> Your point is understood. Encapsulation is a method though, > not a rule.
Then you misunderstood. :) Encapsulation is one of the core tenants of OOAD. I'm not saying CFC's are completely OO; but certainly we should treat them as though they are. Especially now; as they evolve. If we come up with methodologies and follow them > blindly without knowing when the methodology helps us or > cripples us... then we serve the methodology rather than the > methodology serving us. Err.. sure. Lets drop the term "methodology"-- too loaded. Lets say, "best practices". That does not mean my concept is > right... but could you chat with me about why you think this > could be a detrimental approach? I guess it depends on what and how you are approaching something conceptually. If you're using a cfindclude to break out your method logic then its not soo bad.. But feels very dangerous to me. > My concept is to create a CFC that can call the core circuits > settings, and actions. Sounds familiar. In CFMX the CFC is better at doing > this than an UDF, or a Custom Tag. I don't have a problem > with how Fusebox has done things... but for building sites it > seems there is an easier way of doing things. The goal of the > "methodology" used with this type of component is to have > nTier focused applications. MVC is nice... and basically this > is a parallel concept to MVC. I have no idea what you're talking about. > 1. Why would you have to do everything in a CFC to maintain > encapsulation? I suppose while technically you don't; I bet you won't. 2. Would calling an include file be worse than > calling a database or XML file because it contains logic? Yes, completely different things. > (nice nTier question) 3. If external logic is the issue... > wouldn't dependence on an external CFC also break > encapsulation? No. Objects communicate to themselves and other objects via messages. Either by directly invoking object methods, or via event callbacks. Another dependancy can be inheritance. Another tenant of OO. 4. For speed of execution, ease of > implementation and debuggin, training, and such issues... can > you think of an easier way to do things? I don't really have time to. :) Esp. since so many like to do these things for me. ---------------------------------------------------------- You are subscribed to cfcdev. To unsubscribe, send an email to [EMAIL PROTECTED] with the word 'unsubscribe cfcdev' in the message of the email. CFCDev is run by CFCZone (www.cfczone.org) and supported by Mindtool, Corporation (www.mindtool.com).
