In the United States the glyphs in typefaces are not subject to copyright (though the font files may be), so while there may be some copyright issues elsewhere it will be at least PD-US. Incidentally this project will be very useful for wiktionary, I hope it is still progressing ^^ —Muke Tever05:29, 25 August 2005 (UTC)[reply]
Thanks, this link is usefull but we found several stroke order mistakes in this website. We are doing better :] Yug(talk)00:38, 29 May 2006 (UTC)[reply]
I saw it but i have some difficulty to understand .. but 4 version for 必 , hum .. perhaps we should ask a japanese wikipedian for help in translation, the black one are correct and pink one incorrect.
I have a program called DimSum which already has thousands of stroke orders. You could probably ask the author for a copy at his forum at [http:\\www.chinesecomputing.com]. Unfourtunately, the forum has a lot of downtime.
Hi, when I studied the jouyou kanji I checked stroke orders that aren't dead obvious from multiple sources just to be sure. You should be aware that there are some variations in them, even in places you would expect to have correct stroke orders. Such problem characters will need special attention. I think that common errors/variations should be documented (with references).
For jouyou kanji, gahoh is fairly reliable with its stroke order movies, so using it to double-check Japanese stroke orders is a good idea. It agrees with 漢字源 on all jouyou kanji I have checked (which is many) except 衷[2], which differs even in stroke count. 漢字源 uses 10 strokes, gahoh 9. 漢字源's version is 衣+中, where first two strokes of 衣 are written first, then 中, and then rest of 衣 (using stroke orders of 衣 and 中).
Another site, which has many non-jouyou kanjis, is kakijun, which has some peculiarities, but the site owner is of the opinion that the stroke orders there are the most common actually used and that they are not wrong either. The stroke order of 虎[3] and others that have 虎冠 in them (虐, 虚, etc., but not 劇 as the site owner's sources specifically mention the stroke order for that) are examples of this.
In any case, for most characters in the jouyou set, all (modern and even remotely reliable) sources will agree. Of course older works will contain stroke orders and even stroke counts that are not taught anymore, even if they were reliable in their time. Top of 発[4] and left of 収[5] are two cases at least that are relatively often given different stroke orders (or counts). AFAIK gahoh is correct on these.
Finally, thanks to everyone doing this. I for one would have wished for such a resource as you are aiming for when I was studying the characters. -- 82.103.215.23623:23, 13 January 2007 (UTC)[reply]
Dsl pour cette longue absence, je recommence à me replonger dans ce travail. D'autres anim seront bientot publiés (qd j aurais de nouveau le net..).--Wikic16:52, 7 March 2006 (UTC)[reply]
Shouldn't we make sure that all the images are transparent?
We are using GIFs, PNGs and SVGs. These all permit transparency. — Erin(talk)04:26, 17 July 2006 (UTC)[reply]
I have extended Ricardo Ferreira's Greasemonkey extension for the popular Reviewing the Kanji site to display stroke order animations using the files in this category. The Greasemonkey is here.
I wanted to offer my thanks for creating and populating this category, which is enabling a very cool feature.
I am not sure where I should put acknowledgments for use of the images in the Greasemonkey extension. I currently have a comment in the Greasemonkey script. I store copies of the images on my website, but they are embedded as inline images in tooltips by the script. I want to do the right thing, but not sure exactly what it is. The image pages appear to indicate that the images are in the public domain.
I am a wikimedia newbie, so this may be a dumb question, but is there any way for me to easily get a list (or notification) of new stroke order animations as they become available?
I am just a beginner in this project, so I'll make the first reply but probably not the definitive one.
Thank you for making this work useful!
You can use your own copy of the images without problem. Existing animations (including specific japaneses ones) are indeed public domain. Other works, like BW pictures and red ones, are mostly under Creative Commons Attribution Same agreement (CC-BY-SA) licence, meaning that you should give a visible credit to the author and give a link to the licence. I cannot certify that all future animations will be in DP, it depends on the contributors. Anyway, Commons allows only free licence so at worst it would probably be a CC-BY-SA licence too. As there are only a handful of people doing this, if it happens to become CC instead of DP in the future, you can probably group the authors on a small on-screen greeting line with a link to the licence et voila.
There is no RSS feed or things like that. This URL will give you the 250 last changes over the past 30 days in the animation category (no stress; right now, there are not that many changes). The same can be done with the others by modifying the name in the URL. Unfortunately, creation are not easily dinstinguishable among other modifications. Eden200409:10, 18 July 2006 (UTC)[reply]
( To see this use, follow the instruction of this script, install Greasemonkey, install the script, and go to the page for 五 and mouseover the large 五 (you may need to create an account). )
I have now been using these for a while and need to make a decision whether I will continue to use the Wikimedia stroke order animations. While they work beautifully in my site, there are very few (<10%) kanji animated and there appears to be an average of no more than one a day being added. When you combine this with the large number of different stroke order projects at this site, simple extrapolation suggests another 3 or 4 years will be necessary to complete animations for most kanji. Is there a reasonable expectation that this rate will increase? If not, as much as I like these animations, I will not be able to use them for my projects. I'm not complaining because this is a free and much-appreciated service. On the contrary, I'm writing this because I don't want to abandon these images unnecessarily. Unfortunately, I am not in a position to participate in the project myself as I am writing some large Japanese educational programs that will use these (not to mention a more than fulltime day job). Again, don't take this as a negative comment but rather as an attempt for me to continue to base these projects on your beautiful images. Any information on projected status would be greatly appreciated. Thanks! Mps13:46, 4 August 2006 (UTC)[reply]
That's true, we go slowly for the moment. We first have to finish the 300 radicals gif, then everybody who have gif skill will be able to contribut, and we will go faster. Anyway, We think finish this 300 radicals for ~january 2007 only, and the 1000 usest kanji for December 2007.
We go slowly, that true, and we cannot go really faster in a calligraphic quality. But we will go faster after, so keep an eye on our project :] , we do more beautiful and more exact than other work made.
Last thing, your script made a good pub for us, we have one more volonter for the *-order.gif, see User:CharleyGarrett, So thanks you anyway. And please, when we will need more new contributers, talk about this project everywhere.
Thanks for taking my comments in the intended spirit and the update on the timeline. I'm also glad you'll be betting a new volunteer. Being able to use these animations would definitely be my first choice. On a separate topic. I notice some images ending in "...sbs.gif" that don't seem to show stroke direction. Can you tell me what they are?Mps18:57, 6 August 2006 (UTC)[reply]
-order.gif are in a hight calligraphical quality and take lot of time. So first, we have to do the 214 keys. Then, try to do the 1000 commonest characters in this calligraphic style will need probably 1 or 2 year.
-sbs.gif will just show the stroke order , so is really easier, a contributor with just some graphic skill [not calligraphic] will be able to join the -sbs.gif sub-project.
I think it'd be cool to help out with this, but I don't know where to start. To give you a little background, I studied Chinese for 3 years and lived in Beijing for 6 months. I have Wenlin and Photoshop. Where should I start? --Fang Aili01:16, 14 November 2006 (UTC)[reply]
I seen in your user page that you do not have graphic abilities. If you have to learn a full protocole, I think the -red.png is the most convenient.
The second one are the -bw.png, which are more difficult, but we already have a pretty set of ~1.000 pics with some missing which need to be complete.
If you really have time and big volonty and so may learn how to do gif as beautiful as these made by Wikic, you can join the -order.gif .
In summary, I think contribute to complete the -bw.png of the 214_radicals his the most need. Code such H-P indicating the good stroke order.
Great Idea, can someone help a newbie link my attempts[edit]
Hello, I think this is a wonderful site and I would like to help make some order.gif's
I have made the following:
This second one was linked to already I think, but I don't know.
But I am not so certain how to link them or put them on the page--I did try, but my feebleness seems to have gotten in the way...
I am a Chinese teacher and I plan to use gif's like these to free up class time I use to teach characters so I am hoping to make more contributions in the coming weeks and months.
Also suggestions on the .gif's would be more than welcome.
user:Nvtsky (Scott)
Hello stroke order team,
before I come up with my question(s), I would like to say "Thank you" for your good work. Keep going on, there should be many language learners who appreciate your work. You already seem to have invested quite some time into this thing, so I would like to add some information if needed or check if you have some more information for me :).
I came across Chinese stroke order:214 radicals, and especially column "Validated ?" caught my eye, as I am currently looking for a table of character stroke orders, e.g. "牜: P-H-S-T". There are many tables on the net that include stroke orders for chinese characters, but all of what I found is limited to data for points in 2-dimensional space for drawing charactes subsequently (e.g. DimSum has one for stroke order animations, check my edit a few lines up). What I need is semantical information on strokes, not their graphical description in a certain character. You already have some entries for approx. 150 charactes, that at least is a start for me :). A paper I've read states that all characters can be made out of 2000 basic charactes, so 2000 stroke orders should be sufficant. That is still fairly a lot. So what I would like to ask is weather you know of additional sources on strokes and stroke orders (esp. free ones, free as in beer).
Differences in chinese charactes dependant on localeNext question would be: Do you plan to add further stroke order images for korean variations? There can't be too many. But in a picture I made, based on information from Unicode, there are still some obvious differences.
Hello Chrislb, I will just answer to the easier questions :
Question n^2 : Korean ? from the few exemple that I studied, Korean books shown to me exactly the Japanese stroke order. So Korean users will have to give the priority to *-jred.png ; *-jorder.gif, and *-jbw.png. Some [really few] korean file may be need (such as *-korder.gif). But we cannot work on it now. (we still have lot of work before this)
Question n^3 : Graphic variants of one character ?, they are graphic variants function of the graphist who drawed the font. And you can find many different graphies of the same radical in the same font without reason (without historical or semantic background). Unicode graphist haven't build a coherent system of radical's graphies. Example : the Turtle character : unicode.org made a mistake, and unicode.org made several little (stroke) mistakes. [The stroke order being linked to the graphy.]
As you show in your pic, They are also graphic differences with reason. In exemple, inside words : simplified china make rou 月 S-HZG-H-H while traditional make it P-HZG-D-T.
Answer 4 : In Chinese stroke order:214 radicals, we divide in "Traditionnal stroke order", "Simplified stroke order" and "Japanese stroke order". We work only about stroke order.
The case of same number of strokes but with different strokes : In some case, It may be need (But we still did not) to make a difference between "Traditionnal graphy", "Simplified graphy" and "Japanese graphy"[, and "Korean graphy"]. We know it, but we are not there, so we just work on Stroke order differences.
Hi, I don't know where the best place is for asking:
For 直 you have to different zVariants: Image:直-bw.png and Image:直-order.gif. The tables of the Unicode Standard states that the first one is found in japanese and korean standards and the second one in taiwanese and mainland chinese ones. Maybe you can clarify on this. Thanks. --Chrislb16:21, 1 January 2007 (UTC)[reply]
I now understand the trouble. 直 zhí, is not the same japanese fonts and Chinese fonts, and it appear that I have a japanese font (!) to display web pages. The image on Image:直-bw.png have to be re-upload to Image:直-jbw.png. Chanueting was right. Yug(talk)22:48, 7 April 2007 (UTC)[reply]
Chinese-German dictionary HanDeDict uses Chinese stroke order images[edit]
Starting today, the free Chinese-German dictionary HanDeDict http://chdw.de will use stroke order images from the Chinese Stroke Order Project in its writing exercises (http://www.chinaboard.de/chinesisch_deutsch.php?mode=schreibuebungen). On behalf of the HanDeDict team, I would like to thank you very much for all your great work on your illustrations - you are really helping a lot of learners with your great project!
As we think that it takes a lot of work to complete illustrations even for the most common characters, we have not only included your license, but will ask anybody who would like to have more illustrations and knows how to create computer graphics to help you out. Of course we cannot promise anything, but we sincerely hope that you can gain some valuable contributors that way.
Apropos - the license information is included on the writing exercises download page mentioned above, as well as in the writing exercises themselves. We also link to your project on our about page (http://chdw.de/about). Since that is a lot of places, please bear with us that we did not mention every individual contributor's name. That does not mean that we do not value each and everyone's contribution to your project - on the contrary, we are very grateful to each and every one of you. But as we wish you the best of success, we would probably not be able to catch up with your growing team and forget to mention somebody. Even if that were only in one of the places we gave you credit, we feel that that would be unfair to the poor individual we forgot to mention, so in order to be fair and keep it maintainable, we would like to give credit to your team as a whole and link to your page where everybody can see who you are. (By the way - HanDeDict is published under a similar license as yours (CC attribution-share alike), and we have too many editors for other projects to give credit to each one individually, so we require a similar attribution by other projects who use our data.) I hope that this is OK with you - if it is not or if you have any other ideas of how we could cooperate, please contact us at redaktion@chdw.de.
Most caracters have the same stroke order, for stroke order specific to japan we simply add a "j" to the file name, such Image:生-jbw.png.
If you want help on the bw.png project, I think it's not so difficult to creat a protocol for gimp, the "Photoshop bw.png protocol" can be a good starting point.
The Chinese stroke order:Strategy page is now sleeping : I, and other contributors (mainly students) have currently others task to do (study ~).
Hi, I've been watching the Stroke Order Project for some time and worked on a program that tries to make the creation of animated and print-out chinese stroke order easier. Check out the links on my user page ThomasStrohmann05:40, 24 March 2007 (UTC).[reply]
I have an interesting little side project I've been working on for several years that is a GPL'd Chinese-English character dictionary. It is based on an earlier software project whose copyright status is rather dubious, but is a tool that was used for at least a couple of years at the U.S. Army w:Defense Language Institute for Chinese Language instruction. While I'm not a native speaker of oriental languages, I am quite familiar with the electronic rendering of Chinese characters and some of the difficulties in trying to get the information needed by somebody learning the characters.
I was wondering if there are any resources (especially on-line, but books are acceptable too) that would help with trying to determine what the exact stroke order would be with some of these characters that could be used by a westerner like myself to help out with the development of some of these images that are being created with this project. I've seen the protocol documents for this project, and I will say that they are very well detailed except for this one minor issue. Maybe I'm missing something here. --RHorning16:36, 25 March 2007 (UTC)[reply]
For the Stroke order issue, please read : en:Stroke order (starting point, overview) ; 214 radicals (our radical by radical summary) ; www.edu.tw - Taiwan Ministry of Education's 常用國字標準字體筆順手冊 (online authoritative book for traditionnal stroke order) will help you a lot.
Most characters have the same shape in all countries.
For a charaters which is different in Japan, it will be need to :
. use a japanese font,
. name the file with a "J" such as ?-jbw.png .
If you are japanese, you probably have such font in Kaishu/Kaisho style on your computer. Otherwise, you can find a Japanese Kaishu font on the web, on some Japanese website or maybe on http://www.zhaozi.cn. I can't help you more since I work on the Chinese side of the project.
Hello. Since Yug posted a message at the village pump of Japanese Wikipedia, some conversations on what is the authoritative Japanese stroke order have been exchanged between Wikipedians. The information given there may not be anything new to the members of CJK stroke order project, but since I kind of promised Yug to give a feedback, here I try to make a summary of the conversations (See Wikipedia:井戸端/subj/CJK筆順プロジェクトの翻訳依頼 for the original posts).
First, in 1958, Japanese Ministry of Education (now Ministry of Education, Culture, Sports, Science and Technology, MEXT) published Hitsujun shidō no tebiki 筆順指導の手びき, a handbook on stroke orders of 881 "kyōiku kanji", which are the kanji listed on the tōyōkanji beppyo 当用漢字別表, published in 1948 by the Cabinet of Japan (ENWP explains as 1946, but this is the year when Tōyō kanji was published).
This handbook was mentioned in the Standards for Textbook Authorization (教科用図書検定規準) as what school textbooks should follow until 1976. However, after the revision of the standards in 1977, this handbook has no longer been mentioned in the official standards. What it reads now is "In principal, the stroke orders of kanji should follow commonsensical orders which are widely accepted in the society. When the stroke order of a character in semi-cursive script differs [from that of regular script], appropriate explanation must be added." (漢字の筆順は，原則として一般に通用している常識的なものによっており，行書で筆順が異なる字については，適切な説明を加えていること。 MEXT, Standards for Authorization of the Textbooks of Compulsory Educational Organizations [義務教育諸学校教科用図書検定基準], 1999, 1, http://www.mext.go.jp/a_menu/shotou/kyoukasho/kentei/021202.htm. The same sentence appears in the Standards for High Schools [高等学校教科用図書検定基準].)
In Japan, along with textbook authorization, Curriculum Guideline (学習指導要領) published by MEXT controls elementary and secondary education. The guideline for Japanese language education at elementary schools refers to kanji stroke order as "[schools must teach students] to write correctly and following stroke orders, paying attention to the length of dots and lines or how they touch and cross each other." (点画の長短，接し方や交わり方などに注意して，筆順に従って文字を正しく書くこと。) at the first and second school year. (MEXT, "Japanese," Curriculum Guideline for Elementary Schools, 1998. 12, http://www.mext.go.jp/b_menu/shuppan/sonota/990301b/990301d.htm. The edition of 1988 used to include a similar sentence for guideline of the third school year, too.)
This may need to be move too
Some JAWP Wikipedians states that, as their personal experience, it seems that the stroke orders taught at schools nowadays sometimes differ from that of the above-mentiond tebiki. Also two books were given as possible useful references.
Japanese people have been being taught a little different stroke orders using textbooks by different publishers or in different days. That's because stroke orders of each character are not strictly determined.--Widehawk(Takahiro05274)06:03, 13 May 2007 (UTC)[reply]
I think the Stroke Project itself is very great and it should keep running somewhere.
But unfortunately there is one problem; this Wikimedia Commons is NOT an appropriate place for running the project. (Read carefully Commons:Project Scope (Accorind to the Scope, this Wikimedia Commons is a Storage for media files and is used for reference style from the other Wiki Sisiter Projects. )
I am not completely sure about the other Wiki Sister Projects, but why don't you use Wiktionary ?
The Wiktionary would be perfectly matched with the project idea.
Why don't you move it now before its getting too big to move.
Just in case... "Keeping the Letter files on this Commons, the only Description Page should be move to the other place such as Wiktionary. Then, that page on Wiktionary refers to files on this Commons with the reference links."
Just for my 2c, I think the Stroke Project is fine at Commons. The project scope is only really intended to keep out stuff that is not suitable for any Wikimedia project, and keep text and non-media things at their appropriate home. Since the stroke project is so closely tied to the media files, I think it's fine at Commons. But it could also be fine at Wiktionary. I think this is a borderline case and it would be appropriate at one of several projects. pfctdayelise (说什么?) 03:19, 1 July 2007 (UTC)[reply]
OK, I understand. Thank you for your clear explanation. I just wondered if it should be done somehow if it is not appropriate for this Commons project. --SantaClaus10:04, 1 July 2007 (UTC)[reply]
I don't know what SantaClaus is talking about in the Japanese, or it is a translation of the above one? In fact, I cannot see his/her point, this project can be used no matter wikipedia, wiktionary or wikibooks. Being a file database, I think wikicommons is the best place for us to maintain and group all the files. Also, we can combine the Com:ACC and the stroke project and share our resources. Chanueting09:22, 1 July 2007 (UTC)[reply]
Hi! I've found this good program for Kanji: [7] Arlas! 23:22 5 oct 2007
Chinese stroke order progress function calls[edit]
There are a couple of "progression" pages titled Commons:Chinese stroke order:Progress and Chinese stroke order:Progress2. Each has 400 instances of Template:Hanzigallery which raises the number of parser function calls for these pages to 1512 and 1596 calls, respectively. The edit preview page warns that the page "contains too many expensive parser function calls" and "should have less than 500 calls."
We could try to cut down on the parser functions used in the template (e.g. just leave red links). On the other hand loading 1200 pictures for each page (once completed) is pretty heavy. Would there be any objections to reducing the number of characters on each page down to a hundred? --Swift (talk) 05:47, 4 November 2008 (UTC)[reply]
The rationale behind these seems to be to seperate different stroke orders for the same category. See also a discussion in Micheletb's talk archive. I doubt these are the best names for the categories, but seems to arise from a feature in the template that adds the category. --Swift (talk) 18:46, 31 December 2008 (UTC)[reply]
My opinion is that we do not need this 3 categories at all.
It is useful to have the Torder category (for traditional stroke orders), and the Jorder category (for Japanese stroke orders).
The Category:Corder.gif stroke order images contains just one cursive picture, others are categorized at Aorder, which seems also of not much helpful use, but is proposed to collect alternations/variations.
In general, the state of categorizing is not the best one at Commons.
At the moment we have different contradictionary categorizing systems for stroke orders. It may be discussed which one will be the best one, but any straight system is better than different divergent systems.
We should consider creating special stroke order categories for each character. Since Commons is multi-lingual, Category:馬 and Category:国 detract a bit from their usefulness. They are, however, very simple. Hmmm ... just brought this up for comments. --Swift (talk) 06:42, 23 January 2009 (UTC)[reply]
I think it's ok to have easy categories so people can find them while not looking all over Commons. Maybe the category "Radical {{{radical}}}" I introduced with the SOP license should be even more general, too? --Tauwasser (talk) 06:59, 23 January 2009 (UTC)[reply]
Seems wrong to me as well. I found this page that verifies our concerns. All other combinations with 力 on the right I know of also have HZG before SP, like 働、動、勉 etc. However, as I understood it, there were also plans on the table to redo most of those gfx with vectorized arrows, so they gain a uniform look? --Tauwasser (talk) 13:37, 24 January 2009 (UTC)[reply]
It furthermore matches with what my sources give me for the Japanese variant as well as the authorative source for the traditional Chinese variant (both of these are 動).
On a general basis: could you guys add the stroke order like "HZG-P" to the decription template when creating new files? This would make it simpler in the future to check which stroke order this images follows. There might be ambiguous values where two different images have a different stroke order but same resulting textual description, but that should be a much more smaller problem. And when I say adding, I mean in a machine readable way. Additionally the distinct source could be mentioned especially in cases where two sources state differently. Hope I'm not asking too much here, with giving too few myself :) Merci--Chrislb (talk) 10:04, 25 January 2009 (UTC)[reply]
I think such over-view pages are the best way to go as well. The Kangxi radical lists currently do this, and I planned to put that data in the Kanji progress pages. These pages are, I think, a better place to review the stroke order and compare them with sources.
In the case of the traditional Chinese characters, things are fairly simple. We just type in the stroke order and link to the correct page in the online version of the Taiwanese Ministry of Education page for reference.
I've actually been contemplating whether we could (and should) somehow add data to the image files. Data that could be extracted both onto the description pages and into character lists. One (completely untested) idea that I had was to put information on a sub-page/sub-pages of the image (these acting in a way as fields in a database). It would have to be easy to construct and maintain, or we might as well just create an off-Commons database to handle those issues (something I've mentioned to Yug, already). --Swift (talk) 15:36, 25 January 2009 (UTC)[reply]
If we work in SVG, every data can be put into the SVG. But edit one by one each SVG is definitively not convenient. Use a text page with a list of "汉 D-D-T-HP-N" will be far more convenient to edit, and as simple to reuse. Yug(talk)05:55, 29 January 2009 (UTC)[reply]
Thanks for your response JimDeLaHunt. I hope this clarfies things a bit. Please let me know if this is still too cryptic.
The problem: We have files that illustrate the stroke order of simplified Chinese. When they differ from the traditional Chinese or Japanese, those variants are uploaded. We cannot, however guarantee that this has been done. If an image exists only for the simplified Chinese, we can therefore not tell whether that is because the traditional Chinese and Japanese are the same, or because they just haven't been added yet.
Now, we happen to know that the Japanese stroke order for 伐 is the same as for the simplified Chinese (both have been simplified from the original traditional Chinese stroke order) but this isn't explicit. We need to put something in place at Image:伐-jbw.png so that we know which to point to; the simplified or the Chinese.
I have created the example redirect: Image:一-jorder.gif. It seems that it behaves preciely like the original; lists in categories, though these are not listed on the redirect page itself.
The main problem I have with this method is that it is rather messy. Users might get redirected without noticing and modify the simplified, when they thought they were fixing the traditional or the Japanese.
the simplified stroke order image is copied to Image:伐-jbw.png
This is my favoured approach. It's fairly clean in that we're not playing around with redirects and there are less chances of mistakes. If that isn't compelling enough a reason, those copies might just be a waste of space. But then again, these images aren't that large and hardly as much of a waste of time as some of the stuff on here. --Swift (talk) 06:41, 13 January 2009 (UTC)[reply]
I know literally nothing about stroke order, so I apologize if I say anything stupid. It seems to me that the easiest solution would be to have only one copy of each file, and indicate in the description which case applies. In your example, this means that you would edit the description of File:伐-bw.png to specify that this corresponds to both the BW and JBW stroke orders, and include it in both category:Bw.png stroke order images and Category:Jbw.png stroke order images.
You could even edit {{SOlicense}} to include either a link towards the other stroke orders for that character, an indication that the stroke order is the same, or an indication that the other stroke orders are unknown and that the reader can help. Pruneautalk10:19, 13 January 2009 (UTC)[reply]
Thanks for your input, Pruneau! This is certainly an option (we already have progress pages listing which characters have been validated stroke orders), but it requires the user to check every file description page (or progress page) before using a file.
An example is b:Japanese/Kanji/Lesson 9 which uses a template to check if there are Japanese stroke orders available for roughly 100 characters (a drop in the ocean of the thousands used daily that are within the scope of the wikibook). It would be a lot of work for maintainers of Wikimedia projects (and anyone else wishing to use these files) if this had to be done manually.
Other projects using the slowly growing Stroke Order Project (such as b:Chinese (Mandarin), b:Cantonese, and similar projects on other Wikibooks sub-domains) would similarly benefit from an automatic system where validity were determined by existence. --Swift (talk) 14:41, 13 January 2009 (UTC)[reply]
I don't like the solution with copies. Your main concern with redirects are the categories: categories work with redirects, if you put the [[Category:]] statement on the same line as the #REDIRECT [[]].
Btw: What's the most likely case for Japanese character stroke orders: to be the same as traditional or to be the same as simplified? My guess would have been traditional, cause the Japanese system is based on the traditional system and simplified was only invented, when Japanese already existed for long. So, according to my guess, traditional would be a better fallback. Is my guess wrong? --Slomox (talk) 15:13, 13 January 2009 (UTC)[reply]
Thanks for joining the discussion, Slomox. My main concern with redirects is not the categories. My main objections to redirects is that they are harder to maintain and can be confusing. I've already experimented with redirects. Your suggestion to place the category on the same page didn't help with displaying the category info on the redirect page.
Unfortunately, there is no way to order the fall-backs. The different variants have all evolved seperately. Yes, this is one royal mess...
伐: t != s = j
生: t = s != j
話: s != j = t
There are also korean versions(!), but I don't know any details about their stroke order. All this makes a default line of characters difficult to organise. --Swift (talk) 17:45, 13 January 2009 (UTC)[reply]
I was wrong with the "on one line" thing. That was an old workaround. Categories on redirect just work fine without that workaround meanwhile. The problem in this case is a new problem unrelated to the old one to which "write it on one line" was the solution. The problem seems to occur only, if the redirect is in the image namespace. It's a bug (the history for image redirects is broken too) and I reported it to wikitech-l.
Yes, I was aware, all three systems have differences. My comment aimed at likeliness. How many Japanese stroke orders follow traditional style and how many simplified? Is it 10/90, 50/50 or 90/10 percent?
Why do you think, redirects are harder to maintain than hard copies? In my opinion the opposite is true. With hard copies an improvement to one image has to be done to the other copy separately. Well, hard copies on the other side are better in the case, that the 'actual stroke order' is changed in the original. For example, if the Chinese version is wrong but thereby by chance identical to the Japanese one. The Japanese set the redirect to the wrong Chinese one and later the Chinese change their version to the correct one without the Japanese noticing it, so now the Japanese version is wrong. But I guess improvements are more likely than deteriorations. --Slomox (talk) 20:49, 13 January 2009 (UTC)[reply]
Sorry for the late reply; lost the internet connection and couldn't submit it until now.
Thanks for reporting the bug.
Unfortunately, it's not a case of one or the other. The traditional Chinese is (as you assumed) the oldest. The Japanese branched off as soon as the characters were adopted. Some new characters were created and stroke order changed. The simplified Chinese are a much more modern creation, but radically reduced the stroke number and in some cases changed the order. In some cases these changed correspond to those in Japanese.
On the frequency of overlap, I'm afraid there isn't much information. There are about 50-100.000 characters in existence. A well-educated Chinese person may know around 5000 and the Japanese counterpart around 3000 (this has to do with language structure, not intelligence ;-)). Certain recurring parts are systematically written differently in the three schools we're looking at. I'm not familiar enough with the prevalence of these features to be able to say for sure.
I think redirects are harder to maintain because users may not notice when they being redirected. As you mention, however, a change in the license template could address that. Another issue, is the existence of a base-line. However, by doing away with the baseline and annotating the license template properly, any image can become the referred-to-character, and the same character can have multiple ones (e.g. the Korean redirects to the Japanese, and the simplified to the traditional). Thanks for the suggestion, I'm going to look into this! --Swift (talk) 08:15, 15 January 2009 (UTC)[reply]
Thank you for the explanation. I think I see the problem better. I would suggest three steps: 1) use redirects instead of separate copies of the image, where a Japanese or Traditional stroke order is identical to the Simplified stroke order; 2) add a letter 's' for Simplified, in parallel with 'j' for Japanese and 't' for Traditional; and 3) define a clear order of precedence for which letter to use for the real data's filename and which to use for the redirects (Simplified probably has highest precendence to have the real data instead of a redirect, next is probably Traditional). Why? Well, it looks like redirect pages can appear in categories. Note that File:一-jorder.gif appears in Category:一 and Category:Jorder.gif stroke order images. And I think Category:Order.gif stroke order images should show the jorder, torder, and sorder images, perhaps in subcategories. JimDeLaHunt (talk) 06:55, 14 January 2009 (UTC)[reply]
The project page has a few examples of stroke orders under the heading The three schools of writing. The Japanese image for 伐 is absent and in its place was a note stating that the stroke order is the same as in traditional Chinese. The difference in stroke order between the traditional and simplified Chinese is that in the traditional stroke order, the dot comes before the last downward-left stroke but in the simplified one, the dot comes last.
According to http://kakijun.main.jp/, a Japanese friend, and the only book on kanji that I have here in my apartment (Kanji ABC by Andreas Foerster and Naoko Tamura) the Japanese stroke order is the same as the simplified Chinese one. The book states that every kanji with the "halbeard" radical ("戈") is written with the dot last. In Japanese, therefore, this character seems to use the same stroke order as in simplified Chinese as opposed to the traditional Chinese.
I fixed this error, but that was reverted soon after. I have undone that and would ask that anyone provide their sources before changing this back. If anyone wants alternate or more detailed sources for that which I have provided already, please feel free to ask. --Swift (talk) 18:45, 22 October 2008 (UTC)[reply]
1st, sorry for the little edit war, I didn't noticed the revert... O.O [I just had a feeling of "déjà vu" when editing it again]
Authoritative sources can be found there: User:Yug/Stroke_order_according_to_national_rules#Sources. The Japanese are a special cases, since they let bookwriters free to state a stroke order... which should "follow Japanese tradition, or be the easiest stroke order possible". Accordingly, there are no authoritative source in Japan.
In my memory [I looked at many sources, don't remember clearly], the 伐's order in japan is the same than in Taiwan. I may be wrong. Yug(talk)09:57, 26 October 2008 (UTC)[reply]
I didn't found back my web sources, but I remember a bit more.
Japanese made a Kanji simplification in 1946. 骨 and 鬼 [ONLY] have one less stroke in Japan than in Taiwan... if I remember well.
Chinese made a Hanzi simplification in 1956, and went farer than Japanese : 3 strokes radicals changed into 2 strokes ones, stroke order change to be "horizontal writing friendly". 廴，瓦，禸，辶，骨，and 鬼 have one less stroke in China than in Taiwan.
There, Japaneses are more traditional than China. I don't remember exactly what is use in Japan for 伐 (Japan was not my specialty).
Online, the best source is probably http://www.kanjicafe.com/ (you can download the 1500 animations).
We already have some talk about japanese stroke order some years ago (on the archive), but websites have changes their structures, and -for japanese- no way to find any "authorative" source.
Accordingly, we are in the same situation than bookwriters : we have to follow usages, and for unclear cases, make a choice.
Marco have leave the team after the amazing 1000's pic work he done, I'm not able to explain more on this 戈 japanese issue. Yug(talk)10:39, 26 October 2008 (UTC)[reply]
I'd actually already found that archived discussion since starting this section. I guess we should still decide on a policy for the Japanese characters. I think it would be best if we stuck to the previously authorative list in terms of the characters added. Where there are alternatives, we could add those as such.
I think this once again shows the need to set up some sort of a process or structure the project in such a way that we stop falling back on the simplified Chinese and only redirect users when the stroke order has been checked. --Swift (talk) 05:26, 4 November 2008 (UTC)[reply]
I wanted to mention that while the stroke order for this character in Japanese is the same as in simplified Chinese, the direction of third stroke is the same as the traditional Chinese diagram -- that is, from left to right. --Jayme Antonio12:48, 2 March 2009 (UTC)[reply]
The bug/feature request in there (on bugzilla.wikimedia.org) since end 2004.
The function have been create-coded (2008), tested (end 2008), integrate to mediawiki (Jan. 2009), and now need to be ACTIVATE (turn "ON"). This is in the hand of some few KEY programmers (Brion, etc.) It sadly seems that for such "turn on" task, they make this once every 2, 3, 4(?) months or so, to turn on all new functions together. That's pretty slow~ I agree.... But I admit that this rename function will have HUGE consequences on commons. And may need to be think slowly, and clearly.
Hi Christoph, you are right. Unfortunateley a lot is wrong, or at least could be better, all around categorization in the commons, as i said somtimes.
Depending the files you metioned, there happened something: the template (SOlicense) contains some categorization, but a tool (the BOT BotMultichill) erroneously set the flag for missing cat. And somebody tried to heal that by adding the complete wrong "order.gif"-category.
After starting a discussion at User talk:Gmaxwell#Help on editing protected Template:SOlicense, to whom I came because of the protected template, my knowledge increased in the meantime. Now I consider it a more general question. My ideas for categorization may lead to overcategorization.
To use as an examp=le Radical 021: If a picture is in Category:Radical 021, and as well in Category:匕, and Category:匕 becomes a sub-category of Category:Radical 021, it's overcategorized.
Therefore we should find a solution and decide whether
to have a structure without subcats, each pic in both cats (with expansions of templates ACClicense and SOlicense), or
to have pics either in [Radical 021] or in [匕] (as they are now, without template expansion), and then to make [匕] a sub-cat of [Radical 021] (not otherwise!), preferable by a BOT.
I added your requested optional category argument to the protected template; I'll leave it to you and the other participants here to determine the ideal categorization system. I'd personally not worry too much about over categorization, at worst it is still less harmful than under-categorization. --Gmaxwell (talk) 10:09, 4 July 2009 (UTC)[reply]
Thank you, Gregory. I won't worry, but I am still thinking we need a schema which way we should go. I thought it useful to show in the description the radical-number, as i inserted it at ACClicense (...depicting the stroke order of the character {{{1}}}{{#if:{{{5|}}}| (Traditional '''Radical {{{5}}}''')}} in [[:Category:{{{2}}}{{{3}}} stroke order ...) --Sarang (talk) 15:55, 4 July 2009 (UTC)[reply]
Having just stumbled across this amazing project through Wictionary and seeing all that has been done, I feel this is well deserved:
The Teamwork Barnstar
Awarded to all – past and present – who have come together to work on this Stroke Order Project and its seemly endless line of tasks for the betterment of and added knowledge for Wikimedia. — CobraWiki( jabber | stuff )05:58, 23 October 2009 (UTC)[reply]
Unfortunately, I only speak English or I would pass my praise on to everyone doing the same work in the other languages. If you can translate, feel free to pass this Barnstar on to the other language groups who are working on this project. Thank you for all of your continuous hard work. — CobraWiki( jabber | stuff )05:58, 23 October 2009 (UTC)[reply]
.svg)
Done by User:Tauwasser[8]. --Swift (talk) 07:35, 30 January 2009 (UTC)
![[6]](http://www.michael.spertus.name/kanjistrokeorder/kanji_links.gif){kind=link}
![[8]](https://commons.wikimedia.org/w/index.php?title=File:%E5%8F%B0-bw.png&diff=18017798&oldid=14766546){kind=link}
