"Best Tool for the Job" is a Myth
when it comes to web frameworks
Browsing around the Q&A on JavaRanch.com I see a lot of the same questions when it comes to web frameworks. To paraphrase:
"Which web framework should I choose and why?"
I and others often respond with the typical canned answer. This answer comes in many flavors; "There is no golden hammer", "We can't answer without the context of your project", "Every project requires its own investigation to answer this question", etc, etc. The meaning of those answers is all the same, however. But is it true?
The last 3 projects I've worked on I really pushed for, and was able to use, Stripes. I really like Stripes and since I am not a "full stack" person I generally like to pick and choose each portion of my solution puzzle. Stripes is my go to web MVC piece of that puzzle. And I honestly can't recall that during the discovery phase of any of these projects I honestly considered anything else. (on my current project we did discuss Symfony vs Stripes but that is more of a language debate than framework debate). Is that a mistake? I don't think so. If you use, like, and favor Wicket do you really consider something like Struts 2 or Spring MVC when you are getting ready to start a new project? I'd be willing to bet you don't.
I think we all like to think that we are open minded about this sort of thing but the passion on blogs and forums (or flame wars if you prefer) really prove otherwise. Most developers think their tool box contains the golden hammer(s).
What do you think?
Re: "Best Tool for the Job" is a Myth
One of the things I do as a professional, is realizing solutions that were the best available at one point no longer are. At that point, you must be able to admit this, and choose differently.
But when using a full-stack framework, this is generally not possible. This is why I favor the UNIX approach: use different pieces together. And as an added advantage, I can only use those pieces I actually need.