I agree, EJB 3 looks very good. But my point was that the first two iterations of the spec didn't help. And if anything, the non-spec driven Spring framework drove EJB 3 to be what it is today.
Clinton On Wed, Apr 7, 2010 at 11:06 AM, Guy Rouillier <guyr-...@burntmail.com>wrote: > On 4/7/2010 9:50 AM, Clinton Begin wrote: > >> >> This is what has killed J2EE vs. the alternatives. Look at the history: >> > > Not to start a flame war, but ... > > > >> * CMP - Spec. Dead, along with all implementations. >> >> * EJB - Spec. Dead. Spring killed it -- not a spec. >> > > EJB3 is very much alive. The spec people learned from experience; EJB3 > uses straight POJOs. Spring tries to be all things, and doesn't really do > any of them exceptionally well. I don't use it because of its massive > footprint. Sure, you can use subcomponents in isolation, if you can figure > out how to decompose it. But I prefer targeted tools that do one thing very > well. Hence, my attraction to iBATIS. > > > >> * JDO - Spec. Dead, along with all implementations. >> >> * JSF - DOA. Bad idea to begin with, and has failed to unify client >> side Java. Struts, GWT, Wickett, Stripes, ZK, Tapestry, etc. all still >> exist -- and are more popular than JSF -- all without a spec. >> > > JSF use is increasing, not decreasing. My company has committed to it for > several major projects in the last 6 months alone. We've also used GWT for > other projects. JSF is more of a server-based technology than a client > technology, and different applications may need one approach or the other. > I'm sure each of the tools you list has pluses and minuses; I haven't used > them all. Struts I think has seen its day, and usage is declining in favor > of newer approaches. > > -- > Guy Rouillier > > -- > Guy Rouillier > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: user-java-unsubscr...@ibatis.apache.org > For additional commands, e-mail: user-java-h...@ibatis.apache.org > >
