RE: [PHP] 'God' has spoken... :-)

At 11:59 AM 8/15/2005, Jim Moseby wrote:

I'm (we're) still using PHP4. Mainly because there's been no reason for us to upgrade. ie, we're not doing anything that requires PHP5 (and if there is no feature in PHP6 that we have to have, we won't be upgrading to that either). I don't see the problem with this, unless it will be more difficult to upgrade from PHP4 to PHP6, than from PHP5 to PHP6. What I mean is, will upgrading to 5 now make the upgrade to 6 (or 7, or 9, or 11) easier in the future. So 5 has no features that I need right now. 6 probably doesn't either. But when PHP11 comes out and I must have it, will I be able to easily upgrade from 4? Will I expend less time/effort/risk making each upgrade as it is released that I would making a 5 version jump? Maybe its worth the time and effort to upgrade even if I don't need the features in the new release, just to make it easier to upgrade to the next one(s) down the line. Not only that, but as time goes by, community support (this list for instance) for 4 will likely diminish as others move on to later releases. Not only will the die-hard PHP4 users find it harder to get answers to their questions, but their answers to others will become less relevant to newer-version users. Just my .02 JM

A good parallel is what happened with Visual FoxPro and with PageMaker. When VFP 3 came out it broke the mould of FoxPro development, and it was a huge leap to adjust; product was not really mature and VFP 5 was great. VFP6 was even better - but a leap from 3 to 6 was almost impossible because so much had been changed in the interface and feature set. At least MSFT is doing it properly now, and announcing that VFP10 will be it, then no more. PageMaker was worse, but it's so long ago now that my memory's faulty. I think it was around PM3 or PM4 that the file format changed; if you had the earlier version, PM4 and later could not read the older files, you had to obtain a crippled version of PM3, which could only open the old version and save in the new, and then work on those files in the newer version. Of course a lot of stuff got busted. And of course, there is Visual Basic - where development stopped at version 6 and a new .NET language, labelled "Visual Basic" was introduced. So much stuff in VB 6 was broken that moving the code to the .NET language named Visual Basic is well nigh impossible. The problem with PHP 5 is that the ISP's have to be so conservative. There's no tagging mechanism which says "process these files with PHP5, use PHP 4 for everything else." So, based on experience, better to move early than later. I suspect a jump from PHP4 to PHP6 will be huge -- the problem is to move the ISPs. Regards - Miles