VGMdb
Go Back   VGMdb Forums > VGMdb Site Related > Questions and Comments
Register FAQ Calendar Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old Sep 9, 2012, 03:45 PM
realnabarl realnabarl is offline
Junior Member
 
Join Date: Aug 2010
Location: China
Posts: 22
Default About '~' and '〜'

Which should we use in Japanese track title?
Reply With Quote
  #2  
Old Sep 9, 2012, 04:22 PM
dancey's Avatar
dancey dancey is offline
Trusted Editor
 
Join Date: Dec 2007
Location: New Jersey
Posts: 1,294
Default

Do you mean ASCII '~' and Unicode '~'? Never ever use Unicode in English track lists (because of freedb). Other than that it's just determining which is being used.
Reply With Quote
  #3  
Old Sep 9, 2012, 04:54 PM
realnabarl realnabarl is offline
Junior Member
 
Join Date: Aug 2010
Location: China
Posts: 22
Default

Quote:
Originally Posted by dancey View Post
Do you mean ASCII '~' and Unicode '~'? Never ever use Unicode in English track lists (because of freedb). Other than that it's just determining which is being used.
Yes I know the first one should be used in English and the second one should be not. The first charater I said is the second one you said, and the second one I typed is another one.

Well, I just got a wiki page for these characters:
http://en.wikipedia.org/wiki/Tilde

Now it is much clear.

These two can be found in many pages in freedb, VGMdb and MusicBrainz in various album titles and track titles:
~ U+FF5E FULLWIDTH TILDE
〜 U+301C WAVE DASH

They are very similar and both takes two spaces.

This page might be the key:
http://ja.wikipedia.org/wiki/%E6%B3%...82%B7%E3%83%A5

My Japanese is not good, it would be nice if someone would like to translate it.
Reply With Quote
  #4  
Old Sep 10, 2012, 07:28 AM
LiquidAcid LiquidAcid is offline
Trusted Editor
 
Join Date: May 2008
Posts: 1,563
Default

Quote:
Originally Posted by dancey View Post
Do you mean ASCII '~' and Unicode '~'? Never ever use Unicode in English track lists (because of freedb). Other than that it's just determining which is being used.
What's the problem with Unicode and CDDB? According to the CDDB proto specs it supports UTF-8 in level 6.

Reference
Reply With Quote
  #5  
Old Sep 10, 2012, 08:07 AM
dancey's Avatar
dancey dancey is offline
Trusted Editor
 
Join Date: Dec 2007
Location: New Jersey
Posts: 1,294
Default

Quote:
Originally Posted by LiquidAcid View Post
What's the problem with Unicode and CDDB? According to the CDDB proto specs it supports UTF-8 in level 6.

Reference
There's no problem with the specification, the problem is expectation of client support. Anyone using freedb/cddb to query vgmdb for English tracklists is likely not to support Unicode in their filenames, tagger or client. Forcing Unicode characters on non-unicode OS's or OS's that are not natively Unicode is bad programming, bad practice and bad in general.

If you want Unicode, use Japanese.
Reply With Quote
  #6  
Old Sep 10, 2012, 08:16 AM
LiquidAcid LiquidAcid is offline
Trusted Editor
 
Join Date: May 2008
Posts: 1,563
Default

Quote:
Originally Posted by dancey View Post
There's no problem with the specification, the problem is expectation of client support.
Clients can set the protocol level. So there is _no_ expectation here.

EDIT: Ah, I see you mean. So this would require a on-the-fly conversion of UTF8 to ASCII on the server, and that's not going to work unless one specifies to what non-ASCII characters are mapped to.

Quote:
Originally Posted by dancey View Post
Anyone using freedb/cddb to query vgmdb for English tracklists is likely not to support Unicode in their filenames, tagger or client. Forcing Unicode characters on non-unicode OS's or OS's that are not natively Unicode is bad programming, bad practice and bad in general.

If you want Unicode, use Japanese.
I was wondering if anyone here is actually still using an OS which doesn't support unicode. Even W98 has Unicode support, and I honestly doubt that this is still in wide use.

Same applies to filesystem. FAT32 supports Unicode, NTFS as well.

Last edited by LiquidAcid; Sep 10, 2012 at 08:20 AM.
Reply With Quote
  #7  
Old Sep 10, 2012, 08:59 AM
dancey's Avatar
dancey dancey is offline
Trusted Editor
 
Join Date: Dec 2007
Location: New Jersey
Posts: 1,294
Default

Quote:
Originally Posted by LiquidAcid View Post
Clients can set the protocol level. So there is _no_ expectation here.

EDIT: Ah, I see you mean. So this would require a on-the-fly conversion of UTF8 to ASCII on the server, and that's not going to work unless one specifies to what non-ASCII characters are mapped to.
No, there should be no on-the-fly conversion. There shouldn't be any Unicode in an English tracklist. Period.

Quote:
I was wondering if anyone here is actually still using an OS which doesn't support unicode. Even W98 has Unicode support, and I honestly doubt that this is still in wide use.

Same applies to filesystem. FAT32 supports Unicode, NTFS as well.
Support for and actual implementation are two different things. If you don't have a Japanese code page installed then you're either going to get '??' characters (if your OS doesn't support Unicode or it's using the old Windows ANSI file system calls) or you're going to get two one byte characters instead of one two byte character, like '`%', etc. Regardless of whether the OS supports it, it's still up to whatever application to support Unicode and code pages, and on top of that you have to configure your OS to have the code page installed. You can't guarantee that always happens, so stick with ASCII because that is the vast, vast, vast majority of English users will be using.
Reply With Quote
  #8  
Old Sep 10, 2012, 05:02 PM
Datschge's Avatar
Datschge Datschge is offline
Trusted Editor
 
Join Date: Mar 2008
Posts: 704
Default

Quote:
Originally Posted by dancey View Post
There shouldn't be any Unicode in an English tracklist. Period.
So the "English" tracklist should always be limited to the ASCII character set? I think this needs to be set as a rule as I'm pretty sure there are already tracklists called "English" while actually having track titles containing characters of other Western languages not included in the limited 7bit ASCII set.
Reply With Quote
  #9  
Old Sep 10, 2012, 07:18 PM
dancey's Avatar
dancey dancey is offline
Trusted Editor
 
Join Date: Dec 2007
Location: New Jersey
Posts: 1,294
Default

Quote:
Originally Posted by Datschge View Post
So the "English" tracklist should always be limited to the ASCII character set? I think this needs to be set as a rule as I'm pretty sure there are already tracklists called "English" while actually having track titles containing characters of other Western languages not included in the limited 7bit ASCII set.
It really should be. I think there might be a valid argument for extended ascii characters like stuff with umlauts, etc, because they're still 1-byte, but not anything 2-byte.
Reply With Quote
  #10  
Old Sep 18, 2012, 07:38 AM
kami68k's Avatar
kami68k kami68k is offline
Member
 
Join Date: Sep 2007
Location: Germany
Posts: 79
Default

Quote:
Originally Posted by LiquidAcid View Post
I was wondering if anyone here is actually still using an OS which doesn't support unicode. Even W98 has Unicode support, and I honestly doubt that this is still in wide use.
Well I got this cheap mp3 player which I always carry with me, and although it does support unicode, I could imagine this is an area where you can still encounter non-unicode systems. I dont really know though :-)
__________________
www.vrc7.net
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump