|
Post by kerry on Feb 3, 2009 7:09:14 GMT -5
A problem I had for a while was that my website was generating google indexes for my Chinese names in the wrong order. It was following the western convention of putting surname last. Hence Googlbot was seeing my gg-father as 文粹蔡 instead of 蔡文粹. (For this exercise, I've bolded the family name.)
I've now got Chinese names in the right order, transliterations of Chinese names in the right order; surname lists by 1st letter of surname working correctly for Western names (if any), transliteration of chinese family name or Chinese character.
I'm just waiting for the google index to refresh over the next few days after which you should be able to find my grandfather correctly with 蔡文粹.
Essentially, this is supported by the GEDCOM standard but needs to be implemented by your particular software. I'm happy that phpGedView supports this. Perhaps some of the the other site users can verify if this is supported or not. For instance, Geni doesn't support one letter surname search through it's directory - they mandate at least 2 characters for a search.
|
|
|
Post by Doug 周 on Feb 5, 2009 18:59:08 GMT -5
{To view Chinese characters: Go to the top of the browser, click on [View], then click on [Character Encoding], then click on [Unicode(UT8)]}
Kerry, You bring up a very good point about accessing your GEDCOM data from the internet. I tried to test finding my wife's GFather's Chinese name and it was unsuccessful finding data in Geni, MyHeritage, and Dynastree. PhpGedView is a powerful program to allow searches from Google. It looks like you have configured it correctly to be accessible. This leads me to an idea about the CLAN PROGENITOR REFERENCES. When I look thru the Zhou 周 (my surname) Zupu data ( siyigenealogy.proboards28.com/index.cgi?board=clanprogenitor&action=display&thread=843 presented by Kevin), I try to match a name in my Jiapu family tree with the images referenced in the forum. Being illiterate in Chinese, my eyes become crossed when I try to compare the characters. I usually give up before I become too nauseated. If we can scan these characters as Chinese text using an OCR program (for example users.belgacom.net/chardic/cocr2.html ), would you be willing to be the Webmaster of a PhpGedView site that would store these Zupu Clan data? Would other moderators of the various surnames be willing to convert their images into Chinese text? Is this be the correct use of the CLAN PROGENITOR REFERENCES? Doug
|
|
|
Post by Henry on Feb 5, 2009 19:44:09 GMT -5
Doug,
I believe all the image charts that you see on the Clan Progenitor References were all Chinese text before they were captured for posting as a chart.
Philip's information was on a spreadsheet, the Tan lineages were Chinese text in a Word file.
So having them in Chinese text should not be an issue.
But comparing Chinese characters side by side usually works for me.
Henry
|
|
|
Post by Doug 周 on Feb 5, 2009 20:25:52 GMT -5
Henry,
You are correct as usual. It is easier to print out the published Zupu, enlarge the output, and then compare the characters side by side. I was getting lazy and thinking of doing a computer-google-like search. A computer search is better for recurrent searching. However, once you find a match and are able to hook up to a branch, there is little need to do another search.
Doug
|
|
|
Post by Henry on Feb 6, 2009 7:13:50 GMT -5
Doug,
You are also quite correct - sometimes just trying to find that first character in that sea of Chinese characters is very intimidating. Thus, the computer text search to find the first Chinese character can save a lot of time and frustration. Using a computer search to find the next and successive ancestors makes complete sense too.
But, you might want to check on the character first to see if you want to do a computer search on either the “traditional” or “simplified” version of the Chinese character – you may have to do a double search, because a search with no result (hit) may be because the search was for a traditional instead of simplied version of that Chinese character.
Many of the older family/clan genealogy books are handwritten and usually written in traditional Chinese characters and even with the excellent COCR2 program – this may not be good enough to convert them into digital Chinese text.
There is kind of a chicken/egg problem of what to do first. Do you scan the entire book – hundreds of pages and then convert all the thousands of images of Chinese characters into searchable Chinese text and then do computer searches or do you find the ancestor manually by fumbling around with a copy of the Chinese character in hand until you find the links and determine a complete line of descent for the pedigree from the first generation to the present.
The manual method may not be s bad as it might seem – because what you are really doing is following each of the lines of descent until you find the particular ancestor and using the side-by-side image comparison. In my experience, I have had to determine the complete lineage by linking all the various lineage segments through the ancestral nodes, i.e., the clan progenitor references. Once I have this complete pedigree lineage then I go back and “digitize” the image of the Chinese characters and convert them into digital Chinese text.
Now, this is for people like me who cannot read or write Chinese – but maybe in the next life, I will return as a Jinshi scholar and not have to deal with being illiterate in Chinese.
Increasingly, the majority of the recent Chinese family/clan genealogy books that I encountered after year 2000 are published / printed books. The printed form is much clearer and easier to scan and convert into digital Chinese text, than the handwritten family/clan genealogy books. Also, it appears that revision / updates of recent family/clan genealogy books are being performed by a group of clan genealogy experts that are trying to resolve the discrepancies and also extending the family / village genealogy lineage all the way up to Huang Di, the Yellow Emperor, circa 2697 BC.
Henry
|
|
|
Post by kerry on Mar 26, 2009 6:05:04 GMT -5
Apologies for not replying earlier.
Re scanners: I've found scanning to be very difficult. YMMV
I haven't made the entries from my genealogy searchable so that I can find them - I'm quite intentional about making them searchable for someone else! I don't think I'd go through all the hassle if it was just for my personal use.
If your requirements are more focused on a closer audience, the "business case" to my mind is a bit weaker. If it's for personal use, then the paper form is probably better. With practice the physical location of pages and locations on a page (may!) become a useful habit.
I've really only got the direct family links back to my progenitor. I'm adding entries now and then but I'm really slow. What I hope will happen is that some other relative will find the same progenitor, recognise the early generations that are listed and add the links down to their part of the family. (and hopefully they'll be more capable with Chinese characters than I!)
|
|
|
Post by Henry on Mar 26, 2009 8:06:19 GMT -5
Kerry, Please try the little free software package that Doug had recommended: users.belgacom.net/chardic/cocr2.htmlInitially, I was rather reluctant to use yet another scanning mechanism that only worked maybe 45% of the time and with a high hassle factor - as with the Chinese pen scanner I bought etc. But, this COCR2 program is about 98 % accurate and on the scanned page - I am converting Chinese character images into digital Chinese characters that are searchable - in seconds and lots of them. This is really incredible and the most useful mechanism I have for doing Chinese genealogy research. Now, I do not have to ask my friends to type out the Chinese characters for me. For the other 2% that COCR2 cannot recognize, I still write out the Chinese characters using : www.nciku.com/COCR2, nciku, and along with the MW Snap3 program that I use to capture any Chinese text images on my computer screen - I am relatively independent in my Chinese genealogy research: www.mirekw.com/winfreeware/mwsnap.htmlI suggest that you give COCR2 a try - it will save you tons of time and effort. Other colleagues have tried it and they seem to like it a lot. The progenitor efforts are to identify clan progenitors as key points for linking lineages within a clan and provide a fully linked lineage from the most recent descendant all the way to Huang Di, the Yellow Emperor. I believe that all the Clan progenitors on this Forum are probably more than willing to share the lineages that they post on the Forum - in a digital (searchable form) with other members. Taking the clan progenitors one level higher, our most generous colleague, Dr. Philip Tan, is also providing the clan progenitors from many of the clans that diverge from the Huang Di lineage, thereby, providing the relationships between the progenitors of many Chinese clans - awesome. Henry
|
|