Last week, Andi suggested that we (and by "we", I mean the PHP community at large) start preparing for 5.1. He made a call for anything people would like to squeeze in that's almost ready, and suggested that the biggest upcoming feature will be PDO. I agree, PDO is very exciting, and, while I haven't had an opportunity to try it yet, a consistent mechanism for database access in PHP has been long on the list of wanted functionality in PHP. More on PDO later (I hope).
Two other major discussions errupted:
First, Rasmus indicated that he would like to release his filtering
mechanism. Much argument took place. I pitched in my opinion (a couple
times) -- ("This is a good idea, but it should never be turned on by
default. This must be a developer-called mechanism, and not
a sysadmin-called mechanism. I don't have time to deal with another
magic_quotes_gpc-like fiasco."), as did many others, and discussion finally
subsided. Rasmus' filter, whatever it will be called, will be in PECL only,
and not the main PHP distribution. At least for now.
The other discussion was much more amusing (and annoying).
(((Let's set the scene: most of the time (at least on php.internals), I sit on the sidelines and watch. Once in a while, I have something useful to say, but it's usually more wise to sit back and merely lurk, because odds are that someone else with more karma will make the same, or similar comments that I would. Rarely will I speak up about a major issue, although I have done this in the past (and now we have Classkit, and hopefully Runkit, soon). Most people, I'd guess, simply lurk on Internals -- they (myself included) don't have the specific knowledge (Zend Engine internals, PHP Internals, API, etc.) to participate intelligently.)))
The other main 5.1 related discussion was about operator overloading. The topic was breeched when Sebastian Bergmann pointed to Johannes Schlüter's patch to allow this new functionality. The general concensus on the list was something along the lines of "Operator overloading is (syntactical sugar|nearly impossible to debug|voodoo|unnecessary/can be accomplished in many other ways). Someone named Terje Slettebø showed up, seemingly out of nowhere, and started pushing the necessity of operator overloading in PHP. Again, the core PHP guys (the gurus with above-mentioned karma) all but uniformly denounced the idea. Still, Terje pushed. That was his mistake.
You see, Terje had already lost. It didn't matter how good his next point was. It didn't matter how well he extolled the virtues of operator overloading, and it didn't matter who he quoted from which newsgroup on whatever language. Why not? Simple: he doesn't have the necessary karma to fight back.
The PHP community, as I see it, is a near-textbook, yet real-life, implementation of a meritocracy -- you gain "power" or karma, by contributing to the community. What Terje didn't understand is that he'd lost because of this. He hasn't proven himself. Maybe I just haven't seen the results of his labour, but Terje entered the battle only partially armed. Sure, he seems to know his stuff, concerning operator overloading in C++, but his contributions to PHP, if they exist, and to my knowledge, are minimal. He carries no weight, and when the powers that be tell you "no" you can't fight back if you don't have any karma.
Fortunately, entirely too long after the discussion had become useless, but not after a few humourous cheap shots, the list has been nearly free of operator overloading talk for a few days. Unfortunately (for himself, at least), I fear that Terje has dug himself into a hole that he will have a hard time crawling out of.
If I were him, I'd assume an alias, for the purposes of Internals, and lurk for the next 6 months, saying little to nothing. Hopefully nobody will notice that he's the same person, if/when he comes back.
(-:
