I checked my UTF-8 document somemore bits, and found Wiki server may
transrate '0x20' to '0x0a', which destroyes Japanese text.

Like this,

I wrote (say, /mnt/wiki/22/index.txt),

もしもPlan 9で日本語を使いたいなら、私

the byte sequence of abobe sentence will be

[E3 82 82][E3 81 97][E3 82 82][506C616E2039][E3 81 20][E6 97 20]
[E6 9C 20][E8 AA 9E][E3 82 92][E4 BD 20][E3 81 84][E3 81 9F]
[E3 81 84][E3 81 AA][E3 82 20][E7 20 81][E9 81 20].

The runes which has '0x20' in that UTF-8 sequence are destroyed,
but the others are written right. Here, I enclosed one Kanji
character(rune) in [].

Couldn't we stop this translation from '0x20' to '0x0a'?

Kenji