John D Farrar wrote:
Dave...what OS do you use? Could you point out the OS that is "Fully Encapsulated"? That would be a good project for someone to create to show us the evils of all the other systems. HEH... and again... system variables... and program variables are not the same issue.
I think it would be very difficult to show that any operating system is fully encapsulated, but that doesn't mean that almost anyone who writes operating systems doesn't aim to achieve encapsulation. It also doesn't mean that they do, so it becomes something of a hypothetical discussion pretty quickly.
A counter question is to ask you to point out an OS that deliberately breaks encapsulation. I don't know of any, but I'm not an OS guru. Do you? If so, is it generally accepted that it was a good or a bad idea to do that and if so, why?
On the point that going to a new system (based on scope conflicts) could make the CFC's obsolete. Does that mean if the interface has to be updated for the new system that encapsulation isn't any good either. Your arguments fail in critical thinking. You should go take a course in critical thinking to see if you could prose your argument better. You may have a point... but you haven't made it yet.
I tend to look at it from the point of view of code reuse. I may have a future application which requires the same or very similar functionality, but where that functionality isn't based around the CGI variables. If I can reuse the component without touching the code inside it then I can have a reasonable degree of confidence that it will not introduce bugs into the new application. Even if I do happen to be re-using the component in the same manner, if I am on a different platform and using a different web server the CGI variables may very well be different. If I have ensured that the access to the CGI variables is encapsulated then I _know_ there is only one place where I will need to modify the code for the new environment.
Further more... I think Sean's thoughts about encapsulation are "mostly" correct. My debate is the thinking that it is an "absolute rule". I am not likely to agree that calling a cgi variable inside a CFC is a violation of good code. Start there... and I would honestly like to see it if there is a point. Sean supported the point by saying it violated encapsulation. That is what we call circular reasoning.
I think it's a valid challenge anything that hasn't been proven to be correct in your experience, but I'm leaning more and more to the point of view that the rule is to always use encapsulation and look for use cases where using it will cause a problem. If I find one I'll happily break the rule. So far I haven't found _any_ places where it causes more problems than it solves, but every application is different, so if you have a use case where encapsulation is a bad thing I'd like to hear about it.
Lastly... someday we will do away with windows. There will be a better way. So... do you think they should have stopped at windows 95? Windows NT? Which version would have been good enough for you?
I must have missed something in the thread. I'm not sure what you're referring to here.
Spike
--
-------------------------------------------- Stephen Milligan Code poet for hire http://www.spike.org.uk
Do you cfeclipse? http://cfeclipse.tigris.org
----------------------------------------------------------
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 www.mail-archive.com/[EMAIL PROTECTED]
