Woot.. no MVC??06 Oct 2006 | Comments
Our new framework has gone quite a major “upgrade” recently. The database abstraction layer is changed considerably to accomodate the usage of newere PHP’s MySQL interface called MySQLi.
MySQLi is supposedly better than the standard PHP’s MySQL interface. One notably advantage is the support of prepared statement. After using it for a week, I guess the benefit for our system is not that noticeable, mainly because our projects so far don’t use SQL transactions that intensively. I’m just thinking whether it’s worth the effort of upgrading it at all.. Especially when today, I was reminded by my boss, that our system is not MVC based.. and I was stu
ed!nnYes, it’s not MVC based system.. IMHO the effort should be put on creating the framework to be MVC based instead doing things like caching and database layer upgrade. Better for SEO and better for the developers. It will be more maintanable as well I guess. MVC is just so common, that most of Open Source packages would have this characteristics, Drupal, WordPress, Code Igniter they all are MVC-ed. But on the other hand, I also realised that converting to MVC is probably beyond our capability at the moment, it’s just going to be a massive rewriting of the framework especially after the recent changes, should’ve moved to MVC first before even tinkering with other stuffs. Oh well.
Sometimes I wonder how is a decision on adding stuffs to the framework can be justified. I seriously doubt caching is essential for the framework at this point of time, even the database upgrade. I mean just because Drupal does something, doesn’t mean that we have to follow suit right? The only thing that really worth investigating, I think, is the Drupal’s Node system. The thing is, these changes ca
ot be translated to selling points (commercially) directly. That is, we can’t sell the system by saying: our system is now using improved MySQL interface, it also does caching. The client will just go “huh, so how’s that beneficial for me?”. But if it’s MVC-ed, we can say our system is using a url rewriting engine, that will help massively with search engines.
I’m not saying that the changes are not good, not at all, the current framework is way better than previous one, but main concern is: with so much things you can improve/add, why choose one over the other? Still haven’t found better email engine yet, and this is potentially the strong selling point of the system. Still haven’t done search yet…
Huh, I guess I am too in to this software architecture thingy, now I question every decision made for the framework, because in the end, it doesn’t really matter for me :P I just happy code crunching using whatever. Why do I even bother thinking about this, when I have to do my assignment.. sigh.