filter from English to Thai to Thai to English the Alt/Phon field is empty. I think this must be a bug ? It's the same for other languages.

I think the filter buttons should not be required: if you tap a English to Japanese glossary the right listview should update to English to Japanese. It might be good if there is a button to switch direction though.

A glossary folder can have a description but a glossary file doesn't. Can I suggest adding a description field to each glossary ?

Also, when I export a glossary, it retains the id. If I redistribute the glossary, as for example my Thai glossary, what id should I set ?

thanks

koan

You're right, it can be confusing (especially the first time). I plan to improve that but I wanted to make an early release. In the future, when importing a glossary, I will probably warn the user that the imported glossary contains words in more than 2 languages and/or when the languages are not among the selected Languages of Interest (in the Language Preferences).

This way, the user will be able to select the languages of the data he wants. And the Languages of Interest should be set automatically. About the Language Selectors (comboboxes at the top) I could set them automatically when the imported glossary is in 2 languages only but if it contains more than 2 languages, it's ambiguous and I better not touch it. I could set the Filter off though to make sure that we see the imported glossary in the TreeView.

Imagine for example that a medical glossary file has been translated in English, French, Spanish, and Italian and German, the user should be able to import only English, French and Spanish and omit Italian and German. Doing so should automatically select English, French and Spanish in the Languages of Interest Preferences if they were not already selected and the Filter should be turned Off.

Turning the Filter Off can have a confusing/surprising effect some times as you may not see the words in the right pane if the Language Selector is not properly set. This is a "feature" though as it allows a user to translate a glossary in a third or fourth language as described in the Online Help of the Glossary Manager.



I understand, especially now that the glossary files are encoded in binary. Of course, this file should not normally be edited :-) You could have removed all the imported glossaries using red cross button in toMOTko's GUI and import it again.



I'm aware of this. Actually, it's not a bug but rather a choice I made. As you know, the screen's real estate of the Zaurus is quite limited so I chose to display only an Alt field for the Target Language. If I had more space, I would have an Alt field for both Source and Target Language. Practically, we don't need so much an Alt field for the Source Language as this language is usually your mother tongue or a language you're very proficient with. I preferred to spend the space on the Comments field. So the language of study should always be set in the right Language Selector in order to see the Alt field.



Your idea is good but doesn't work for a Glossary containing translations in more than 2 languages like the previous use case I mentioned before.

The Language Selectors (as I call them) and Filter are very convenient when you're learning more than one language. If you're just learning Japanese, for example, you should leave the Language Selectors to English, Japanese and Filter On, and never touch them again.

If you're learning more than one language, the Filter is very convenient as the TreeView can become quite loaded. If you take a look at my screenshots, you will see.

The Language Selectors are not intended to be used as directional selector but they can be used as such if you don't care about the Alt field. I admit it can be confusing to get used to it. The proper way to set the direction of study is by checking Revealing Orders options in the Quiz Preferences.



I will think about it. Once again, it's a trade-off I made. The space is limited and I thought it was not essential to have a description for a glossary. Maybe I should add a button besides the title which would invoke a dialog containing Description, Author, Date and other metadata. There is enough room to fit a TextField just over the list of words but I'm not sure if it will be pretty, especially in Portrait mode.



It doesn't matter, the number will be ignored when importing the glossary. Instead, a new number will be generated in function of the state of the destination Glossary Manager. You can set it to "1".

Thanks a lot for your feedback. I will think about ways to improve the usability of the application.

@Frederic

I updated my glossary - lots of editing! I changed the layout as we had talked about.

One behaviour I think is a bug: if I start toMOTko and select English Japanese my Thai glossary shows up in the treeview and if I select it then the glossary view says "English" at the head of the first column and "Japanese" at the head of the second column, even though it is showing my Thai words.

thanks

koan

P.S. File and new screenshot at my Thai on Zaurus page - see my .sig

I imported your glossary. Thanks for the update. I will publish it on toMOTko's website shortly.

When you say that you see you the Thai words, I think you mean that you see your words, but only the English parts, in the English column, and you see a Japanese column that is empty, right?

If so, this is not a bug but a feature... Even though it's not an obvious one.

First, you mention that when you start toMOTko, you see your vocabulary while English/Japanese are selected. I have a few questions to ask : How do you start toMOTko? Do you really launch it from scratch or do you unsuspend the Zaurus and select the application (that is already opened)? I asked that because in my case, when I start toMOTko, the filter is On (in other words, the "Show all glossaries and terms" is unchecked). Could you tell me if the Filter is On when you start toMOTko?

If the filter is Off, your glossary should be visible even though English/Japanese are selected. You should then turn the filter On and the Thai glossary should disappear from the TreeView. I think that most of time (except maybe when you import new glossaries or translate existing ones), you should always turn the Filter On (i.e. "Show all glossaries and terms" should be unchecked). When the filter is On, you will need to set the Language Selectors to English/Thai to see your glossary. At the moment, the state of the Filter is not saved when we quit toMOTko. It's on the todo list.

Second, I explain why it's a feature and not a bug... As mentioned in my previous message (and in the Online Help), this feature exists in the case you would like to translate a glossary in another language. For example, if I have a animal glossary in English, Japanese and I want to translate it in Spanish, I will turn off the filter and select English, Spanish. In the term list (right pane), I will see the animals in the English column but the Spanish column will be empty. Then, I can select each word and edit it (pencil button). What happens in such a case is that I edit the "en" and "es" <translation> item of the terms.

So in the end, if I may express me in XML, you get something like that :



So you can learn the same words in different languages. Once a glossary have been translated (partially or completely), you can turn the filter On again.

If you select English/Japanese or English/Spanish or Japanese/Spanish, you will see the same glossary but the data will be displayed in the appropriate language.

This way, it's possible to distribute a glossary in several languages in only one file. That's something I intend to do to write a glossary of animals in English, French, Spanish and Japanese. I could also write a glossary about fruits, vegetables, verbs, etc. People will be able to import only the languages they want. For now though, all the languages are imported.

With this feature, you don't need to edit the XML file to translate an existing glossary, you can use the built-in editing facility of toMOTko.

Finally, I think that if I would put "Non available" or "Not defined" in red, instead of empty entries in the term list that could be easier to understand. What do you think?

Thanks for your feedback. Sorry for the long posts.

No, that's not quite right.

1. I run toMOTko by tapping the icon, it is not running in the background.
   
2. The filter shown is the English-Thai, only Thai glossaries show
   
3. I change the filter to English-Japanese, all glossaries show (including Thai)
   
4. Tap on Thai glossary - the glossary view shows English-Thai but the column heading for Thai says Japanese. Thai words are show, rendered with Japanese font.

Do you mean the filter is set to a language and not blank ?
Yes, as explained it is English-Thai.



Well, yes, but I'm not seeing empty entries.

I think it's a good idea to support glossaries that have more than 1 language as some of us are studying more than 1 language, but I think it's problematic for toMOTko's design as toMOTko is a flash card program not a dictionary. I think features like adding images might be more important.

thanks

When I talk about the filter, I mean the "Show all glossaries and terms" item in the Actions menu.
Filter is On if Show all glossaries and terms is unchecked.
Filter is Off if Show all glossaries and terms is checked.




Here are some screenshots of what happens when I perform these steps :

(SC1) Before first step : toMOTko is not started yet.


(SC2) After first step : toMOTko has just started.


(SC3) After changing to Eng/Jap. I don't see your glossary. It should appear after the Politics glossary.


(SC4) To see your glossary, I turn off the filter. I click the glossary and I get this :


In SC4, the result is as expected (as mentioned in previous posts).

So I'm unable to reproduce the problem. Could you send me a screenshot of the Z just after the last step and also a zip/tar file of your /home/zaurus/.toMOTko directory. Maybe it could be useful to reproduce the bug.



The idea to add images (and why not, sounds) is good. Thanks! It's not my priority but I may implement it eventually.

About the design of toMOTko, I agree it's not perfect but, hopefully, it can be improved.

@Frederic

I discovered the "bug": I had kept an old version of my glossary file (with the original en and ja markup). Naturally, toMOTko picked it up and renders as English - Japanese. So there is no problem. Except with me!

Nice screenshots, I'm glad you have installed my file!

thanks

That's a good news. Thanks for telling me.

I put your glossary on toMOTko's website in the Glossaries page.

Thanks, Frederic, for the new version.

A quick question: the "Word/Expr." box seems to be cut off at the bottom (it is smaller than the "Alt./Phon" box. I have played with the settings but it is still cut off (and the letters/characters in that box are cut off at the bottom as well). I'd post a screenshot but I forgot the key combination and the folder where screenshots go.

Long time no hear from you, Kurochka.

It should not be unless I misunderstand you... Look at the screenshots on the website. Does the Word/Expr field look cut off? Does it look the same on your side? Try setting different fonts in the Preferences. If it doesn't work, try restarting the application. In most cases, it should not be needed but It could help. I had such a bug once but I thought it was fixed.

To take a screenshot : Fn + Shift + C. The image file should be in Documents/Screen_Files on your SD card (most probably).

I use your program all the time.

Here is the screenshot. http://allyoucanupload.webshots.com/v/2001251220182324392

I use efont (because it has all the Unicode ranges I need in a nice size) for the field portion, and it does this.

I'm glad to hear that. I hope you like the multilingual support.



I'm able to reproduce the problem. It happens when using a too large font for the Label Fonts. I will try to find a fix in the week-end. Meanwhile, I recommend using a smaller font. I use unifont medium and the result looks ok. Maybe not as nice as efont though. I really like the style of japanese rendering in efont. Very cute. I still prefer lcfont for toMOTko because I find it easier to read.

@kurocha

I think I fixed the bug. It was a tough one. I spent 7 hours on it. I don't quite understand the problem actually but I have found a workaround that is not perfect, but hopefully, good enough. At least, with efont, the layout is fine. It will be available in the next release.

Thanks for reporting the problem.

anki has an emacs front end, but I haven't tried it yet

I'm currently desiging a leitner app as a local web server using a browser (curses if you like) as the interface, primarily for heisig, but should be flexible enough for general use

nice, but is the leitner? doesn't look like it from just a quick look at the man page...

Actually, I've never used it more than 5 minutes so I don't know. The man page mentions a continuously improved algorithm for word selection. I think you should give it a try. It was easy to use/configure if I remember well.

Thank you for your hard work!

I released version 0.7.0 of toMOTko. It's a little bit earlier than I thought but I wanted to make some fixes available for some annoying bugs.

Here is the change log :

- Fixed duplicate folder id bug.
- Improved layout update for QuizFrame when changing font.
- Added Language Selection dialog box for import/export.
- Shortened some strings in Spanish to make them fit in Portrait mode.
- Save the state of expanded/collapsed tree nodes in Glossary Manager.
- Display Undefined string in gray when a glossary contains no word in
a language shown in the list of terms in the Glossary Manager.
- Fixed lost selection bug when switching language or changing the state
of the filter.

@kurochka

Tell me if you can use efont properly with no cut-off fields in the Quiz.

I'm going to take 3 weeks off going back to my home country. Don't worry if you don't hear from me for a while.

Thanks, once again.

Yes, the new version does not have a problem with efont. If I am not mistaken the height of the windows has become static (does not depend on the size of the font) to keep efont happy. This have led to smaller examples window in the laptop/horizontal orientation. I can live with that, I will just scroll more if the example is large. Also note that with large font sizes the buttons are being eaten by bottom task bar. All this is not important. The hotkeys for yes and no would fix it

Will take a look. Thx for the hint.



Nice news! My main concern is about speed. I have yet to find a program as fast as real index cards, esp. if it comes to going over vocabulary you more or less know.
That's why I'm looking into curses instead of a "nice" GUI. Speed! Don't know if a browser based solution will cut it there.

Detlef

@Frederic

Hi, I have some suggestions for 0.8.0.

I can't organise my glossaries. I want to move them around subfolders.
I guess this is what you will do when you implement cut and paste ?
I also would like to be able to sort glossaries and folders.

When I don't select "Show all glossaries" and specify "Language1" > "Language2" I think I should still see all the folders, even if they are not Language1 or Language2.

I think each glossary should have some notes to explain who made it, where it comes from etc. If you change the glossary title label to "Title" then there would be room for an (i) icon which could be tapped for more information.

When you edit an entry, it would be nice if there was a way to navigate forwards and backwards to other entries.

The menu "Edition" should be "Edit", in prefs "Digraphes" should be "Digraphs" for the English UI.

thanks

koan

Is it just me or does the edit function does not save modifications? Maybe this is a permissions problem but it worked with the previous version.

Just to clarify, the only affected field (the one that does not save changes) is the comment field. The text fields work fine. This is weird.

I'm back to Japan.

Thanks for reporting this bug. I can reproduce it so I will fix it soon. This occurs only with the Edit button of the Quiz. The Edit button in the Glossary Manager works fine on my side.

Hi, I would like to make another request, though I'm not sure everyone will like it..

When importing files, Tomotko searches automatically for importable files... and this takes about 8 minutes on my sl-c3200. So I'd rather save time and search/select it myself if possible.

What do you think?

Exactly. The Cut & Paste feature should take care of this. In the present version, Cut doesn't work but Copy & Paste work for terms. So if you can't wait, you can reorganize your words by recreating the folders (that's easy to do), copy and paste the terms in the newly organized folders. Once you're done, you can then delete the obsolete glossaries with the red X button.



I've discussed this issue previously in this thread with Kurochka, if I remember well. I don't think this feature is essential. The folders and glossaries are already sorted alphabetically. I could add 2 buttons (arrow up, arrow down) to translate a tree item from one position up or down but this would burden the Glossary Manager's interface too much, IMHO. Because of that, I suggest you to prefix the folders and glossaries titles that you want to be sorted in a specific way with numbers like 01, 02, 03.




To see a folder when "Show all glossaries" is disabled, the folder must contain a glossary with some terms in the language specified in "Language1 > Language2". If you think about it, it makes sense. What's the need to show an empty (or irrelevant) folder? Most of the time, you don't want to see it. When you do want to see such a folder, that's where the "Show all glossaries" comes handy so you can insert words in it. I think that this behavior is what most people want, like 98% of the time. I may add a preferences setting to change this default behavior in the future but I'm not convinced that it's a good idea.



This is on the todo list.




I will put it on the todo list and consider it later.




I will update the strings shortly. Thanks.

Wow, that's incredibly long! I was not aware of this. This is probably caused by the Qtopia's default FileManager class that I'm currently using. I've read somewhere that there is another implementation of FileManager (provided by Sharp's API?) that could do a better job and ask for a specific location. I will look further on this.

I haven't encountered this problem on my 3100. It's snappy here.

I think it depends on the size of your storage devices. On my SL-C1000, I'm using the built-in 64 MB memory and a 1 GB SD card so it's fast enough. According to what I read on the Zaurus's FAQ, the SL-C3100 has built-in memory of 128 MB with no hard disk so the speed is probably comparable. The SL-C3200 has a 6 GB hard disk (or so I think) so that would explain the delay as Qtopia's FileManager object looks for all files with an extension ending by .xml under user's home directory. If you have a lot of files, it may takes some time to scan all of them.

I uploaded 3 new glossaries for Thai:

• Numbers
• Time
• Everyday phrases

They are not exhaustive but I think they are comprehensive.

Please see the "Thai on Zaurus" link in my sig.

thanks!

Thanks Koan, I will take a look to your new glossaries during the week and add them to toMOTko's website.

@Matthis

I borrowed zfiledialog code from ZBEDic application and integrated it into toMOTko's import/export feature. This will probably solve the problem you mentioned before. Similarly, keyboard shortcuts will also be inspired from code found in ZBEDic. I intend to make a release as soon as this is implemented.