I don't know why you would want complex inheritance hierarchies or even persisted CFCs. In fact, you may never need either. However, when you encounter a project that calls for either you might like to know ahead of time how CFCs would handle it.
You are trying to twist my statements into some generality like CFCs suck. However, I have been clear each time; CFCs have scalability issues in certain use cases; not across the board. Whether you care about those use cases is irrelevant. Matt Liotta President & CEO Montara Software, Inc. http://www.montarasoftware.com/ 888-408-0900 x901 > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf > Of Jeff Battershall > Sent: Wednesday, March 26, 2003 10:33 AM > To: [EMAIL PROTECTED] > Subject: RE: CFC scalability problems (was : RE: [CFCDev] MVCF at > benorama.com) > > Jeez, Matt. Why would we want "complex inheritance heirarchies" anyway? > That can be a poor design choice in certain cases. I don't doubt that > what > you're saying is true, but, as Benoit points out, one could simply favor > composition over inheritance as GofF recommend. And the fact that Java > might be better for certain things - does that make CFCs unusable, > unscalable? And who says CFCs are the magic panacea? This is not an all > or > nothing proposition. > > They're a great option for architecture reasons and for client support > reasons. I'm sure they'll get better over time. My experience thus far > with them has been very very positive. Call me a fool, call me an early > adopter, but I'm gonna stick by my CFCs. > > Jeff > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf > Of Matt Liotta > Sent: Wednesday, March 26, 2003 9:52 AM > To: [EMAIL PROTECTED] > Subject: RE: CFC scalability problems (was : RE: [CFCDev] MVCF at > benorama.com) > > > > Yes, I was refering to CFC method vs. a UDF and custom tag. CFCs have > > limitations for complex inheritance hierarchies > (performances, > > lack of super()). > > But anyway, if you require a very complex object model with complex > > inheritance hierarchies, you should not go for CFCs but directly for > Java > > (with some CFCs wrapper/facade if required). > > > Suggesting that Java should be used instead of CFCs just proves that > certain > use cases for CFCs have problems. > > > As for the overhead associated with stateful CFCs, I am not aware of > that > > (and it seems that Sean Corfield is not either). > > > Sean should certainly be aware of the issues as he was involved in the > threads that discussed it. > > > I am on CF-Talk (probably one of the most active technical CF list) > and I > > don't think I've ever seen this problem mentioned. > > I just made a search on "stateful CFC" and "persisted CFC" and it gave > no > > results. > > > These issues were discussed on this list and CFGURU. > > > They might have some overhead as you said. > > But I'm not sure it will prevent you to build scalable sites. > > > Because the overhead is too high. > > What I think is interesting is that people keep assuming that certain use > cases with CFC will be fine. Don't assume; test! I have tested most of the > CFC use cases and am aware of what works and doesn't. Again, I state that > complex inheritance hierarchies and persisted models for use in MVC result > in scalability problems with CFCs. > > -Matt > > ---------------------------------------------------------- > 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). > > > > > ---------------------------------------------------------- > 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). ---------------------------------------------------------- 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).
