Hello there SPF PHP developers.
I’ve got a challenge in front of me and before I start in on it, I’m realizing I need a little [perhaps a lot] of advice to be sure this goes right.
I’m starting a new service that I of course anticipate to grow quite quickly and while our team gets funding in order to hire out, we’re also working to build out the initial version of our system. Problem is, none of us are current coders, so it’s time to relearn and quickly. That’s where I need some advice.
A little background… Long ago I was pretty proficient with Perl coding together small scripts and up to full ecommerce applications that handled millions of transaction. After a few years of that I picked up a decent amount of php developing smaller applications and a number of fairly popular, but simple, addons for vBulletin 1-3.
While I’ve kept some of those tools going here and there, it’s been many years since I’ve called myself a developer and ideas like MVC are a bit unnerving – I’m a marketer, not a coder, but must become one again. My business partners have similar backgrounds – understanding, coding experience but no domain expertise in the modern world of php.
So, knowing all that you know about php development I’m hoping to get some ideas on where to start to insure success.
[INDENT]Would you pick up a framework like Zend or cake to force ridigity? Gain access to more secure approaches? Or perhaps something lighter to just gain common functions?
Would you point at a particular templating engine or work within php to use the native support?
How formal would you be in adopting MVC type approaches at the risk of having to learn more advanced coding techniques?
Do you have any suggestions on what to avoid, be it particular systems, handicaps, or methodologies?
Knowing that we will be implementing features from Facebook and other social services with open platforms / APIs, does this change the libraries or methods we should be thinking about?
[/INDENT]
More than any of those specific questions, I’m broadly looking to get input from those who know the field. This is not a build that must last forever but it does need to be good enough to justify the effort and support a launch for 6-9 months so I’d like to be sure we approach it and do things decently at worst.
Thank you.