|
 |
Brun_Ugle flies again (TAC 2012 team い) | ||
Share with:
| ||
 Language Learning Forum : Language Learning Log |  | |
| Brun Ugle Diglot Senior Member Norway brunugle.wordpress.c Joined 6618 days ago 1292 posts - 1766 votes  Speaks: English*, NorwegianC1 Studies: Japanese, Esperanto, Spanish, Finnish | Message 113 of 276 28 April 2012 at 8:55am | IP Logged | ||
|
We haven't actually started anything yet. We are waiting for the official start on May first. People are posting things to test the bot. Everything will be erased on 30. April so we can start for real on 1. May. At the moment all I'm posting is nonsense. I'm just trying to see how it will add things of different lengths and how it will come out when I post by different methods. When I start for real, I will only post when I finish a book. If I did it each day or something, I would have to figure out how many pages since the last time I posted, and that's too much work. Plus it becomes easier to make mistakes. I'm sure some people will post as they go, but I won't.
I think the real challenge will be to figure out how many words I've written. Since everything is run together in Japanese, it isn't so easy to count. Also, there is often a little disagreement on what constitutes a word. I see things divided up differently sometimes in different dictionaries. But I won't worry about it too much. I think I will just go by the "close enough" principle. 1 person has voted this message useful | |||
| g-bod Diglot Senior Member United KingdomRegistered users can see my Skype Name Joined 5980 days ago 1485 posts - 2002 votes Speaks: English*, Japanese Studies: French, German | Message 114 of 276 28 April 2012 at 10:28am | IP Logged | ||
|
I think in Japanese the standard is to count characters rather than words. I can't think of any better way to do it.
I'm getting very excited about the challenge. I want to start already! 1 person has voted this message useful | |||
| Brun Ugle Diglot Senior Member Norway brunugle.wordpress.c Joined 6618 days ago 1292 posts - 1766 votes Speaks: English*, NorwegianC1 Studies: Japanese, Esperanto, Spanish, Finnish | Message 115 of 276 28 April 2012 at 12:43pm | IP Logged | ||
I don't know. If you count characters, you'll get a very high word count compared to if you'd written the same thing in another language since most words are more than one character. Of course if you wrote something more than a few lines long, it would be tedious to count every word, but it would be possible to count all the words in just a few lines, take an average of that and multiply by the number of lines. I doubt I'll be writing very long things anyway. Mine will probably be more like,"Today I woke up at 7. I read a book and I went for a walk." I'm not sure I can manage much more, but maybe with some heavy dictionary work.... 1 person has voted this message useful | |||
| g-bod Diglot Senior Member United KingdomRegistered users can see my Skype Name Joined 5980 days ago 1485 posts - 2002 votes Speaks: English*, Japanese Studies: French, German | Message 116 of 276 28 April 2012 at 2:21pm | IP Logged | ||
|
Yes, I guess what I meant is that, for example if I am set assignments in Japanese, the number of characters are stipulated rather than the number of words. And I know my version of MS Word automatically counts characters for Japanese text, but words for English text.
Anyway I did a quick comparison between 100 words of some random English text I typed and some Japanese taken from an assignment I'd written. I used the same font and size to see how many characters of Japanese would fill up the same amount of space and it came up with 250 characters. So maybe 250-500 characters might be a reasonable equivalent for 100-200 words? I also tried counting the words in the Japanese text (estimated around 97) but it's really tricky. How many words are there in 暖かくなるのでしょうか? Edited by g-bod on 28 April 2012 at 2:22pm 1 person has voted this message useful | |||
| Brun Ugle Diglot Senior Member Norway brunugle.wordpress.c Joined 6618 days ago 1292 posts - 1766 votes Speaks: English*, NorwegianC1 Studies: Japanese, Esperanto, Spanish, Finnish | Message 117 of 276 29 April 2012 at 7:48am | IP Logged | ||
|
Log for 2012.04.22 - 2012.04.28 inclusive
Now it’s only two days until the Super Duper Challenge begins. I hope they manage to come up with a final version of the rules and the twitter bot before it begins. I never thought I’d use Twitter, and that has actually kept me from doing Tadoku, but I’ve found out it’s not that bad. It’s even easy, which wasn’t really one of my worries, but I thought you had to use a mobile phone. I didn’t really want to do that. Now I wonder if I really need this log, a blog, and twitter. It might be overkill. I’ve been rereading HP 1 for practice and because I wanted to read something before the challenge started. I’ve found that I have to read a lot more per day than I have been doing. I’m probably going to have to read about 4.5 hours these last two days in order to finish before the Challenge. I need to finish a book every 6 days on average and at my reading speed that means close to five hours per day. Of course, HP 1 is fairly long, but falls a little short of the 2 book mark, so if it was a little longer, it would count as 2 books and then it might not be so bad. Plus, I was so stressed some of the days that I could barely concentrate and sat staring blankly at the pages for ten minutes at a time. Even so, my reading speed when paying attention is only about 5 minutes per page. I can see that my life and my study plan for the next 20 months is going to consist mostly of reading, the occasional movie, and writing on lang-8, which will probably take me more like 5 minutes per word. Reviewing the Kanji: Time = 1:19. Read the kanji: Time = 2:26. Reading: Time = 20:39. iKnow: Time= 7:35. Total for period: 32 hr, 0 min Total since start of TAC 2012: 411 hr, 25 min Total since I started keeping track (2011.11.06): 605 hr, 48 min Only 1882 hours, 42 minutes and 20 seconds to go ;-) 1 person has voted this message useful | |||
| Brun Ugle Diglot Senior Member Norway brunugle.wordpress.c Joined 6618 days ago 1292 posts - 1766 votes Speaks: English*, NorwegianC1 Studies: Japanese, Esperanto, Spanish, Finnish | Message 118 of 276 29 April 2012 at 8:19am | IP Logged | ||
Yes, I know how Word counts them since I tried it too. And it really is hard to count. Some people might say 暖かく なる の でしょう か, others would say, 暖かくなる の でしょう か. Very difficult. But taking an average like you did, is a good idea. I tried a slightly different way. I took some Japanese and the English translation, so the text was the same and I got 100 Engish words to 200 Japanese characters. So I think that multiplying the number of Japanese characters by 0.5 (my answer) or 0.4 (your answer) gives a good estimate. Maybe we should take the average and say 0.45. It will always be a rough estimate in any case. Plus some people like kanji more than others and that will also affect the word count. Some will write 下さい where others might write ください, for example. Also, if we took the translation in a different language like German, where they connect words together to make new words, we'd get a different answer still. I think I'll just go with 0.45 and say that's good enough for me. Edited by Brun Ugle on 29 April 2012 at 8:23am 1 person has voted this message useful | |||
| Woodsei Bilingual Diglot Winner TAC 2012 Senior Member United States justpaste.it/Woodsei Joined 4795 days ago 614 posts - 782 votes Speaks: English*, Arabic (Egyptian)* Studies: Russian, Japanese, Hungarian | Message 119 of 276 29 April 2012 at 10:35pm | IP Logged | ||
|
Reading your post on word counts I just though of something. Some people use text
parsing for Japanese, I think initially, to help them know where they are when reading (i.e. grammatical relations, but then it shows words too) I don't think it really boils down to either kanjied words, or only kana, though. I don't know, maybe parsing is the way to go? Here are two parsing tools: Langrid Alternative KNP Interface It's basically one of the above two, but with it's own interface. I like the layout of the first one better :) Thoughts? Edited by Woodsei on 29 April 2012 at 10:40pm 1 person has voted this message useful | |||
| Brun Ugle Diglot Senior Member Norway brunugle.wordpress.c Joined 6618 days ago 1292 posts - 1766 votes Speaks: English*, NorwegianC1 Studies: Japanese, Esperanto, Spanish, Finnish | Message 120 of 276 30 April 2012 at 9:41am | IP Logged | ||
|
The parser is great. (The first one. I couldn't make the second one work at all.) It did just fine with kanji too. However, I don't see any way to make it count the words. If it doesn't count them, then we would have to count by hand, which would quickly get tiresome. Do you know a way to count without counting?
Edit: Actually, I just noticed that it isn't so good with verbs. It divided できません into 3 words でき ませ ん, います into two いま す and 取り組んで also into two 取り組ん で. So I think the word count would still be way off. Edited by Brun Ugle on 31 December 2012 at 1:40pm 1 person has voted this message useful |
|  | 
Printable version
|
| Copyright 2024 FX Micheloud - All rights reserved No part of this website may be copied by any means without my written authorization. |
