Re: Proposing a new extension : Autoload
| From: | Derick Rethans | Date: | Wed, 12 Dec 2007 18:13:34 +0000 |
| Subject: | Re: Proposing a new extension : Autoload | ||
| References: | 1 2 3 4 5 6 7 8 9 10 | Groups: | php.pecl.dev |
| Request: | Send a blank email to pecl-dev+get-4986@lists.php.net to get a copy of this message | ||
On Wed, 12 Dec 2007, Pedram Nimreezi wrote:
> On Dec 12, 2007 12:31 PM, Antony Dovgal <tony@daylessday.org> wrote:
> > On 12.12.2007 20:26, Pedram Nimreezi wrote:
> > >> Those 4 things you just mentioned serve a distinct and useful purpose.
> > >> They are are not redundant with respect to one another.
> > >>
> > > Well.. they are in fact, since I'm only (and usually) talking about
> > > performance..
> > >
> > > include/require = SLOW, include_once,require_once = VERY SLOW
> >
> > Please share you knowledge and help us to make it faster.
> >
> You know I'd like to see what Derick thinks about this... Has anyone
> ever looked at his Vulcan Logic SRM engine? the super Alpha which he never
> has time to work on which can offer a real long-term fix for the speed
> slowdown?
I think SRM is a bad idea, it is against everything that PHP is not
meant for. THe strength of PHP is it's huge potentional for scalability.
A solution like SRM just adds a single point of failure and a bottle
neck. In situations where speed actually matters this is possibly the
worst architecture you can pick.
However, this has all totally nothing to do with includes being slow. As
a matter of fact, I find that statement as-is hugely uneducated as there
is no benchmark to prove anything like this.
> So why don't we truly share the knowledge... and help to find a
> solution for the root problem...
The root of the problem is people blabbering about stuff they have
little clue about.
regards,
Derick
--
Derick Rethans
http://derickrethans.nl | http://ezcomponents.org | http://xdebug.org