C

bit fields -- they aren't well specified by the language and how they work is compiler dependent and architecture dependent. It's often hard to find where a particular symbol is defined in a large mass of code, esp. if that symbol is produced by a macro. Which reminds me... The preprocessor is a rather ugly hack, amenable to all sorts of abuse. lack of standard sized integers (remedied by uint*_t lately, but there is lots and lots of old code floating around out there with custom typedefs or #defines for DWORD, WORD, BYTE, etc.) Lack of something akin to Perl's cpan.org (would love to be wrong about that one.)

编辑: 在考虑C的CPAN时，我想……我该怎么称呼这样的东西，然后想到了“ccan”，然后谷歌了一下，我看到了这个: http://ccan.ozlabs.org/

不过，它似乎还处于起步阶段。

JavaScript

Function object syntax: f = new Function( "foo", "bar", "return foo+bar;" ); (It takes n arguments, the first n-1 are arguments for the function, then nth is the actual function, in string form. Which is just silly.) Function arguments can be repeated. f = new Function( "foo", "foo", "return foo;" ); The last repetition is the only one ever used, though: f( "bye", "hi" ) // returns "hi" f( "hi" ) // returns undefined E4X should just die. My users are always complaining that it doesn't work the way they think it will. Let's face it, when you need a page and a half of psuedocode for a setter, it's time to rethink things. A standard notion of stdin/stdout/stderr (and files!) would be nice. null != undefined It's irritating to have to handle them both. Sometimes it's useful, but most languages manage to limp along fine with one.

SAS

从不有自己的想法(一切都是借来的)。 贪婪于庞大的数据集。 使用Java，但从未学习过什么是对象。 窃取Perl，但将其隐藏在其数据步骤中。 统计学家总是撒谎!

C#

我对c#非常满意，但这两个真的让我很恼火:

Constructor-based initialization for immutable classes is less convenient, less intuitive (when you read the code you don't understand what you assign to what), has less IDE backing than inline object initialization. This makes you lean towards mutable classes inevitably. I know this has been mentioned before, but I strictly have problems with initialization syntax for immutable classes. switch is too verbose. Whenever I see a situation where a switch would be proper, I'm really inclined to use an if..else if.. just because it's more terse (~30% less typing). I think there should be no fallthrough for switch, break should be implied, and case should allow comma separated list of values.

Scala是我最喜欢的语言。五件讨厌的事?容易:

Takes a long time to learn properly. I know you can write Scala as a 'better java'. That is what we used to say about C++ and C too. I agree this is an inevitable consequence of the deep ideas in the language. But still ... Methods vs. Functions: def f(x: Int) = x*x defines a method f, not a function f. Methods are not functions despite a lot of early Scala tutorial material blurring the distinction. The language tries to blur it too because if you supply a method in some places where a function is expected it is accepted. Do we have to have both methods and functions? Yes it is fundamental. But it was initially confusing to me. Composing classes or objects from mixins in the 'cake' pattern is prone to NPE's. e.g. trait X { val host: String; val url = "http://" + host } is a mixin that will NPE on instantiation, or not, depending on its position in the class declaration. The compiler could tell you if it will fail but doesn't. (In 2.7 anyway.) It is hard to diagnose the problem in complex inheritance graphs. Arrays in 2.8 rely on implicits to mesh with the main scala collection types. But implicits are not applied everywhere. An Array can be supplied where a Seq is expected. But an Option[Array] cannot be supplied where an Option[Seq] is expected. I know there are no completely 'right' ways to handle java Arrays. Type erasure. Enough said.

Common Lisp

conditions aren't classes (since classes came later), even though their interface is almost identical some of the names are just weird, e.g., flet / labels (only difference: scope), and defvar / defparameter (only difference: behavior when already defined), or any of the bit-twiddling functions (dpb, ldb, etc.) packages are ... really hard to get right -- every time I think I understand them, they don't do what I want built-in data structures and functions aren't as generic as they could be (e.g., why can't I define my own hash function portably?) multiple namespaces for functions, variables, etc. (I'm not opposed to this in principle, but CL made it too complex; Norvig has said he can't tell from the spec but there appear to be at least 7 namespaces)