Re: Where is the value ? (Re: SessionManager)

From: Date: Sun, 01 Jul 2001 20:37:22 +0000
Subject: Re: Where is the value ? (Re: SessionManager)
References: 1 2 3  Groups: php.pear.dev 
Request: Send a blank email to pear-dev+get-521@lists.php.net to get a copy of this message
Luc Saint-Elie wrote: > > At 17:32 27/06/01 +0200, Tomas V.V.Cox wrote: > > >I really don't agree here. The XLM_Parser class saves you to type a lot > >of code! Just extend the class, define a StartHandler, EndHandler and > >cdataHandler, and you get'll a parser for your xml format. Isn't this > >enough add-on value? > > Hello, > > A ?basic end user? point of view here?. > > XML_parser has probably a lot of value for a high level developer (who > probably doesn?t really need such a class by the way) but no real value for > the average end user , I mean no ?real programmer?.. and they are (we are > ;-) ) a lot in the PHP world. > > The reason why is the complete lack of documentation. > There is a 2 nd century Chinese proverb that says ?a high end Ferrari is > useless if you don?t know how to drive a car? Absolutly true. PEAR is enough old to become a real big and nice proyect. If we want to reach this point I think we must change a little the mind and put two simple new requirement for new accepted software: - Documentation. I don't mean a 50 pages pdf full ilustrated. For the first time having a explanation of the class (purpouse and features) and some usage examples at the top of the source code could be enough. - Mantain. Developers who submits code to Pear at least should attend bugs reports, if noone fix or discuss it in for example two or three days. Open Source proyects used by a great amount of people are quite different to hacker code exercices. It implies some other (I know, sometimes boring) work, that people should have in mind. Users are very needed and welcome in Pear as they really give an added value to Pear when they meet with a bug or submits a patch. > > The most the technology is new and/or not well known (and its the case with > XML) the most documentation is important.. perhaps as important (if not > more important) that the raw performance of the script. I'm not a XML Wizard like Stig (who wrotes the XML_Parser class), but I'll write some basic documentation for the class sometime this week (hope that he or others could review it). I feel a little sad when hear from people that doesn't use some great software because of the lack of docs. > On the complete other hand, DB.php is a delight to use? not because of the > overall performance of the package (I'm confident in your capacities.. but > because of the excellent work you did on the documentation front > (http://vulcanonet.com/soft/?pack=pear_tut ) that combined with the package > itself add A LOT of value. This work makes the package a must have for this > ?basic average end user? that I am, far , very far beyond the question of > database independence (that by the way, I'm not sure it is a real big stuff > in a web world, we don't change the database every now and then) > As I normaly work with other developers, for me is a must to have useful documentation. But perhaps sometimes Pear developers aren't pear users, so think that they don't get any advantage developing for Pear. Docbook <http://www.docbook.org/> is for far the standar and the best documentation publisher system (used by the PHP Manual), but I also know that perhaps it is too much for writing quick and small docs. I've develop my own ultra limited system (the one I used to build the doc for Pear DB), that reads standar Pear definition packages (package.xml), tracks last modification times and highlights php code. I'm glad to offer it to developers who wants to write fast docs. Regards, Tomas V.V.Cox

« previous php.pear.dev (#521) next »