+1...for that AJ But if you are going to use someone else's code. make sure. Make for damn sure... it has a wealth of CURRENT documentation and a lively community.
My mantra... (a.k.a. rant) I don't use software that tries to solve all problems. The worst software grows out of goal-less development. Look for something with clear, concise, decisive design decisions. The *nix moto... do one thing and do it right. LESS CODE IS BETTER! A good example that will hit close to home for some and I hope won't start a flame war is Drupal. Until recently it was a bloated mess. It has started to clean up it's act come version 7 but in my opinion it still does WAY too much for it's own good. I don't use software if the documentation is not up to snuff. Don't take api/(insert language here)doc as enough. When you are getting up to speed in a new technology you need much more handholding or face hair loss. You need cookbooks, working examples, CURRENT api and method signatures. etc. I dont use software if it hasn't already been put into production some very visible places. Use something you know works. Why be the first on the block to see if X scales? You know how I know what scales. I have seen larger stacks use it in the past. Safe bets are best. I dont use software if it hasn't had at least 2 minor releases. Unless you need a specific feature of a bleeding edge release just use what you know will work! How do you know... because it's been around for a while and people have used it and tested it. I dont use software if it has no community involvement. (Don't underestimate other peoples schedules... think about how busy you get during your average work year. You may just have a project dropped from under you. One core developer on an open source project is not enough to bet the house on.) I prefer my software to come with unit/functional tests. If nothing else they can be used as examples. But the piece of mind you get as a developer if you know what you are using has been tested thoroughly is worth more than the most amazing untested functionality you can dream of. Just think of all the times you banged your head on your desk trying to figure out why something isn't doing what the documentation said it was supposed to do. Only to find out yes... it is a bug in the library you are using! I hate using software. Less is better period. If you can get away without using an extra daemon then do it. Don't add 5 more caching layers and 3 more database abstractions just because they are the new cool acronyms. It's only going to make your code less maintainable. Over architecting and early optimization are what keep projects from ever getting to launch. The same thing goes for languages. Don't write your services in more than 1 language if you can avoid it. If you are writing webapps from the ground up limit yourself to one backend and one frontend language. Don't add caching until you need it! The same thing goes for sharding of your database schemas! KISS... across your entire stack. Not just your code. Whew... sorry about that. Back to the topic at hand. Symfony 1.X is probably still more widely used than Symfony 2.x and both are good. Although 2.x is a far more abstracted and multipurpose framework 1.x is still great for the usual 3 tier webapps. Both are very well documented. CodeIgniter is a really great lightweight framework. If you are looking for something straight forward, well documented and full of examples this is also a good choice. Zend had so much steam early on and then where did it go? I feel like they have been sleeping. Yii has been around for a while and same with Kohana but both have less numbers than the above. I have to revisit cake. I haven't used it in a while but for a while it was seriously suffering on the documentation front. If I was to pick a framework to run with on a new project I would probably pick Symfony 2 because of it's forward thinking architecture. It can seem overly complex so be warned. If you haven't already heard of things like dependency injection or service containers you may feel a little lost. The good thing is that there is a bunch of documentation to get you up to speed. Also it makes use of useful things like ESI and... HTTP... who knew! Cheers, Jesse On Wed, Sep 14, 2011 at 3:40 PM, Ajai Khattri <a...@bitblit.net> wrote: > On Wed, 14 Sep 2011, Kristina Anderson wrote: > >> The framework choice is always less important than having a >> well-designed database underlying your code. > > One could argue, that there's no replacement for sound engineering and > that the greatest framework is your own brain. > > Its important not to get caught up in any hype surrounding one tool > or another. There's that (now classic) post from the CD Baby guy who was > all set to jump into Rails but ultimately rewrote his own PHP framework > instead. > > As he says in the post: "But the main reason that any programmer learning > any new language thinks the new language is SO much better than the old > one is because he's a better programmer now!" > > http://www.oreillynet.com/ruby/blog/2007/09/7_reasons_i_switched_back_to_p_1.html > > > -- > Aj. > > > _______________________________________________ > New York PHP Users Group Community Talk Mailing List > http://lists.nyphp.org/mailman/listinfo/talk > > http://www.nyphp.org/Show-Participation > _______________________________________________ New York PHP Users Group Community Talk Mailing List http://lists.nyphp.org/mailman/listinfo/talk http://www.nyphp.org/Show-Participation
