CIFS vs. NFS and other filesystems (was Client for Samba Networks)

[Steve Langasek]

On Tue, Dec 18, 2001 at 11:00:46AM -0800, Jeremy Allison wrote:

> Multi-length characters suck. Period. Hard to program, lead to buffer
> overruns, hard to traverse in reverse....

> UCS2 is at least a fixed length encoding. That's why I was so cross
> with Apple for adding multi-length ucs2 encoding in their version of
> Samba :-(.

Meaning that they implemented UTF-16, or some other variable-length
Unicode encoding using 16-bit widechars?

> If 16 bits isn't enough, then go to 32 bit. 2^32 characters should be
> enough so that every weird-ass language (including Klingon) that needs
> compose character pairs can select a unique codepoint for each pairing...

> Human languages just aren't that complex.

In what sense is traditional Chinese[1] not a human language? :)  

Steve Langasek
postmodern programmer

[1] "Ah," say the Unicode apologists, "but all modern Chinese documents
are written using /simplified/ Chinese!"  True, but how would we feel if
Unicode could be used for all human texts /except/ for Shakespeare,
since he's dead and therefore obviously of no concern to computer
standards?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
Url : http://lists.samba.org/archive/samba-technical/attachments/20011218/ee8f323b/attachment.bin