renzhe Posted October 31, 2011 at 09:53 PM Report Posted October 31, 2011 at 09:53 PM There, I've said it. I started going through it recently to make sure that I'm not forgetting any important vocabulary, and I'm spending more time correcting entries than studying. It feels like at least 1 in 10 is wrong. Either the tones are obviously wrong: zuo3 for 作, guang2 for 光, or the translation uses some obscure meaning without mentioning the obvious one, or is often simply wrong. If you are going to use Anki, do yourself a favour and import the Mnemosyne HSK deck instead. It has a small number of issues, but nothing this grave. I understand that this was done by a volunteer, but I don't recommend taking a chance with it if you are a beginner. EDIT: yes, I'm referring to the old deck, ~8,800 entries, the most popular one when you select "download decks". 4 Quote
jasoninchina Posted November 1, 2011 at 05:20 AM Report Posted November 1, 2011 at 05:20 AM I have the New HSK Word List and it appears to be spot on. I have completed levels 1-4 and I often double check definitions and tones. Perhaps you have one of the older versions. 1 Quote
creamyhorror Posted November 1, 2011 at 05:52 AM Report Posted November 1, 2011 at 05:52 AM Yeah, you should specify which deck you're referring to, because there are a bunch. A while ago I got the "New Hanyu Shuiping Kaoshi" deck with 5003 cards and coloured readings to have a look (from Anki's public deck repository). It seemed okay, a few hundred cards in. The meanings are just bog-standard CEDICT ones as far as I can tell. I would recommend this one because of the colours. Quote
Olle Linge Posted November 1, 2011 at 06:00 AM Report Posted November 1, 2011 at 06:00 AM The problem with "popular decks" is that there is a snowball effect; once people start liking something, everybody else does too, regardless of the quality. We can only see the number of downloads in Anki, not whether people actually liked or even used the deck in question. Therefore, I think it's our responsibility to clean up as much as possible, especially for popular decks that might otherwise become nuisances for lots of students. Renzhe, if you've I identified such a large number of errors, would it be possible to contact the author and describe the problem? If there are better decks already in Anki or elsewhere, I doubt s/he will hesitate to remove the bad one. 1 Quote
OneEye Posted November 3, 2011 at 04:40 PM Report Posted November 3, 2011 at 04:40 PM Re: Snigel's post, I posted a deck a while back called "Taiwan All Grades" or something like that, which has proven to be fairly popular (as such things go). I feel pretty bad about it now because a few days ago I took a look at it again for the first time in a long time and realized there were a lot of odd things going on. Not incorrect, necessarily, but odd. I got the definitions and pinyin by running the list through an old version of Pinyin Toolkit, which resulted in things like 中 being listed as "zhòng, to hit (the mark)". Not wrong, but not the first thing I think of when I see 中 either. Not to mention, now that I think about it, I was probably wrong about that list being "what Taiwanese kids learn in school, in order". I don't remember where I got that info from, but I highly doubt the last character learned is 菇. I'd love there to be a corrected version of the deck, but there isn't yet, and I don't have time to do it. If anyone else is up for the task I'd be happy to take my deck down. Quote
Olle Linge Posted November 4, 2011 at 06:35 AM Report Posted November 4, 2011 at 06:35 AM I think this is a problem inherent in sharing decks online. I mean, if you use someone else's decks, you can't just assume that they are perfect. However, I do think a serious flaw in Anki is that popularity is measured by number of downloads. This is clearly bunk because sometimes the only determining factor for which deck to download is how many people have used it before. Thus, the worst of all possible decks (not yours, ) might still look like the most "popular". Also, it requires quite a lot of effort to keep large decks updated. I think the only reasonable way of doing it is if you still have those words in the deck you use yourself and have them tagged according to which deck they belong to . That way, you can constantly update. However, that assumes that you keep all your flashcards. I do, but it still doesn't cover everything. I share nine decks in Anki, five of which I keep updated. One or two of the remaining decks I can update if I want to, but I don't really change much in them so I probably won't update them either. Two are from textbooks I added years ago using ZDT, so I don't have them at all in Anki (they're beginner textbooks using simplified Chinese). So, if everyone would conscientiously keep all their words tagged, it would be much easier. I do, but I realise most people don't. I think an upgraded system for sharing decks would be needed to solve this problem, perhaps including some kind of rating system for decks. What do you think? Quote
laurenth Posted November 4, 2011 at 08:32 AM Report Posted November 4, 2011 at 08:32 AM Although I'm very happy to use a high-quality shared product like Anki, I would do nobody a favour if I contributed back by sharing my own deck. It contains information coming from several sources (translated into English or French, based on the dictionaries I happen to be using), tailored to my needs, with mnemonics that wouldn't mean anything to anyone else, funny etymologies, etc. Conversely, I tend to stay away from shared decks, as they often feel "machine made". I believe that the process of creating a deck, typing characters, looking for definitions, examples, etymologies, etc. is part of the learning process. My Anki deck faithfully records my learning process. So my Anki deck is my companion - almost too private to be shared Quote
Olle Linge Posted November 4, 2011 at 09:00 AM Report Posted November 4, 2011 at 09:00 AM Yeah, I feel the same, but I come to a different conclusion. I share some textbooks that, as far as I know, no-one else is sharing. Sure, apart from the information from these textbooks, my decks contain lots of extra information on etymology, usage, character components, mnemonics and other notes. Still, this is in addition to the other information. I reason that it's much better to have an accurate and updated deck available if people want it than to have nothing. Of course, my decks contain more information that people usually want, but they can simply stop reading after the English translation if they wish (most of my other comments are in Chinese, so they're mostly useless for students of that textbook; I have of course added these comments much later than I added the actual flashcard): I'm not saying you should do the same, of course. I mean, I would never dream of sharing my master deck publicly. Not only does it contain too much personal notes that won't make much sense to other people, it also contains lots of words that are... hm... not sure I would like to publish with my name on the deck. Still, I keep my old textbooks tagged, so exporting these tags to new decks and uploading to Anki is easy. Sure, I believe it's better for people to create their own decks, but if the choice is between using pre-made decks and not using SRS at all, I'd prefer my students (and others) to use SRS with pre-made decks. Quote
Recommended Posts
Join the conversation
You can post now and select your username and password later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.