There's buzz going on right now about the demise of Mach-II, and its "replacement" by Model-Glue. I'm excited to see Model-Glue continue to grow in popularity, but I'm hoping we can avoid sensationalism and spreading Mach-II FUD. My conclusion is the same as it usually is: be pragmatic.
Here's a few thinking points on the topic:
1. Use what's right for you.
Mach-II's author (Ben Edwards), has a more formal background in OO/Java development than I do. My background is equally split between ColdFusion and Flash. The difference in our backgrounds shows through in our frameworks, and I think one of the reasons for Model-Glue's "burst" in popularity is that it uses more terms and concepts familiar to Macromedia developers. On the flip side, it may be a "weird" framework for those coming from the Java side of the fence. So, Choose what's comfortable for you and your team, not what's trendy.
2. There's a lot of investment in Mach-II.
There are many Mach-II applications out there happily churning away. I'm sure there are many under development as well. There's been thousands of dollars of money spent on Mach-II training. That's not all going to go "foomp!" and disappear.
3. It's all about the Model, anyhow.
If you've built a decent Model, it really doesn't matter (and shouldn't be that hard) to go between Model-Glue, Mach-II, and MVC Fusebox. If your development investment is primarily in the framework you choose, something's wrong on a lower level.
4. Of course Mach-II has a slower release cycle!
...and so will Model-Glue when it hits 1.0. I'm adding a few more features people need to simplify their development lives, and not going to go much beyond that. There's only so much an MVC framework should do to begin with.
(On a side note to Ben Edwards, congrats on the new house and marriage. I bought a house last fall and got hitched in late April, and I know exactly how busy life must be for you right now!)
5. Competition drives innovation
Mach-II's largely been alone in the OO ColdFusion arena. Its existence drove the vision and goals of Model-Glue. It wouldn't surprise me to see the opposite occur sometime in the future.
6. Community participation
It makes sense that people are hearing a lot more from the Model-Glue "camp" than the Mach-II "camp." A lot of the people who have adopted Model-Glue (and helped in its development) are the same people that write a some of the more popular ColdFusion blogs and participate heavily in the community. This doesn't mean that Ben and Hal aren't listening. The Model-Glue listserve is still small enough where I can reply to almost every message - if current trends are any indication, this won't always be the case.
Also, I'm just as likely to put the kibash on feature requests as Ben - while the 0.5 -> 1.0 cycles have been community driven, there's only so much this framework should be doing. After 1.0, you can expect me to be very stubborn in terms of feature requests.
7. LGPL (somewhat off the "Mach-II is Dead" topic)
Model-Glue is released under the lesser GPL. This leaves you free to alter it to suit your own needs. If I won't add something you want, please feel free to do so. If you'd like to work on the framework after 1.0, let me know - I'd like to insure that the framework remains community-driven, and wouldn't mind sharing "reign" over the codebase. However, I'll still feel most comfortable being the 'final' say over what is 'official' Model-Glue - a lot of time has gone into things like naming convention and consistency, and I think a large part of the frameworks ease-of-use is due to this.
3 comments - Posted by Joe Rinehart at 8:15 AM - Categories: Model-Glue