Re: HTTP Class

On 5 Jan 2001, Manuel Lemos wrote: > In PEAR repository it is assumed that there is only one way to do > something right. In PEAR repository there seems to be only space for one > class of each type. This leaves absolutely no space for others to come > with different approaches to do the same. > I think that's hardly the truth here. Based on the current state of things, there is no consensus on which kind of way to go with PEAR. Sometimes it looks like Stig is finally going to finish the draft of the website and other times we go back to talk about Coding Standards. I'm not saying that is a problem, since PEAR is supposed to be contributed by the developers, but my point is that the PEAR repository right now is hardly following any rule written in stone. If you want to contribute something to PEAR, or suggest that it changes somehow, just send an email here proposing it. > For instance, it would be impossible to have in PEAR repository 8 page > template classes or 6 table wrapper classes like there are at the PHP > Classes. > Like I said, on the current state of things, nothing is really decided on how PEAR will accept or deny contributions. (if I'm wrong, please correct me) > >uploading PHP classes. It REQUIREs login authentication to view > >and download classes which negates the possibility of automating > >updates or any kind of standard mirroring scheme. > > That's not accurate, stay tuned for the upcoming chinese and korean > mirrors. > This almost sounded like a joke. Do we need to wait for a mirror of some sort to have anonymous access to the repository ? > Anyway, since you bother to question the way the PHP Classes site is > accessed, I'll explain you the importance of the way it is. Every > downloaded file is accounted accurately. There is a top downloading charts > that is absolutely accurate since nobody can spam the site to increase the > number of downloads artificially to raise any positions in the charts. The > charts show the number of real users that downloaded each class. > > This is important because each author can at least be morally compensated > for the effort of sharing their code for free. I lost the count of how > many authors have thanked me for having the site working that way. > Motivated authors contribute more than once for the joy of the PHP > community that access the site. The upload average is of 2 classes per > author. > This is not true. You could just create a system where you would log all types of downloads, being of registered users or not. It would be even more interesting, since you could have the two different statistics, one for "anonymous" access and the other for the registered one. Why is it so important to make the user register on the site if you can still get the same statistics on this other way ? I'm sure you already thought of that and have a good reason tho. > IMHO this is plain ridiculous. It sounds like the PEAR repository > maintainers would be doing you a favour to let you contribute with your > hard work that your are giving away for free!?!?! > I believe you have the wrong perspective here: the maintainers are already giving this "developer" something for free, also created using their "hard work". Why shouldn't they have some sort of control of what goes inside their fruit of the "hard work" ? > I already said it and will say it again, the way I see it this is the > communist applied to the so called "Open" software development. Notice > that basically if you want to contribute you have to give your work away > for free, which is a fair rule), but only if the central committee (which > thinks is above all and everybody) approves it, which is an unfair rule and > that was precisely why communist regimes have fallen apart. > > So, please be serious when you bother to criticize the work of others that > effectively contribute for the joy of being part of the PHP community. > Like another person already noted, this committee has control over their own work, and the future development of their project. However, since this is an Open Source Project you can always just fork the code and run your own little committee. > This sounds like an invitation to leave. I am getting used to get this > kind of "stabs" once in a while from certain members of the PHP core > developers team just because they disagree from my point of view. You may > not realize just yet who you are hitting, but some day you will. Never > mind, I don't want to go where I am not wanted! :-( > I believe you know better than that. I find hardly to imagine that you were not expecting this kind of reply to the message you sent talking about your site. Don't get me wrong, your site is very important to the "PHP Community", but you never saw someone from PEAR or whatever going into the mailing lists of PHPLIB to talk about how the other implementations of their Database Abstraction Layer are better written and whatnot. The same thing would eventually happen. Cheers, Joao -- João Prado Maia <jpm@phpbrasil.com> http://phpbrasil.com - php com um jeitinho brasileiro