I have something against PHP, but there's a deeper point to be made here. A language that does nothing more than reinvent a popular language will not succeed, even if it does a better job. To succeed, a language has to provide a good solution to a problem a lot of developers are having. Some examples:

* C - assembly and machine languages are not portable

* PHP - the barriers to entry for server-side web programming are too hugh

* Java - C++ becomes too error-prone with large teams

* Clojure[0] - there's too much complexity in programming that isn't inherent to the problems being solved

* Ruby, Python - Perl is really useful, but a little too messy

[0] Calling Clojure popular may be pushing it, but there seems to be some momentum building behind it.