Better handling of multi-word LingQs
under review
A
Al Bedo
Mentioned previously in various posts but let me add this formally as a feature request.
In Vietnamese, and possibly other languages too, a single English word is very often split across 2 or more words in Vietnamese. As an example, the word breakfast translates as bữa sáng. But LingQ currently treats these as 2 independent words rather than a single entity. It's as if in English LingQ were to treat 'break' and 'fast' as 2 disconnected words and supply all the meanings for the 2 individual words but not the combination.
Ideally, two changes are needed to LingQ: First a look-ahead function for each word to see if the word combination ranks higher in frequency than the separate words. If so, to make 'break fast' as the pre-LingQ (ie bữa sáng in Vietnamese) rather than the 2 individual words.
Second, if a LingQ for the word combination is present then LingQ should suppress showing the individual component words in the vocabulary list in page or sentence view. Otherwise the vocab list gets very bloated and cluttered making it difficult to spot the meaning for a particular word.
I suspect that this second feature would be easier to implement than the first. Implementing both would be the perfect solution. But the second feature alone would be real progress.
Mark Kaufmann
under review
Mark Kaufmann
Yes, it would be nice to improve this. We have some ideas and will take your feedback under consideration as well.
A
Al Bedo
Mark Kaufmann I should have added that what I am describing is not an edge case in Vietnamese. Other than the simplest of words like many pronouns and prepositions (which quickly become known words anyway), perhaps 30-40% of all words in Vietnamese are these di- and tri= syllable/word combinations. As a result, the vocabulary list in sentence/page view can literally become 3 or more times longer than the real list of actual LingQs, which is both unwieldy and confusing to hunt through.
Also it becomes commonplace to see confusing sentences like the attached image for 'I like Sundays' where Sunday is 'chủ nhât'. A lingQ has been created for the word-pair chủ-nhât, but LingQ still wants to show me chủ and nhât as separate unknown words, in addition to the combination LingQ. It's even more distracting when chủ nhât becomes a fully known word, with no background colouration for the word-pair.
Mark Kaufmann
Al BedoThat's an interesting phenomenon for Vietnamese I guess. Not sure how best to deal with that. Since those have been identified as words, our system can't ignore them. Even if we can show the grouping, the separate segments will still be considered words because of the spacing. We will have to think about that.
A
Al Bedo
Mark Kaufmann The issue is less that LingQ should deal with this automatically (though that would be the perfect solution of course) but that there is no obvious manual workaround either.
What could work would be a checkbox for each component word for 'ignore if present in a lingQ phrase'. This would be set manually by the user along with 1/2/3/4 etc. If this option were checked then if the component words were present in a LingQ phrase then:
- The individual component words would not be shown in a page/sentence vocabulary list, but only the complete LingQ phrase.
- Highlighting would be ignored for the component words but only used for the LingQ phrase.
By the way, I am johnd2 (John Dann) in the LingQ forums. In making the original feature suggestion Canny forced me to register a new user name and I never use my real name for these things online.