What's best to use?

[Patty]

So I want to develop a web based tool that manages bookings, does email / sms notifications and takes care of MIS / Stock / Finances etc. I know how to do html, php and sql. Is it best to use these languages or a frame work such as codeigniter?

[Hamster]

A framework like codeigniter (which doesn't interfere too much) have a lot of the donkey work already done for you. Your development speed will be increased a lot using such a framework.

On the other hand, frameworks have got your basic CRUD interactions etc built in and may not be optimal for an extremely busy application. And it might feel constricting at times if you don't know it that well.

Personally, I like writing stuff from scratch (I feel that reinventing the wheel can make it rounder). Very bad for business but lots of fun for the developer

[Wovenhand]

If you use CodeIgniter you will be using those languages anyway.

Edit: You'll need to understand the MVC pattern if you use any of the frameworks.

Another edit:

Personally, I like writing stuff from scratch (I feel that reinventing the wheel can make it rounder). Very bad for business but lots of fun for the developer

Couldn't agree less.

[_kabal_]

Couldn't agree less.

+10000

if you are writing any code that isnt business specific, you are wasting your and your client's time and money

[foozball3000]

+10000

if you are writing any code that isnt business specific, you are wasting your and your client's time and money

Where's the fun in that? Programmers who have fun while coding (even if it's a little longer to code) give far better products. There's more to development than just boring code and business logic.

[_kabal_]

business is about making money, not having fun.
fun comes from writing amazing maintainable, scalable, extendable business applications, well thats just me. there is nothing less boring than turning businesses into software

[Hamster]

Couldn't agree less.

You honestly think writing business code is more fun than screwing around with framework code and stuff?

+10000

if you are writing any code that isnt business specific, you are wasting your and your client's time and money

FTFY

[Wovenhand]

You honestly think writing business code is more fun than screwing around with framework code and stuff?

What?

You said you enjoy reinventing the wheel. I don't.

[Hamster]

What?

You said you enjoy reinventing the wheel. I don't.

Different strokes I say. To me it is way more fun optimizing, redoing or writing the caching/logging/security framework stuff than designing the same ol' workflows, screens etc. over and over again. Besides, doing that allows you to learn better ways of doing and not doing things.

[Wovenhand]

Different strokes I say. To me it is way more fun optimizing, redoing or writing the caching/logging/security framework stuff than designing the same ol' workflows, screens etc. over and over again. Besides, doing that allows you to learn better ways of doing and not doing things.

Fair enough, I'm probably just jaded. I won't even touch a framework if it doesn't have built-in auth, for example.

[_kabal_]

if you find yourself having to rewrite caching/security/logging, etc frameworks, you chose the wrong framework in the first place.

maybe I am just spoilt as a java dev with the miriad of incredible boilerplate frameworks out there, that I can concentrate on the really interesting stuff, like integration into legacy systems

[Wovenhand]

if you find yourself having to rewrite caching/security/logging, etc frameworks, you chose the wrong framework in the first place.

maybe I am just spoilt as a java dev with the miriad of incredible boilerplate frameworks out there, that I can concentrate on the really interesting stuff, like integration into legacy systems

I think that's one of PHP's biggest problems: decision fatigue and the lack of a single focus like Rails or Django has.

[Hamster]

if you find yourself having to rewrite caching/security/logging, etc frameworks, you chose the wrong framework in the first place.

maybe I am just spoilt as a java dev with the miriad of incredible boilerplate frameworks out there, that I can concentrate on the really interesting stuff, like integration into legacy systems

Our project is too big and versatile for of the shelf frameworks. Can't afford that dependency so it has to be in-house. Not quite a framework but more base libraries that need fixing and some conformity after 5 years of deving.

[diabolus]

Well if you are into Rapid Development and prefer delivering business applications with alot of the "obvious" things already built in (updating/deployment/security/connections/etc) , also look into something like Microsoft Lightswitch , especially since it is essentially C# and SQL . You design your database and your application is literally generating itself.

Linky (not alot of devs seem to know about this, use it, don't use it ) : http://www.microsoft.com/visualstudio/en-us/lightswitch

Obviously if you got alot of spare time and are bored, you can create your own framework that does exactly the same thing, but in my case, i am of the type that does not want to end up in a situation where i have to maintain and support a heap of applications because i coded it from scratch (and a ton of bugs and features need to be added over time at your own expense). So using a Lightswitch type of framework, hey, 2 days and you got your app out, it is also much easier to support and maintain and hand over.

...so yea, depends on where you see yourself in 5 years i suppose (and whether you are working for a Software Company or a Non-IT company) . Everyone i know who did the "lets write our own framework" thing, are still pretty much coding on the same framework until they leave, alot of devs might see that as coding yourself into a rut...especially if you can't carve yourself a path out of your creations .

Trust me, all these little "tools" your create, whatever tech you use, you are going to own it. So look ahead, it is all roses thinking of creating your own platform, but 2 years later when you are still fixing bugs on the GUI, you might not feel the same about it


Then again, i'm an engineer , a computer science type tends to think differently about these things. However, the original question is clearly telling me this is a business application, and definitely not something that is exactly "new" or "revolutionary" in terms of business applications.

[Tim the Techxpert]

Well that was an interesting debate andd one that developers have discussed and debated for many years. Back when mainframes were computers and workstations were unheard of, developers used libraries of code because it was written to standards, worked and if it needed changing then the same changes were applied to all applications that ussed it. We used to have the same debates about write your own or use the standard one so has much changed.
In many ways it has now there are plenty of tools to develop your applications in and I would say use the tool that best suits the job.
As an application developer it is your job to get a working applicatin out as soon as possible.
If you want to spend your creative talents in writing code from scratch then why not create a development tool?
Anywho I digress so have you considered Dreamweaver for the development?
Whatever tool you choose I would not develop from scratch time is too short and time is money.

I am Tim and I look forward to hearing how the application turned out.