You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@tapestry.apache.org by Chubby Johnson <mr...@hotmail.com> on 2004/04/17 19:02:38 UTC
Tapestry fundementally flawed - ?
It seems to me that Tapestry is fundementally flawed.
The fancy page pooling may save a few milliseconds, but will easily allow
the application to return Visitor A data to Visitor B--the ultimate "no-no".
I can't think of a worse design decision, especially since cycles are so
cheap these days.
Anyone?
-Chubby
_________________________________________________________________
FREE pop-up blocking with the new MSN Toolbar get it now!
http://toolbar.msn.com/go/onm00200415ave/direct/01/
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by Erik Hatcher <er...@ehatchersolutions.com>.
On Apr 17, 2004, at 1:02 PM, Chubby Johnson wrote:
> It seems to me that Tapestry is fundementally flawed.
Yeah, I suppose the NHL, Darden, and the many other companies
developing with it are completely out of their mind, eh? :)
> The fancy page pooling may save a few milliseconds, but will easily
> allow the application to return Visitor A data to Visitor B--the
> ultimate "no-no".
How 'bout some proof to back this up.
> I can't think of a worse design decision, especially since cycles are
> so cheap these days.
We look forward to your contributions to fix the flaws :)
Erik
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by Cherry Development <av...@cherrydev.com>.
On Apr 20, 2004, at 5:15 PM, David Ezzio wrote:
> I'm willing to give you the benefit of the doubt. You didn't mean to
> be hostile, you just don't know a lot about Tapestry.
>
> The deal is this: the pages contain their components which may contain
> more components. In addition, they refer to XML configuration and
> HTML templates. So there is a fair amount of "static" state that does
> not change while the application is running. The dynamic state is
> contained in the page and component propeties. Some of these
> properties are "persistent" in Tapestry terminology -- meaning that
> their state is swapped back and forth from the HttpSession object at
> the beginning and end of each request. The remaining "transient" page
> properties are simply initialized at the start of each request. Net
> result: a page may be used for several requests from several users,
> but each request acquires the page after it has been initialized with
> the state that belongs to the user making the request. Its elegant
> and efficient, but it is a little complicated.
I'd actually argue that it is not elegant. I actually consider this to
be Tapestry's roughest edge at the moment. I would love to see a
change made that would allow a developer to use a "one page instance
per page request" model if they choose to. After having spent several
years developing in WebObjects, which is very, very similar to Tapestry
except in that particular respect, I've found that even after 8 months
working in Tapestry, I find the WebObjects method much more natural and
elegant. I'm sure a way could be found to overcome the large amount of
setup required for each new instance of a page. As I see it, all the
'static' state could be pooled, and each page instance could simply be
a thin wrapper around that state.
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by David Ezzio <de...@ysoft.com>.
Hi Chubby,
I'm willing to give you the benefit of the doubt. You didn't mean to be
hostile, you just don't know a lot about Tapestry.
The deal is this: the pages contain their components which may contain
more components. In addition, they refer to XML configuration and HTML
templates. So there is a fair amount of "static" state that does not
change while the application is running. The dynamic state is contained
in the page and component propeties. Some of these properties are
"persistent" in Tapestry terminology -- meaning that their state is
swapped back and forth from the HttpSession object at the beginning and
end of each request. The remaining "transient" page properties are
simply initialized at the start of each request. Net result: a page may
be used for several requests from several users, but each request
acquires the page after it has been initialized with the state that
belongs to the user making the request. Its elegant and efficient, but
it is a little complicated.
David
Chubby Johnson wrote:
> It seems to me that Tapestry is fundementally flawed.
>
> The fancy page pooling may save a few milliseconds, but will easily
> allow the application to return Visitor A data to Visitor B--the
> ultimate "no-no".
>
> I can't think of a worse design decision, especially since cycles are so
> cheap these days.
>
> Anyone?
>
> -Chubby
>
> _________________________________________________________________
> FREE pop-up blocking with the new MSN Toolbar – get it now!
> http://toolbar.msn.com/go/onm00200415ave/direct/01/
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by "Kevin C. Dorff" <kd...@kcp.com>.
But if your page as a form and you are using pageBeginRender make sure
to check if you are rewinding.
Kevin
based@free.fr wrote:
> Jeff Emminger wrote:
>
>> >> but will easily allow the application to return Visitor A data to
>> Visitor B--the ultimate "no-no".
>> guess you didn't read about initialize() and detach()?
>
> hmm, http://wiki.apache.org/jakarta-tapestry/Gotchas
> '
> initialize() is sometimes mistakenly used as the place to initialize
> properties of a page. Actually, initialize() is used to set properties
> to the state they should be in when returned to the pool--in other
> words, when you are done with the page (typically set properties to
> null). Instead, implement PageRenderListener and then initialize your
> properties in pageBeginRender().
> '
>
>>
>>
>> Chubby Johnson wrote:
>>
>>> It seems to me that Tapestry is fundementally flawed.
>>>
>>> The fancy page pooling may save a few milliseconds, but will easily
>>> allow the application to return Visitor A data to Visitor B--the
>>> ultimate "no-no".
>>>
>>> I can't think of a worse design decision, especially since cycles
>>> are so cheap these days.
>>>
>>> Anyone?
>>>
>>> -Chubby
>>>
>>> _________________________________________________________________
>>> FREE pop-up blocking with the new MSN Toolbar – get it now!
>>> http://toolbar.msn.com/go/onm00200415ave/direct/01/
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
>>> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>>>
>>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by "based@free.fr" <ba...@free.fr>.
Jeff Emminger wrote:
> >> but will easily allow the application to return Visitor A data to
> Visitor B--the ultimate "no-no".
> guess you didn't read about initialize() and detach()?
hmm, http://wiki.apache.org/jakarta-tapestry/Gotchas
'
initialize() is sometimes mistakenly used as the place to initialize
properties of a page. Actually, initialize() is used to set properties
to the state they should be in when returned to the pool--in other
words, when you are done with the page (typically set properties to
null). Instead, implement PageRenderListener and then initialize your
properties in pageBeginRender().
'
>
>
> Chubby Johnson wrote:
>
>> It seems to me that Tapestry is fundementally flawed.
>>
>> The fancy page pooling may save a few milliseconds, but will easily
>> allow the application to return Visitor A data to Visitor B--the
>> ultimate "no-no".
>>
>> I can't think of a worse design decision, especially since cycles are
>> so cheap these days.
>>
>> Anyone?
>>
>> -Chubby
>>
>> _________________________________________________________________
>> FREE pop-up blocking with the new MSN Toolbar – get it now!
>> http://toolbar.msn.com/go/onm00200415ave/direct/01/
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
Re: Tapestry fundementally flawed - ?
Posted by Jeff Emminger <je...@jeffemminger.com>.
>> but will easily allow the application to return Visitor A data to
Visitor B--the ultimate "no-no".
guess you didn't read about initialize() and detach()?
Chubby Johnson wrote:
> It seems to me that Tapestry is fundementally flawed.
>
> The fancy page pooling may save a few milliseconds, but will easily
> allow the application to return Visitor A data to Visitor B--the
> ultimate "no-no".
>
> I can't think of a worse design decision, especially since cycles are so
> cheap these days.
>
> Anyone?
>
> -Chubby
>
> _________________________________________________________________
> FREE pop-up blocking with the new MSN Toolbar – get it now!
> http://toolbar.msn.com/go/onm00200415ave/direct/01/
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
RE: Tapestry fundementally flawed - ?
Posted by "Howard M. Lewis Ship" <hl...@comcast.net>.
Yes, that's a great way to get some attention.
Now, if you want to learn the way to get *good* attention, please read:
http://jakarta.apache.org/tapestry/problems.html
and
http://www.catb.org/~esr/faqs/smart-questions.html
Then wait a while before posting an insulting email.
Following that, you can read the documentation to learn why your basic supposition is flawed.
--
Howard M. Lewis Ship
Independent J2EE / Open-Source Java Consultant
Creator, Tapestry: Java Web Components
Creator, HiveMind
http://howardlewisship.com
> -----Original Message-----
> From: Chubby Johnson [mailto:mr_chubby_johnson@hotmail.com]
> Sent: Saturday, April 17, 2004 1:03 PM
> To: tapestry-user@jakarta.apache.org
> Subject: Tapestry fundementally flawed - ?
>
>
> It seems to me that Tapestry is fundementally flawed.
>
> The fancy page pooling may save a few milliseconds, but will
> easily allow
> the application to return Visitor A data to Visitor B--the
> ultimate "no-no".
>
> I can't think of a worse design decision, especially since
> cycles are so
> cheap these days.
>
> Anyone?
>
> -Chubby
>
> _________________________________________________________________
> FREE pop-up blocking with the new MSN Toolbar - get it now!
> http://toolbar.msn.com/go/onm00200415ave/direct/01/
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tapestry-user-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: tapestry-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tapestry-user-help@jakarta.apache.org