The Register poses the interesting question Is the web's love affair with PHP over? They quote an EDC survey that suggests that developer takeup of PHP, Perl and Python have been slowing down in the past year. I don't know who they surveyed, but here are some possible reasons from my own observation:

There is no doubt that PHP has a place in the enterprise. PHP provides a good way of developing interactive HTML user interfaces, be it for a small business or a large one like Lufthansa, who are using it for online-ticketing. But perhaps expectations have been unrealistic. PHP is not a general purpose language: don't expect to do intense number-crunching quickly or be able to write device drivers with PHP. And PHP is designed to work in a certain way: if you bring your preconceptions to the table, expecting that you can build a multi-threaded web application with persistent object store with PHP, then it will not work.

But even if you understand the above pros and cons of PHP, there are other potential show stoppers. I admit I don't have the answers to all the questions below:

1. Where are the web frameworks that work well with the corporate standard database, be it Microsoft SQL Server, Oracle or DB2?

   I'm not aware of any popular SQL Server, Oracle or DB2 web framework for PHP. There are probably a few such beasts developed in-house, but not freely available. For example, the company I work for has what I like to think is a pretty good web framework that runs on Oracle - but it isn't open source.

2. Are there PHP software standards apart from the PHP extension libraries and Zend API? Is there software certification such as J2EE?

   No and no. Well ok, there is PEAR the PHP code repository, but the quality of PEAR is so uneven that you might prefer to write your own code from scratch.

3. How do i interface with large commercial apps such as SAP, Peoplesoft, Oracle Apps, Siebel or JD Edwards?

   Things are getting better here. There are tools such as SAPRFC for calling ABAP (which is very good), the upcoming IBM and Oracle PHP integration packages, existing SOAP interfaces, and SQL for extracting database information. However for most popular applications, PHP support is an after-thought.

I don't think that the Web's love affair with PHP is over, but I do think that for larger businesses downloading and customizing PHPNuke would be a pathetic joke. Big web apps require a team of good programmers; and that means that PHP growth in larger companies will be limited by the scarcity of skilled PHP developers who can create high quality apps from scratch, and who can convince management to let them do so. It is no accident that Yahoo's adoption of PHP coincided with their hiring of some of the most talented PHP developers in the world.

It is certainly possible that this is an open source problem, and not merely a PHP one. Python and Perl could be suffering from similar issues to some degree or another (framework limitations, problems interfacing with existing commercial apps, lack of standards). But the difference between PHP compared to Python and Perl is that the two languages are sufficiently general to be used in many niches. PHP remains primarily used for web sites.

PHP will continue to be popular with startups, and small and medium sized companies. But in the corporate world, I think growth will be slower.