There is an announcement on the official PHP website about PHP 4 reaching end of life at the end of this year, with no more development beyond the 31st December 2007. Security patches will continue for a further eight months after this date, but after August 2008 PHP 4 will be no more.
As one can imagine, this decision has caused a strong split of opinion within the PHP community. I suspect that many users will be extremely annoyed by the announcement, even though their applications should work under PHP 5 without any changes. If they don’t, chances are that it’s a problem with their script, although I doubt they will see it like that. On the other hand, GoPHP5 is waving the flag for the ‘move to PHP 5’ group, listing projects and hosts which have pledged to move to PHP 5 by February 2008—although this is something of a moot point now that we know that PHP 4 will be discontinued anyway.
My personal thought on all this is that both the PHP team and web hosting companies are to blame for the slow transition to PHP 5. The latest major version of the software has been out for three years now, yet the PHP team has done very little to push system administrators to upgrade, nor have they provided any major incentives for users to want to move to the latest version and put pressure on their providers to upgrade.. Web hosting companies have also failed to even offer PHP 5 in many instances, leaving users with no option but to continue developing for version 4.
Matt has also weighed in to the debate with his recent post, On PHP. I agree wholeheartedly with his point about the PHP core team killing off a popular product (for all its faults, PHP 4 has undeniably been a success) without thinking about why people haven’t upgraded to PHP 5. This could be laziness, in which case the PHP team needs to take steps to ensure that there are incentives to overcome this, e.g. new features, improved security model and smoother updates.
I think what amazes me the most about all this though is that we really should be looking towards PHP 6 by now, yet PHP 5 still isn’t adopted by the majority of hosting companies and end users. If it’s taken this long to move from 4 to 5, how long will moving from 5 to 6 (which has lots of useful improvements) take?