PHP4, The Energizer Bunny?

-> Digg
-> Reddit
-> Del.icio.us
-> Dzone

How long can PHP4 keep going? Well, the way things currently stand, the answer could be “forever”. According to recent statistics, PHP4 still holds a whopping 90% of the market share. It is no wonder that the vast majority of PHP applications have either not added PHP5 support or have continued to maintain themselves as primarily PHP4 applications.

I’ve heard time and time again that all we need is that “killer” PHP5 application to push everyone forward to the next version. Well, where is it? I’ve been waiting for two years to see this “killer” application (yes, PHP5 was released 2 years ago). The problem with the “killer” application is that every application you can imagine wanting has already been done in PHP4, and many done very well. To come up with an application that will make people want to switch to PHP5 would take something no one has ever seen before and would excite them beyond belief.

The next problem we run into is that PHP6 is coming. So, where will that leave us? The positive thinkers in the PHP world say that PHP6 will be the reason that people leave PHP4. They say that we will have a large chunk of people that will suddenly leave behind their PHP4 ways and jump straight to PHP6. Yeah…right.

What it all boils down to is that PHP4 is still currently supported. There have been a plethora of maintenance and bug fix releases for it over the past couple of years. “New feature” development has been frozen for the PHP4 branch for about that period of time, but that does little to persuade people to stop using it. As long as bugs are fixed and security holes are patched, people will continue to use PHP4 and application development teams will continue to code for and support PHP4.

What we need is for the PHP core developers to finally agree to simply stop supporting PHP4. I know I will be thrown under the bus for this, but they could learn a great lesson from Microsoft on this one. Microsoft clearly publishes when they will stop supporting something, and then they stop supporting it. (Yes, sometimes Microsoft extends the deadline, but let’s not focus on that.) The PHP core developers need to set a deadline, say January 1, 2007, and then simply stop ALL releases (bug and security included) of PHP4. That would give everyone time to upgrade to PHP5, do their testing, and be done with everything all at once. Until such an ultimatum comes, PHP4 will remain the reigning champ and new PHP versions, no matter how exciting or groundbreaking, will not even be worth the development time invested in them.

1. 1. manofsteel says:
      Jul 7, 09:40 pm
      
      

      I think when Cpanel and Red Hat Enterprise make the switch, PHP 4 will be done. Until that happens, alot of people are forced to use PHP 4.

      
      
2. 1. Matt Wade says:
      Jul 8, 01:04 am
      
      

      I totally agree about cPanel. Certainly setting a deadline for dropping PHP4 bug fixes would force cPanel’s hand wouldn’t it?

      
      
3. 1. manofsteel says:
      Jul 8, 05:04 am
      
      

      The gamble of forcing people away from PHP4 is to think they will go to PHP5 rather than Ruby, Python, or the other players in the crowded space of scripting languages. I wonder how long PHP can stay relevant with the inability to get people to migrate to PHP5.

      To extend my earlier comment, I think it would be suicide for any killer application to only have the ability to run on PHP5. (Assuming this is an application such as IPB that users can download and install. Otherwise people don’t care what language the application is built from.) To do so greatly reduces the number of people that can run your application. In the case Cpanel, you would have a large portion of the market that would have no ability to use PHP5. With RHE, you cannot install PHP5 because you break your service agreement.

      Unfortunately there is not alot of alternatives to Cpanel. There is Plesk and Directadmin, but most people are comfortable with Cpanel. Unfortunately, many webhosting customers are tech challenged, but have to force themselves to learn enough about hosting to get something done.

      Maybe the killer application would be a good alternative to Cpanel. Something that gives the great control, but also delivers the newest tools available.

      
      
4. 1. Jason says:
      Jul 8, 10:45 am
      
      

      It’s a bit dated now (published 10 months ago), but I wrote an article for Linux Magazine that charted the progression of PHP 5 adoption, includes some statistics pertinent to the U.S.’ largest web hosting providers, and key projects. While some of the data is surely out-of-date, you may find it interesting.

      http://www.linux-mag.com/content/view/2175/

      (free registration required)

      
      
5. 1. Rutger says:
      Aug 4, 05:35 am
      
      

      You cannot force people to move on to a new version. Microsoft can do that because no one can fix Microsoft products except Microsoft itself. If support for PHP4 is dropped, it will probably fork and we end up having two php’s.

      
      
6. 1. SwiftCoder says:
      Jul 17, 04:07 pm
      
      

      Not too mention that cheap webhosting with PHP5 is still hard to find… while PHP4 is on every $2/month host arround.