Honest question here. I agree with the "virtues" of separation, but why do
people think this (let's use the Java world for an example) concept seems to
be mainly something of interest to "Enterprise" shops. Why don't little guys
do things this way? Is it just they don't see the light... or is it more
that it is more critical benefit when, as you said, "there are multiple
applications to work better together"?

I haven't read all the books and talked to all the religious gurus of
encapsulation, heh. Could someone tell me what the current take is on when
this is of benefit, or do they just think it is always the right thing to
do? Let's say you have a couple of insurance offices and want to move some
of your business to a web platform. Your budget is limited... is it reality
that something on the level of Mach ii is needed? Should the project be
abandoned if you don't have the budget to do it that way? Or could a
procedural code that took advantage of well formed objects to separate out
the data work very well? (Especially if there were a methodology that guided
the creation of simplified procedural applications?)

I have heard Steve Nelson argue in favor of procedural applications, while
others argued in favor of event oriented object based applications.
Currently I am working on code that favors the simplicity of procedural and
benefits by the separation of objects. If I were programming in Java... the
case could be made for doing it all that way. Yet in CF we have UDFs, Custom
Tags and such. It is a different environment. I am in favor of people being
"able" to do full object oriented CF development, but still question if the
object thrust some give is slanted. (People writing books, basing authority
of others writing books. Books that sight examples of poorly written
procedural code as examples of why it is wrong is bad logic. And the
strengths of procedural code don't eliminate the strengths of object any
more than the strengths of object orientation eliminates procedural.) 

Isn't the actual goal code reuse? What if you use code reuse with custom
tags, UDF Libraries, and CFCs... and where the code is not reused build with
procedural scripts? (Separating the logic from the content of course...)

Thanks,

John Farrar
SOSensible

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Ken Ingle
Sent: Friday, November 12, 2004 7:53 AM
To: [EMAIL PROTECTED]
Subject: Re: [CFCDev] Separating cohesive applications

I agree with Roland, here at our office we have migrated to Mach-ii,
and one of the first things we did for all applications was to create
a controller cfc that extended mach-ii and intern called the actual
working cfc's to do the work (which do not extend mach-ii).  The
thought here is that it does not matter if we use mach-ii or not, our
CFC's are still usable.  It is a little extra work up front, but it is
well worth it in the end.  It really provides a lot more flexibilty in
the application, it is also allowing multiple applications to work
better together, without being coupled to one another.

Ken.

----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email
to [EMAIL PROTECTED] with the words 'unsubscribe cfcdev'
in the message of the email.

CFCDev is run by CFCZone (www.cfczone.org) and supported
by Mindtool, Corporation (www.mindtool.com).

An archive of the CFCDev list is available at 
[EMAIL PROTECTED]

Reply via email to