A Brief History of PHP

[Vin DSL]

Interesting read...

PHP: Three Versions, One Promise
December 26, 2005
By Sean Michael Kerner

PHP has long been a favorite dynamic programming language for Web developers. After 10 years of growth, 2006 looks to be a banner year for the open source language as its collaboration framework, partner ecosystem and the language itself are expected to grow and gain new users.

Look for a confluence of three distinct branches of the PHP language, with versions 4, 5 and 6 all likely to be out in the market in 2006. It will also be a year in which PHP will continue to challenge .NET and Java as the online language of choice for developers.

According to data from Netcraft, over 22 million domains currently use PHP. A November study from SecuritySpace.com reports that PHP is actually the most popular Apache Web server module, with some 44 percent of all surveyed servers running it. By contrast, Perl, another popular open source language, comes in at 11 percent...

Bottom line:

Three Versions: Triple Confusion?

Three versions of PHP in one year? Sounds like a recipe for confusion. Not quite, supporters say.

Hostway, a large Chicago based ISP, expects that it will support all three PHP versions when they are available.

"We make an effort to keep up with new versions of useful technology such as PHP while not abandoning those customers that still rely on older versions," a Hostway spokesperson said. Ebay doesn't see three PHP versions as an issue either.

Gutmans argued that three versions out at once is a sign of its strength.

"It's actually very much different than Microsoft, which forces people to upgrade," Gutmans said. "Over time there will be a natural migration and when too many people lose interest [in an older versio], it'll stop being maintained at some point," he continued.

"But the good thing is that it's open source, so if a company today has built their application on PHP 4 and wants to keep running that for the next ten years they will have the source code where they can maintain it themselves."

Complete article: PHP: Three Versions, One Promise

[the_ancient]

I disagree, 3 versions in 1 year is a bit much. Esp. for us developers that like to keep programs upto date.

You send more time making sure your old code is still compatitable you dont have time to write new code

[Gwaihir]

Esp. for us developers that like to keep programs upto date.

So your argument is that it hurts your pride?

I mean: there's nothing stopping you from continuing to run older apps on older versions for a good while to come. Security fixes and all are still made.

Besides: when you do seriously fix up your old PHP4 apps you can easily make them PHP6 proof right away. Anything new you design in PHP5 will only break in 6 if you use rather deprecated options.

[Ron]

I don't think that it's a pride thing; I think it's more about trying to stay on the latest version, and having to test, re-code and re-test production apps in a parallel environment for a month or two to ensure no new bugs have been introduced by either the coding changes or the environment changes.

Not that *I* have such worries anymore....

Some companies have more than a bulletin board and a shopping cart.

[Gwaihir]

I don't think that it's a pride thing; I think it's more about trying to stay on the latest version

Not pride? Then what? WHY try to stay on the latest version?

Sorry, but I'm not familiar with any serious companies tossing proven technology out the window just because there happens to be a new version out.

[Spathiphyllum]

Sorry, but I'm not familiar with any serious companies tossing proven technology out the window just because there happens to be a new version out.

As a general observation of large companies, it is not unusual for middle-management to insist upon change of proven methodologies to justify their position and/or shift it to appear plugged in to the latest trends. In fact maybe assigning such an indictment to just the middle management is too narrow. I suspect many a successful technical implementation has been usurped and replaced by the latest and greatest incarnation without the appropriate review to justify the change.

My point is that proven systems are often dumped because anxious managers want to present the facade of expertise and forward-thinking for personal remuneration and not because a shift is appropriate or even desirable from a corporate profitability point-of-view. The problem is that underlings need carry out shortsighted or under reviewed policies, which is always expensive and time-consuming, and suffer fallout when the change fails. I further suspect that those managers with the keen insight get promoted only to leave the cleanup behind. In the case of PHP, companies are free to pick their lineage and sustain it... so open source is a bonus. Conversely, there is no guarantee that developers will stick with the most appropriate and supported line where the brightest creators' minds are debugging the latest version. That is the demerit.

With multiple lines progressing in parallel, it is inevitable that confusion will ensue. Granted, companies are free to handle whichever path they want; however, it would seem to me that focusing the energies on one line for longer periods of time would seem the most appropriate for creators and developers alike. I just wonder if the PHP community is getting ahead of itself with the rapid "evolution" of its code. Are there competing PHP communities trying to gain some upper hand in the codes development path? I'm just a bit confused by the rapid change to justify so many versions co-evolving.

I must concede plenty of ignorance re PHP so this ramble need be taken with a shaker of salt.

[Gwaihir]

Are there competing PHP communities trying to gain some upper hand in the codes development path? I'm just a bit confused by the rapid change to justify so many versions co-evolving.

I must concede plenty of ignorance re PHP so this ramble need be taken with a shaker of salt.

I wouldn't say they are competing. However, there are different groups working on different aspects and that sure has something to do with this. PHP5 has been long in the making. While that was being completed with a fairly decided upon set of features, some others already started working on other 'big' features that had missed the cut for PHP5. Main part: pervasive unicode support.

Now, with PHP5.1 development out of the way, the time has come to merge all that into the main branch. With that getting underway the question was will this be 5.2, or what? Seems at that point some realised the upgrade was a bit too big for a .1 release already and some others jumped onto the bandwagon with other things that didn't make the PHP5 cut. Result: what's being staged will be labelled PHP6.

[Ron]

This thread erally shows the difference between corporate IT folks and non-corporate IT folks.

You try to keep up with the latest version where possible because third party apps frequently require you to stay current, or lose their support.

Versioning of all the myriad versions of whatever is out there and keeping your servers somewhat standardized is one of the single largest headaches there is. Next to dealing with upper management, of course.

Having an app with 3 major versions in a year just makes for the possibility of so many headaches... even in our own shared enviroments, can't you just imagine someone needing php5 for some application, but his php-Nuke or phpBB app won't run on php5, or some version issues with an app and MySQL version.... etc., etc.

Some installations have many apps with many requirements and can easily conflict with another app's requirements.

You try to keep up to date without the impetus of another app because USUALLY in the real world of commercial OS, releases are backwards compaitible... some apps may require minimum versions to function, but will work fine with later versions. Usually.

YMMV

[the_ancient]

So your argument is that it hurts your pride?

.

Pride? No

I develope Commerical Apps for resale across a wide range of server set ups

With 3 versions out I either have to limit my custom base by focusing on a single release or code so my apps are backward compatiable across 3 verions with in a single year.


Creates a bigger headache than is needed. Development and advancment is great, and I am I looking forward to some of the features in php6, however most users are still used php4, so if I build a app today in php5, I need to make it compatiable with 4 and then I need to turn around in a few weeks/months and retest it on php6

[Gwaihir]

Auch.. ok I see your predicament.

Personally I mainly code for single use, meaning I can code for the specific setup available (and often have a large say in that too).

Still, I think / hope that you won't find it that hard to take 5 and 6 in one stride. Of course, you can't use the 6 advantages yet then, but it sounds quite doable to (re)code apps in 5 that won't break when 6 comes around.

[the_ancient]

Auch.. ok I see your predicament.

Personally I mainly code for single use, meaning I can code for the specific setup available (and often have a large say in that too).

Still, I think / hope that you won't find it that hard to take 5 and 6 in one stride. Of course, you can't use the 6 advantages yet then, but it sounds quite doable to (re)code apps in 5 that won't break when 6 comes around.

We will see when 6 comes out

Right now I am still focused on 5 with 4 compatiblity....

That has not been too bad, we will see what actually changes with 6 and go from there.

But 4 was out for how long??? and now they jump from 4 to 6 over night

[Gwaihir]

But 4 was out for how long??? and now they jump from 4 to 6 over night

As I said before: I think those are related matters. It seems that the long time it took to get 5 out led to others already doing a lot of development of 6 in parallel.

For me personally it "is just getting interesting", so I am now coding for 5 with 6 in mind. O, the luxury of not having to support 4.

[Vin DSL]

Smart man!

Rasmus Lerdorf, the creator of PHP
PHP became popular because it eliminated most of the tediousness of writing CGI scripts...

The problem it solves is ugly. Ugly problems often require ugly solutions. Solving an ugly problem in a pure manner is bloody hard. PHP's aim is to make solving the web problem easy.