While you may already have considered this possibility: may I suggest that Word net is also worth a look as it has some thesaurus like properties and is already incorporated into its own software. (Princeton Wordnet —> wordnet.princeton.edu) www.globalwordnet.org/ —> www.asianwordnet.org/ —> asianwordnet.org/ Including Thai.
NB: wordnet.princeton.edu/wordnet/license/ (Free — must cite the Princeton license) While we Europeans can soon feel at home with many dialects of our language: with romance; Germanic and scandinavian roots; intertwined histories for at least 10,000 years — Thai is spectacularly different. Many concepts in asian language do not translate into European wordnet categories very well. Lexically or logically they may seem to but practically would give surprising results. However, Word Net may have some application? I'd consider video clips as one of many possible ways of bridging impossible language gaps. Manga bridging would be "Awesome": wish I could do it ! ¡ May have to train myself with manga.mem decks ? George On 25 Mar 2012, at 03:11, John Francis Lee wrote: > I have made a (half complete) English-Thai-English version of the Longman > Dictionary of Contemporary English (LDOCE) defining dictionary and I would > like to make it easy to install on flash disks, so that Thais can use it to > learn English and English speakers can use it to learn Thai. The idea would > be that, like a virtual memory page fault, you would start out with all the > words 'on disk' and none your own memory but each day would learn 5 new > words... add them to your mnemosyne regimen, anyway. > > The words, definitions and example sentences are in an sql database, so I can > write out an xml file for mnemosyne from that. Apparently I've started this > project before because I found an example that seems to work, with entries > like... > > ------------------------------------------------------------------------- > > <?xml version="1.0" encoding="UTF-8"?> > <mnemosyne core_version="1" > > <category active="1"> > <name><LDOCE></name> > </category> > <item id="_0"> > <cat><LDOCE></cat> > <Q>a (Det)<br/><center></center></Q> > <A><sound src="sounds/LDOCE/a/a.mp3"> > (Det)<br/><center></center></A> > </item> > <item id="_0.tr.1"> > <cat><LDOCE></cat> > <Q> (Det)<br/><center></center></Q> > <A>a (Det)<br/><center></center> > <sound src="sounds/LDOCE/a/a.mp3"></A> > </item> > <item id="_1"> > <cat><LDOCE></cat> > <Q>ability (N)<br/><center>I have the ability to write English > sentences.</center></Q> > <A><sound src="sounds/LDOCE/a/ability.mp3"> > ความสามารถ (N)<br/><center>ฉันมีความสามารถ ที้เขียนประโยคภาษาอัก > ฤษ</center></A> > </item> > <item id="_1.tr.1"> > <cat><LDOCE></cat> > <Q>ความสามารถ (N)<br/><center>ฉันมีความสามารถ ที้เขียนประโยคภาษา > อักฤษ</center></Q> > <A>ability (N)<br/><center>I have the ability to write English > sentences.</center> > <sound src="sounds/LDOCE/a/ability.mp3"></A> > </item> > > -------------------------------------------------------------------- > > This works, kind of. But it shows both the English-Thai and then the > Thai-English cards one after the other. > > I'd like to be able to choose which way to go, and I'd like to add sound in > both direction, so if the Q is ability the A would be ความสามารถ and that's > would you'd hear when you hit Ctrl-r. Similarly, If the Q were ความสามารถ the > A would be ability and that's what you'd hear. > > The idea would that you would choose which category you wanted... say <LDOCE > th-en> or <LDOCE en-th>. And in future we might add <LDOCE en-cn> and <LDOCE > cn-en> other other categories. > > Another consideration is source code control... since many people will have > individual copies of the dictionary I'll need some way of being able to > incorporate changes into a 'master' copy of the dictionary and a means of > people synchronizing their individual copies with that master. That would be > especially useful for adding new languages, where someone might just start an > <LDOCE en-xx> and an <LDOCE xx-en> from scratch. > > I suppose I could use mercurial for that, although I don't know much about > mercurial or any distributed source code control system. I think the drill > involves creating a 'merge tool' for updating the master. > > If anyone has suggestions on how I should best go about this I'd appreciate > hearing them. > > Thanks. > -- > "This message may have been intercepted and read by U.S. government agencies > including the FBI, CIA, and NSA and/or the present government of Thailand > without notice or warrant or knowledge of sender or recipient." > > John Francis Lee > 246/3 Thanon Kaew Wai > Mueang Chiangrai 57000 > Thailand > > -- > "This message may have been intercepted and read by U.S. government agencies > including the FBI, CIA, and NSA and/or the present government of Thailand > without notice or warrant or knowledge of sender or recipient." > > John Francis Lee > 246/3 Thanon Kaew Wai > Mueang Chiangrai 57000 > Thailand -- You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/mnemosyne-proj-users?hl=en.
