I'm not sure I agree with all (or even most) of Erik's critique, but
I do agree with one of his points. I'd rather see the simple things get
simpler and more elegant in Tapestry than see more advanced "complex" stuff
work its way in.
I can always write my own complex stuff for the weird situations I
find myself in. I tend to look to frameworks to make the easy stuff easy and
let me focus on the hard stuff. It's sort of like why I write all my stuff
in java these days instead of C++. Once in a blue moon I'll run into a
situation I'll get annoyed at and say "this would be easy in C++", but 95%
of the time it's just cleaner and faster in java.
That's where I at least would love to see tapestry evolve.
Simplicity and elegance, not brutish and powerful.
--- Pat
> -----Original Message-----
> From: Erik Hatcher [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, July 13, 2005 5:51 PM
> To: Tapestry users
> Subject: Re: Future of Tapestry in questionable
>
>
> On Jul 13, 2005, at 1:37 PM, Joel Trunick wrote:
> >
> > I think WebObjects "got it right", and Tapestry supplied the Java
> > solution, along with now Shale (Struts 2.0), not to mention Wicket and
> > Japple. It's relatively clear to me that the Tapestry way of
> > separating
> > concerns is a winner and the approach should win out in the long
> > run (at
> > least for more intensive development efforts).
> >
> > Ruby on Rails has figured out how to do OR with ActiveRecords, but
> > haven't figured out the presentation layer (still ala JSP), so it's
> > personally not yet interesting to me as a web framework.
>
> I've been building an application with Rails for the past couple of
> months and the JSP-like nature of its templating turned me off at
> first glance. But don't let that influence you yet - there is a fair
> bit of cleverness there that beats JSP and taglibs. For example,
> templates are processed inside-out, similar to the @Border way of
> doing things - the nested view could conceivably control what the
> "layout" does. While there is no rewind cycle to automatically bind
> parameters back to objects, the namespacing of form fields and Ruby's
> elegant syntax makes this trivial.
>
> Rails AJAX integration as well as spectacular DHTML effects are
> trivial to use. In Tapestry it is complex still - even though the
> tight Tapestry JavaScript glue is present.
>
> While the complete separation of concerns of Tapestry is what drew me
> there, it is not necessary to follow this strictly and dogmatically.
> While this may be a tough thing to hear on this e-mail list, I would
> venture to say that a skilled Tapestry developer would be slower to
> build the same web application than a skilled Rails developer, and
> that the Rails application would be able to have a much nicer modern
> look and feel to it almost trivially while the Tapestry application
> would struggle to compare. There are many other factors that can be
> compared here such as maintenance, scalability, testability and
> performance... and I would even go out on a limb and say that Rails
> still has an advantage. Why is that? Simple the language it is
> written in.
>
> I don't want to sound fatalistic about Tapestry, but merely speaking
> from my experience and observations. Tapestry would do well to
> evolve based on such critique... simpler, not more complex, tight
> AJAX integration, clean URLs and elegant ways to map into them,
> simpler hyperlinking, and so on. This is an opportunity! The
> community needs to pick up the ball and contribute in any way
> possible to make this happen and to keep the positive momentum flowing.
>
> Erik
>
>
> > The point is if you believe it's the technology for the long haul,
> > it's
> > a moot question whether Tapestry dies, can simply switch to whatever
> > makes it (as some WebObjects developers have done). But, you will be
> > ahead because you will understand the structure and concepts (same as
> > those that started learning OO on C++, Smalltalk, etc.).
> >
> > Joel
> >
> > -----Original Message-----
> > From: Ahmed Mohombe [mailto:[EMAIL PROTECTED]
> > Sent: Wednesday, July 13, 2005 12:22 PM
> > To: [email protected]
> > Subject: Re: Future of Tapestry in questionable
> >
> >
> >> Recent views expressed by Howard on his blog and discussions on the
> >>
> > net
> >
> >> makes me question the future of Tapestry. He is sounding more and
> >> more
> >>
> > pro
> >
> >> Ruby on Rails and a bit anti Java. So the question is would Tapestry
> >>
> > end up
> >
> >> along the way like one of those abandoned OS projects with no
> >>
> > development
> >
> >> activity whatsover? This is something I think worth thinking about by
> >>
> > we
> >
> >> Tapestry users.
> >>
> > Interesting.
> >
> >
> >> To be honest, I'm taking a serious look at Wicket now.
> >>
> > Well, what's the guarantee that Wicket won't be abandoned long before
> > Tapestry? :). Besides, it's a little to much buzz around Wicket -
> > something that looks like a soap bubble :) (IMHO).
> >
> >
> >> What do you think?
> >>
> > An official statement from Howard would be the best IMHO :).
> >
> > Ahmed.
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
