Depends on the font. In many cases, a default font is selected such that it shows Latin characters rather well and other alphabets ok, but support for non-alphabetical unicode characters may be spotty. You can see all these characters if you use a font which contains a wide selection of unicode characters, but you may not want to switch to it as the default. Alternatively, you may see some of these characters in fonts which are designed for specific portions of unicode.
All this has little to do with the browser or operating system. As long as you have a font installed and a browser that can use the font you select.
Another way to control the looks is, of course, to specify the fonts on the web page in html or css. (See font-family.)
Better yet- Twitter fully supports other languages such as Arabic and Japanese rather well (or at least AFAIK).
While it seems simple, loads of programs and services don't support them well. Can you make a HN account with a name in one of those languages and have it show well? (maybe you can, but I'm unsure)
There are plenty of Unicode symbols that can be used in web design as an alternative to images. Combine that with CSS gradients and you can cut your page load times by a lot.
'Amazing' is unjustified hype. They're much like Wingdings font characters from ten or twenty years ago.
Particularly unimpressive are the pound sign £ (UK currency, available anywhere anyway), and ı which renders as a featureless vertical bar on that page but a subscript bold number one when I paste it here. Which is about the kind of awkwardness I expect with unicode - I don't know what it is or how to pronounce it or which services will accept it or which clients will display it in which ways. Unicode is a proper fallen log in the road, it's like someone dumping a Курица word in the middle of a sentence - I don't know the language or the script, I can't pronounce it or remember the shape accurately and can't mentally work with it in any useful way. Except it's worse because while ♌ might be a word or character or sound in some human language, ✺ and ❀ are not.
All Unicode characters do have actual descriptions. And always a reason why they are in there. If you don't misuse any character you won't have any problems. Why the Unicode hate?
I don't know for sure - bad experience with it once, perhaps? I get fed up of IT, particularly stuff which is massive and complicated and finicky, particularly when it was designed by committee and foisted upon me and I don't benefit from it.
Other people just don't see the virtue in keeping their stuff simple so that it keeps my life simple. They should care less about showing off their "ability to put a telephone in a font" chops and more about how much I don't want to deal with whether the telephone is upper or lower case or the consequences of some wag putting it in their contact field and causing poorly written client software to crash when displaying it. Actually, that hasn't happened to me yet, but I'm just waiting for the phonecall where someone's web filtering software meets unicode domain names and falls flat on its face. "What site are you trying to go to?" "Some unpronouncable symbols".
When we/you/everyone can agree how to display tabs then we'll be getting somewhere and we can address the little squares of undisplayability.
It's not just unicode though, most sofware and hardware feels the aeriform wrath of my grumpy scowling.
