Once in a while I get questions about license (why not BSD?) or future (why not change it and break backward compatibility? why not move to python3).
Well. There are many frameworks that do that. I wrote web2py because I do not like it. Web2py was not designed to please web2py contributors (sorry). Web2py was designed to please the users and make sure their code does not break as we evolve web2py. This allows us to guarantee them backward compatibility and allows them to be sure they can invest in web2py long term. **This is its strongest selling point. This is what we should emphasize and make clear to our friends.** This does not mean we will not keep improving web2py. We have been doing it constantly. This does not mean that there cannot be forks that take a different direction. Although those forks cannot be called web2py in order to avoid confusion to users. This does not mean there is life after web2py. There will be something and we'll learn from our mistakes but whatever follow web2py has to be drastically different for me to have an interest in it. This does not mean that web2py is perfect as it is. What is perfect anyway? It is subjective. (This reminds me of that housewife that keeps rearranging the kitchen. Is there a better arrangement than the one the other family members have already learned and allows them to find utensils quickly without thinking?) This does mean that we, developers and contributors, may sometime have have to put extra work to fit new features into the existing design. I say may because I do not think it was a problem so far. This does not mean that we cannot rearrange the code internally. I am open to experimentation. This does not mean that we cannot create new programming paradigms (for example new widgets that contain validators) as long as existing syntax continues to be supported. This does not mean that we cannot use other repositories for web2py code although I think we agreed some time ago to move to mercurial + google code. I am still leaning that way since launchpad is too slow. I apologize for the delays but the book took precedence. My main job here is to make sure that patches do not break existing code and conform to the web2py spirit. This is not negotiable. Some people have criticized the current exec in environment where some modules are already imported. They say it is not Pythonic. I do not care. That is one of the main distinct features of web2py: do_no_repeat_yourself in web2py takes precedence over explicit_is_better than implicit. This is what makes Rails still more popular than Django. web2py moves one step further than Rails by giving more things an implicit default, including views. I do not think this is a weakness. This is another selling point. Moreover the current design has been developed to avoid conflicts between apps that need different versions of the same libraries or libraries that have name conflicts. Without it the imports would become depend on the order of items in pythonpath and the order of execution of apps. This is a big can of worms that we do not have to deal with. Currently you can import an app in web2py written by somebody else with its own modules and you do not need to worry about conflicts. Some of the critiques have come from members of the python community who have a vision about what a web framework should be: it should be more like Pylons or more like Django or more like Cherrypy or more like one of the other web frameworks. I respect their vision and I am not asking them to use web2py nor to contribute to it. I take a more pragmatic approach. My ideal web framework is one that makes life easier to those who want to learn it for the first time and, secondly, to those who use it at a professional level. If this means breaking some programming paradigms that are considered standard in the python community then be it, as long as we conform to good security and software engineering standards. I do not agree with everything the python community does. In particular I do not think that Python 3 was a very good idea. I do not think the changes were significant enough to justify breaking every existing python program out there. What makes me happy is that people who have actually tried web2py seem to like it. Massimo --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "web2py-users" group. To post to this group, send email to web2py@googlegroups.com To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/web2py?hl=en -~----------~----~----~----~------~----~------~--~---
