Dette emne er markeret som "i hvile"—det seneste indlæg er mere end 90 dage gammel. Du kan vække emnet til live ved at poste et indlæg.
1Choreocrat Første besked:
At the moment when I'm adding I'm using the standards from my university catalogue (one of the ones you can search through but not containing my books). Are there any hints about entering them?
Also, as far as I can see there's no CJK (Chinese/Japanese/Korean) language support on LibraryThing. Is there any chance of that happening at some stage?
On the other hand, it's good to see that there are some people with a Chinese interest - some people have translations of Lu Xun and Ba Jin, among other things.
The biggest problem is finding books at add time. There are some high quality library databases out there, and the work has been done to identify all the connection details (for instance, here). But no action has been taken to add these libraries. The fundamental problem is presumably that there is no economic motivation, since there is not really a potential user base in China and the odd linguist or Sinologist does not really count.
So, the best bets are in the States: university libraries, the Library of Congress, and the Boston Public Library. The strategy I use is first to find the book in WorldCat and then see which library it lists that is also a LibraryThing source. Even with this, the majority end up manual entry, unfortunately. But even then, I aim to copy and paste the data from some OPAC rather than just transcribing it myself.
What specific CJK support did you imagine LT might have? I haven't actually noticed anything special in Chinese language OPACs. Have you? I think the problem is more generally a lack of support for non-Latin Unicode.
It's better than no support, which would be like refusing to accept those characters, or mapping them all into ?. In the early days, there were a couple of times where all the non-Latin titles got zapped, but those are thankfully behind us. The data you enter is preserved.
As of recently, Search within one's own library has been fixed to match. So, if you search for 話 it finds something. However, this does not work when searching globally.
When matching works, LT ignores non-Latin, which means that all books with only foreign scripts look the same to it, which is of course a mess. I have also seen strange behavior when matching non-Latin author names.
Touchstones in Chinese work as of recently. Like: 美国地图册.
Alphabetical sorting of non-Latin is still unpredictable. But even library interfaces that work seem to just use the underlying Unicode code sort order.
Non-Latin tags probably do not work. Best to just avoid them.
Based on those limitations, the strategy I use is to always include transliteration as part of the title and to only use transliteration for the author name. If you do find a university library that has the book, it will be using ALA romanization, so that's how it will start out. I then add the original script by hand (followed by the transliteration in brackets). I put the original script author in the Summary field. Modern libraries will have the original in a MARC 880 field. For instance, you can see it if you click on "See all MARC records" for Beijing gu cha ming si. I have asked on several occasions for this to be automated, but nothing has come of that yet. Anyway, this way searching and sorting can be done via the transliteration, which is better than nothing.
I am not saying this is the right way to do it, but it has worked okay for me, given what we have to work with.
I have no idea what the titles actually are saying, but as long as the folks in the building can read them, I'm happy.
He told me that most American computers treat orien-tal characters as graphics - and there are 14,000 you should know to be considered literate.
Our computers are so old that printing from the Chinese language computer - or even a Chinese language website - has overloaded our printer. I think I'll hit Craig's list and see if I can find someone getting rid of a high-capacity printer or computer.