Re: Re: [metabase-dev] MDB news

Hello, On 09/14/2002 04:57 PM, Gabor Hojtsy wrote:

| Professional developers abide to rules of the company they work for. | However PEAR is not a company and developers do not have change their | style to be able to contribute for free. Demanding people to bend and | become less productive to be allowed to contribute is a very ungrateful | posture. | | You make it sound as if developers have the duty to conform to rules | that they are not paid to work by, as if they are obligated to | contribute. Since they are not obligated to contribute at all, obviously | you will see a lot of qualified developers refusing to contribute. Do | you think that is a good thing for the sake of one style only? I have not participated in this discussion before because the subject says MDB news and I am not really interested in being on one side or the other.

The thread was about how much the rules of PEAR made Lukas had much more work and take a long time to make MDB comply with the PEARifying instead of focus on much more important things like migrating all existing drivers.

BUT as for coding standards, all those can be achived by a code formatter, which are available for free on the net. So no need to stick to one coding style while writing scripts. Consistency is the most important thing while you build up a complete classlib. The biggest difference between PHPclasses and PEAR is that PHPClasses is a listing of individual projects, while PEAR aims to be a collection of core classes (much like the Java core classes). This also determines the rules contributors need to follow in each project.

I am afraid that goal is unrealistic. Java classes were developed by professional developers paid by Sun. They were planned ahead to integrate and complement each other conviniently by professional software architects. Their development was managed by dedicated software engineers. The number of Java classes provide by Sun is huge. PEAR is a class library contributed by independent developers on a good will basis. Developers contribute with what they have and not because it was part of a professionally conceived master plan. PEAR has no leadership nor a development plan. Therefore, there is no guarantee of quality and suitability of PEAR, that encourages people to users adopt it for all their needs for entreprise application development. As much as you would like that PEAR was professionally developed as Java classes by Sun, it is unrealistic that it will ever get there, especially because PEAR is a project without a plan that says we are here and will go there where and we are commited to satisfy the needs of serious applications as the classes we developed were planned ahead to fit together. Eventually that will happen in one aspect or another, but the way it now, it does not make much sense to compare it with Java classes. If you are interested in PHP in something planned and documented ahead with the commitment to provide a base framework of classes, you may try BinaryCloud although there are others: http://www.binarycloud.com/ BinaryCloud was planned ahead by Alex Black. Unlike some people of PEAR, he has the feet on the ground, he did not all from the scratch. He looked around for ready to use solutions of classes that solve base problems well and developed a framework around such base classes. He did not force any developer that had ready to use components to change the style of their code because what matters is that the classes satisfy the project needs well and without the need to reformat the code that already works reliably.

I have read two articles about PHP vs. .NET at a popular webmaster site. Both of the articles are quite misleading, but the comments include many points on PEAR...

I would not be bothered to enter any platform wars. The truth is that what you use today you most likely will scrap in the future for something better. No matter how much you care for PHP today, chances are that in maybe 5 years you will be using something else. So, you would better stick to your current needs instead of hoping that what you use today will be what you will use forever. -- Regards, Manuel Lemos