And wizards are helpful when you are not familiar with the application of
the tool, they are a friendly introduction to doing something.

Agreed, but ideally, the tool should be so elegant and self- explanatory that its natural interface is itself a friendly introduction. Obviously that's idealistic, but as ideals go, I think it's a good one.

My point is really that if you find yourself building a wizard, you should first ask yourself what in your design makes a wizard necessary, and see if can be simplified.

don't keep me on the edge of my seat, but what does "your" Tap 5 not have
that makes it better than Tap 4?

Umm... hmmm ... good question. Some ideas off the cuff:

I find myself writing too many throwaway get/set methods that are there only for the benefit of the template, and unused by any Java code. Perhaps some sort of implicit property mechanism could fix this.

The newly consolidated For component is an improvement, but still complex and confusing. One of the few things I miss about JSP is the standard taglib's iteration mechanisms.

Because of the way its link encoders work, Tapestry unnecessarily requires many separate servlet-mapping elements in web.xml in situations where a single mapping could cover many URL encodings. (I had an argument with HLS about this in Jira. Sorry, Howard.)

I'm sick of typing "ognl:".

There's perpetual confusion about passing parameters between pages. Perhaps the many types of link components could be consolidated into a single Link component, accompanied by a consolidation of the underlying services. I realize this is architecturally complicated, but from a user's point of view, it makes sense: I can link to any page (but this one by default), optionally invoking a listener, optionally passing it parameters. Why shouldn't that just be a single Link component? (The complexity of the answer to that question suggests that simplification is possible!)

So there's some ideas. They may be bad ideas, but I do think further aggressive simplification of this sort is a good thing. It's the sort of thing Tapestry is already doing; no reason to stop at 4.

Cheers,

Paul

_________________________________________________________________
Piano music podcast: http://inthehands.com
Other interesting stuff: http://innig.net



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to