Recently at the JavaRanch we had a promotion for a book on Tapestry. I like the book and took the time to say so.

There were a reasonable number of conversations during the promotion about Tapestry in general and how it worked, but there were also several about comparing Tapestry to other web frameworks, and (to me) suggesting that since there where other more popular candidates, that Tapestry should not be considered. Now to be fair to Alexander, he was there to promote his book and shouldn't need to support all-things-tapestry or be knowledgable on all available options in order to answer all if the "compare and contrast" questions.

I believe these questions were asked in good faith, but then after seeing the blog Why people hate Tapestry (or even Howard) I felt I needed to reply, and then follow up in this blog with my own thoughts.

I agree. I am intrigued by Tapestry but not necessarily a 'fan', but I saw comments asking why Tapestry should exist given that there are things that are 'better' already available. But given this logic, does it mean we should have discourages Struts when pure JSPs were considered 'good enough'?
It may be that Tapestry (or even the current version) isn't the final step towards component driven web sites, but I'd hate to think that this was taken as reason not to try.

Now I'm no fan of Struts and never really have been; I don't attack it or spend my time ranting and raving, but I'm not recommending it for any of the work I do. Any of the other webby frameworks would be evaluated before use too. There are a large number of web frameworks cropping up all over the place and everyone seems to want their own to love and hold, and it is possible that this rejection of frameworks beyond the fringe is in part responsible for the attitude towards Tapestry. But to quote myself "I'd hate to think that this was taken as reason not to try".

Tapestry intrigues me. Using a component-based model for web frameworks rather than action-based intrigues me. It's all very interesting and may or may not be the next phase of web frameworks, but I would rather see people trying things out and failing rather than accepting the same solution and trying nothing.