问候大家，
我想知道是否有人注意到这种不一致并知道如何解决它。
示例：http: //jsfiddle.net/p5G2N/
前提：
多个单词由未编码的 & 符号连接；没有空格。
<p>Goodevening&goodnight</p>
结果：
Chrome/IE：无法将字符解释为 & 符号，从而破坏了预期的大写样式。显示为：Goodevening&Goodnight [不正确 - 注意晚安的大写 G，因为它是一个单词，所以不应该大写]
Firefox：设法正确解释未编码的＆符号并将字符串视为单个单词。显示为：晚安和晚安[正确]
问题：
如何让 Chrome/IE 尊重样式（大写）？
& 应该用过&。从W3 规范：
作者应该使用
"&"(ASCII decimal 38) 而不是"&"避免与字符引用的开头混淆（实体引用打开分隔符）。作者还应该"&"在属性值中使用，因为在 CDATA 属性值中允许字符引用。
TL;DR: Your expectation is incorrect. goodbye&goodnight has two word-separation points, one on each side of the ampersand. Whether the ampersand is encoded or not is irrelevant.
As far as I know, CSS doesn't fully specify what a "word" is, but there is a recommendation to use the Unicode standard word separation algorithm, which you can find here (UAX29).
An informal summary is that a word is a sequence of letters, numbers or "Connector_Punctuation" symbols (ties), and possibly containing "MidLetter", "MidNum" or "MidNumLet" symbols (there's a list in the referenced document), depending on the immediate context of the symbol. & is not in any of those categories, so a UAX29-conformant word-separation algorithm should split words before and after an &.
A word separation algorithm may take language into account. Indeed, it may do just about anything, but it's supposed to be unsurprising for a native speaker of the language. Non-programmers would probably be surprised if word&word were considered one word.
如果要&在 HTML 中使用 - 符号，请&改用。