PS. I think the TG mission is clear, but the goal is not. I think getting a stable framework that can be distributed with ubuntu amd do it within a year would be a worthy goal. Sent from my Verizon Wireless BlackBerry
-----Original Message----- From: [email protected] Sender: [email protected] Date: Sun, 31 Oct 2010 11:00:48 To: <[email protected]> Reply-To: [email protected] Subject: Re: [TurboGears] Re: A modest proposal for the future of TurboGears I echoI the same except for one thing. Either go with pyramid with full force not maintainig the past or not go at all. You guys are great but maybe spread to thin. Combining forces would seems like a more logical approach. But on the other hand until a decision is made there is not a need to put tg2 on hold. I can restart my project when a decision is made. I have done it many times in the last couple of years. I fear trying to merge may have the opposite affect and cause a split. Sent from my Verizon Wireless BlackBerry -----Original Message----- From: Shane <[email protected]> Sender: [email protected] Date: Sun, 31 Oct 2010 03:21:07 To: TurboGears<[email protected]> Reply-To: [email protected] Subject: [TurboGears] Re: A modest proposal for the future of TurboGears Hello All, Saw this post and felt that I should put in a word. I started learning and working on TG2 almost a year ago. Coming from an embedded development background, my last experience with Web development were some C cgi's written for a class assignment about 15 years ago. I had chosen TG2 because I liked Python and figured that TG, with the "best-of-breed" choices made for me had fewer possible ways to go wrong than Pylons. I tried Rails, but went with TG because I liked Python and SQLAlchemy. The idea was to start development of what was then just a quick prototype of a business Web application. Like many quick prototypes, this one (somewhat unintentionally) expanded into into an alpha then a beta project, which is now actually generating some revenue. Between the alpha and beta, I completely ripped up everything I had done and essentially started over, as the alpha had taught me much about how NOT to write a TG app. I wrote my first widget, a wrapper for the jqGrid library which, though it does work features some incredibly ugly code that just barely gets the job done. As someone who very much started from scratch (twice), I think I can second a few things that have already been mentioned: 1) Documentation: Has come up, but let me emphasize the point a bit more. I have actually gotten excellent help from the groups and the quality of the quick start docs I think is rather good. The problem for me is that the docs and examples don't go very far beyond the basic of application. For example, the fairly simple matter of saving form data to the SA object (uniqueness checking, URL mangling) gets fairly complicated if you have a large number of fields and can't/ don't want to use something like Sprox (because I have so many jqGrids for data entry). Making the leap from the quick start case to the real-world case isn't well documented and many solutions found on the groups provide multiple ways of doing the same thing, which brings us to #2... 2) The issue of "One way" vs flexibility raised earlier in Christoph's post I think could have a fairly simple solution that can be addressed with documentation rather than code. Retain and add whatever flexibility the expert developers need, but provide very in-depth examples and cookbook best-approach methods that push the new developer along a well trodden (and tested) path. When they get better, they can then experiment more with alternative and more complex solutions. For example, there must be about ten different ways presented in the docs and newsgroups to fill a SingleSelectionField. For something so common, there really should be a single, obvious method. 3) I think a lot of you guys - and this is said with the utmost respect - may be too friggin smart for your own good. Stop and think for a second about the range of knowledge that you keep wrapped up in your head. When I do an embedded project, I have a 200 page data sheet for the chip, a copy of "The C Programming Language", some liquor, and I'm ready to go. Since starting this project, I have bought a pile of books at least three feet tall sitting next to my desk - JavaScript for widgets, SQL, SQLAlchemy, Apache, Python, the TG book, the Pylons guide, HTML/CSS layout plus online resources for cherrypy, Genshi, etc and most of which have virtually no relation to each other in function. The learning curve for moving into Web apps is steeper than anything I think I have ever tried. Most of the people I know that are looking for a Web development platform aren't Web developers at all but capable people from outside fields (medicine and microbiology seem popular right now) that have an idea and want to develop a prototype before hiring a specialist. In my opinion, TG2 is quite good technically, it "just" needs more documentation which unfortunately can't be done in a weekend sprint. Like Uwe, I have some pretty severe time constraints on assisting the community as I try to get this project off the ground, but I think it is certainly in my interest to try. I can help to write some CookBooks, but my hesitation has always been a concern for food poisoning -- that I may be giving a bad recipe. - Shane -- You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/turbogears?hl=en. -- You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/turbogears?hl=en. -- You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/turbogears?hl=en.
