RE: [phplib] [RFC] Future of phplib From: Chris Johnson (chris <email protected>)
Date: 07/24/01

> -----Original Message-----
> From: Björn Schotte [mailto:bjoern <email protected>]

> * Chris Johnson wrote:
> > To me, that means moving PHPLIB to PEAR results in:
> > 1) I have to rewrite large portions of my several applications which
> > currently use PHPLIB.
>
> Moving PHPLIB to PEAR is for "future" applications, i.e.
> new applications that you write.

Point taken. I may well use something from PEAR for new applications,
written from scratch in the future.

Most of my work in my 22 years of developing software has been extending
existing software, or leveraging one piece of software to build another.
You've perhaps seen the old joke: There are really only 4 programs ever
written in the world. The rest were just cut and pasted from them.

That is to say, many applications have many bits and pieces and
functions and whole algorithms in common. If I've implemented it once
using PHPLIB, even developing a new application means I'll more likely
REUSE my old PHPLIB-based code, than to find NEW code from PEAR, even if
the PEAR code may be better.

> > 2) I end up with PEAR code that is slower and less efficient.
>
> Prove it.
>
> > 3) I have to wait another year or so before PEAR gets its
> act together
> > and actually releases this stuff in a usable state.
>
> Prove it.

I have no need or time to prove those statements. I am the decision
maker for my development. If I FEEL that way, without hard proof, then
I will DECIDE the way I feel. I may or may not decide that a certain
project should have deeper analysis or more facts. But in my
experience, one can NEVER completely prove one software choice over
another by a complete collection of facts, with no subjective opinions.
It just doesn't happen -- there are far too many variables and
judgements involved. Thus, I make my best, experienced judgement based
on what I know and what I infer. I may be wrong, but I can't analyze a
choice to use PEAR DB versus PHPLIB DB abstractions for a year before
deciding. I have to decide today, or at the latest, tomorrow.

> > and as CPAN currently is) is a great idea. But the current
> > implementation and politics of PEAR are detrimental, and
> the bottom line
> > is not attractive.
>
> Why don't you do something against that instead of
> mourning and weeping around?

Who's mourning? Who's weeping? When PEAR and its masters mature enough
and PEAR becomes valuable to me, I'll help out. There's only one of me,
and there's only 24 hours in a day. I don't have time to help every
poor down-trodden project in need of help. 

-- 
Abbestellen mit Mail an:   phplib-unsubscribe <email protected>
Kommandoliste mit Mail an: phplib-help <email protected>