Closed stayhero closed 12 years ago
Like your proposal.
I like RoR although I have never used it before. Looking at that, that will help in RAD. I do not mind learning it though.
We should also look into number of people who are familiar with RoR. If fewer people knows about RoR or Ruby then we should stick with PHP.
As for Zend, I have used Zend 1.10.* version before. I will have to check out ZF2 that is quite interesting with lots of new features.
I like Bootstrap from Twitter, but it will interrupt the existing css classes we will use in future. It does not have css naming convention like jQueryUI does.
You won't regret learning it, promised. I switched a year ago, and now I hate working on PHP based projects. ;-) I like using Rails because almost all commonly needed stuff is baked in. I'm not a huge fan of ZF anymore. Symfony2 looks better IMO but it's still PHP and IMO the documentation is not as good as Rails'.
Regarding Bootstrap: I don't think that we get problems with css classes because when we decide for that CSS framework we will build our CSS around that. I doubt that we're going to use more than one css framework. jQuery UI is different because it's common that you integrate only parts of that into existing html/css, like a calendar widget. They must namespace it because there would be a high risk for naming collisions as their stuff is added to existing code/css.
Otherwise if you create a website from scratch using your custom css you would need to namespace all your stuff as well. And that's not very common (because you're the only one using your css). It'd be different if we would want to create html/css code that we would offer to third-parties as a widget?
On the backend, I'm most familiar with ASP.Net MVC, and have also dabbled with Rails and Django.
It's probably most important to choose whichever language/framework we're all most proficient in - and I'm happy if that turns out to be Ruby on Rails.
Not much of a frontend guy, but like the suggestions Chris made.
I'm all for Rails, and whatever framework you guys think would work. I've used 960gs and blueprint, but looking at Twitter Bootstrap, it's not that much different.
What else do we need to get started? I'm working on setting up a server on my home pc for testing.
I guess we decided for Rails. What about commonlisp?
What about further technology stack? I guess jQuery is a no-brainer? For templates I would choose HAML and SCSS. Anybody who doesn't want to use that, if yes, why? Learning takes maximum a day. HAML helps keeping an overview about your html, and makes refactoring html quick and easy?
We also go with mysql for the MVP?
What about the design? Do we use Twitter Bootstrap? Anything other? Zurb? 960gs would work but we'd need a design on top of that (e.g. for forms etc). Or should we purchase a template on Themeforest.net?
HAML/SCSS, Twitter Bootstrap.
I can setup a testing server on my home pc or Chris if you are willing to host it on your server that'll be better.
Other than reading the tutorials, I haven't used either HAML/SASS before (usually cautious about magic abstractions), but willing to give them a spin on this project.
jQuery for sure - along with jQueryUI and Modernizr where appropriate.
What about using Heroku for hosting? -----Original Message----- From: Christian Sent: Sunday, November 06, 2011 7:36 PM To: SaqibS Subject: Re: [youPM] Decide on a technology stack (Rails, PHP, etc) (#1)
I guess we decided for Rails. What about commonlisp?
What about further technology stack? I guess jQuery is a no-brainer? For templates I would choose HAML and SCSS. Anybody who doesn't want to use that, if yes, why? Learning takes maximum a day. HAML helps keeping an overview about your html, and makes refactoring html quick and easy?
We also go with mysql for the MVP?
What about the design? Do we use Twitter Bootstrap? Anything other? Zurb? 960gs would work but we'd need a design on top of that (e.g. for forms etc). Or should we purchase a template on Themeforest.net?
Reply to this email directly or view it on GitHub: https://github.com/halixand/youPM/issues/1#issuecomment-2647138
I'm cautions about too much magic too. That's BTW why I don't suggest Coffeescript instead of Javascript. But HAML isn't really much magic. But it helps really if you write the HTML by hand. It's much clearer than HTML. You'll never search for a missing closing tag or similar. And you save a lot of typing (closing tags, erb-tags <%= %>).
SCSS is almost CSS, just with templates ( never need to copy&paste colors, font-definitions or cross-browser markup like rounded-corners, moz-rounded-corners etc..
Regarding hosting: Alex is signing up for AWS, he wants to use the free instance as a start. Heroku: We could use that, but I think 5MB database size might be too small even for development, and if we want to use stuff for background processing (delayed_job or Resque in AWS) we'd need to pay for workers.
And I think for production use heroku is too expensive. $200 for a database with 1.7GB RAM? Wow... :-O
Ok, sounds good.
EC2 is only free for 31 days - but that should be good enough for testing.
-----Original Message----- From: Christian Sent: Sunday, November 06, 2011 9:14 PM To: SaqibS Subject: Re: [youPM] Decide on a technology stack (Rails, PHP, etc) (#1)
I'm cautions about too much magic too. That's BTW why I don't suggest Coffeescript instead of Javascript. But HAML isn't really much magic. But it helps really if you write the HTML by hand. It's much clearer than HTML. You'll never search for a missing closing tag or similar. And you save a lot of typing (closing tags, erb-tags <%= %>).
SCSS is almost CSS, just with templates ( never need to copy&paste colors, font-definitions or cross-browser markup like rounded-corners, moz-rounded-corners etc..
Regarding hosting: Alex is signing up for AWS, he wants to use the free instance as a start. Heroku: We could use that, but I think 5MB database size might be too small even for development, and if we want to use stuff for background processing (delayed_job or Resque in AWS) we'd need to pay for workers.
And I think for production use heroku is too expensive. $200 for a database with 1.7GB RAM? Wow... :-O
Reply to this email directly or view it on GitHub: https://github.com/halixand/youPM/issues/1#issuecomment-2647756
OK, we have the amazon service free for a year :) just don't go over the limits. Chris will set it up, does everyone need an account?
I finished the setup of the AWS micro server. It was a pain compiling Ruby on that kind of slow machine. ;-) Send me an email with your public ssh key so that I can add it to the servers authorized_keys list.
I guess we decided for Rails 3.1.1. finally and we can close this issue. If there is anything you need clarification for (hello ryan ;-)) I think it's best to simply create a new issue.
We need to decide on a technology stack. Everyone can propose one.
I would propose:
Other options for the server side application would be IMO:
or
I would choose Rails because I'm familiar with that and from my experience offers the best way to build rapidly a MVP. Although I'm having experience with PHP and Zend Framework 0.6, I don't have real world experience with Symfony2 or ZF2.
Furthermore, for the client side, I would suggest using a CSS/Framework like Twitter Bootstrap or Zurb.
Do any of you have experience with that frameworks? Please feel free to make other suggestions.