PlaneShift

I can confirm that for linux (gentoo here).
The windows version tho is able to use the umlauts, as far as I remember, but they won't get displayed on a linux client (makes /tell with donari kind of hard, since he uses umlauts all the time :D:D:D)

the font ps uses [liberation sans regular] does not support unicode. If it shows up in any system it is bugged. PS is ment to be an english only game and will not as far as i know support other languages fully to this extent.

PlaneShift doesn't suppport Unicode at all, this is not font related. This means that the only special (ascii) characters PS can show are !"#$%&'()*+,-./0:;⇔?@[\]^_`{|}~ (I think these are all). If some ascii characters do not show on Linux that might be a bug, but it may also be a local issue.

IMHO PS should support unicode. I mean, PS is not an _english_ game, it is an international game and in /tell, /group and /guild other languages than english are not that uncommon. Some words might even have a different meaning, depending whether the normal letter or the umlaut is used. Some of the ingame languages also use umlauts from spanish and french, which some people might want to use.

Changing this from a bug report to a feature request might be useful :)

PlaneShift is an English game, really. And what in-game languages use umlauts? Enkien and sorts are still fan-made. Perhaps some day unicode support will be implemented. It's one of those "in the long run" things I think. In the mean time I don't think the devs will worry you can't talk about das frühstück with your friends ;)

Besides all this, are you sure that umlauts (^^ or "kitty emotes") can be typed and shown on Windows, but not on Linux and that this is a issue not only you are having? If so, open a new bug report.

I disagree that PS being an English game has anything to do with the question whether it should support international characters. Just take a look at anything Tolkien wrote - many of the names require accents or tremas and gain a lot of their sense of wonder from that - Húrin, Lúthien, Udûn, and Oromë are simply different from Hurin, Luthien, Udun, and Orome.
So, while accented characters may be less important for normal chat, it's really a pity that they aren't practical in names.

For background information on what umlauts are and how they are used, see http://en.wikipedia.org/wiki/Umlaut_%28diacritic%29 .

Re the ^ sign: I specifically mentioned it as working in my original message, so "kitty emotes" can be typed.

I just prepared a text file with äöüÄÖÜß in it and loaded it into a book.
äöüÄÖÜ unsurprisingly display as blanks, ß surprisingly displays as ß.
All of äöüÄÖÜß are counted as two characters when stepping through with the cursor, indicating PS is not aware of multibyte character sets (this is consistent for "no Unicode support").

So it seems that there are several aspects, all related to handling multibyte characters:
1. The Linux client of PS filters or does not get multibyte keyboard input (at least on an UTF8-configured Ubuntu like mine).
2. The Windows client does not filter (at least on German Windows - likely from what Michael says but unconfirmed; this might be because Windows is using a single-byte character set for text input in Western locales).
3. The font used in Linux does not have äöüÄÖÜ, or it uses a different encoding than Unicode. It happens to display a Unicode ß as a ß.

Marking new, as this is a feature request. Whether ß or ö should be possible in PS is upto the devs. I think talking German in tells is not a problem at all, so having these characters may make sense. In the meantime I suggest just to use ss and oe instead.

What happens if you set <fontname> in ../PlaneShift/data/prefs.xml to a unicode font?

the server does not support it so it would have to be updated as well as i understand or more of the client.

As far as I can tell from a user perspective, it's a server issue.

1) Clients need to associate keyboard input with code points, and to make sure that each code point gets displayed as the same glyph across Windows, Linux, and OS X.
Since the mechanisms differ across platforms, this needs to be done in client code.
I don't know whether CS is Unicode-aware though. If not, extending it for Unicode support may be a lot of work.

2) The server just accepts byte streams from clients (when somebody types a message or writes to a book), and sends them to clients (when the client wants to display a message or a book's content).
As long as the server does not look into or manipulate the contents of the messages, it should be largely unaffected.
(I think it removes swear words though. Unchecking "filter swear words" has no effect for me.)

@Bilbous: It's possible that the glyphs would display correctly. Still, the edit window for books will choke because it does not know that deleting a character may require deleting one, two, or three bytes. In other words, you can create invalid Unicode sequences when editing, which may mean the client might display "funnily" or even get a buffer overflow.
So it might look better to try that, but it would create bugs.