thr3ads.net - Rails spinoffs - AJAX vs. SJAX and single threaded javascript [Oct 2007]

If this information is useful, please help other people find it:
Share via:

arosenzweig

2007-Oct-04 20:36 UTC

AJAX vs. SJAX and single threaded javascript

Hello,

Here''s a mind bender. Normally, we want all our Web 2.0 actions to
happen asynchronously. This allows for multiple actions to occur and
finish when they are ready. We can invoke the XMLHttpRequest
"synchronously" but there are some dangers with that. If it takes a
long time for the server to respond, the web browser will be
completely frozen during that time, giving the appearance that it has
crashed. With that background, here''s my issue...

I have a form with several text fields. When an "onchange" occurs the
data is saved to the database via AJAX. So you can imagine entering 20
items and after you tab off of each one that data gets immediately
saved to the database without further action from the user. Let''s call
this an "auto-save" and it works great.

Now imagine that in addition to the text fields, there are a few
buttons on the page. Preferably, I''d like the clicking of a button to
"wait" while any pending XMLHTTPRequests are active. Please correct me
here if I''m wrong but I believe my options are:

1) Make all auto-save calls "synchronous".

2) Leave auto-save calls in async mode and pray that they finish
before the "onclick" of the button. You can also show a message
"hey,
an auto-save is busy" when you click but that is annoying. The user
shouldn''t be blocked out and require a second click because the auto-
save needed 100 milliseconds to complete.

3) Do some polling in the "onclick" of the button and don''t
allow the
button action to continue until all XMLHTTPRequests are complete.

Right now I''m doing #2 above. Most of the time, users are tabbing and
the auto-save happens quickly so they never notice any quirkiness.
However, when they do decide to type a value, not tab off, but
immediately click the action button they get the warning message about
the auto save in progress. All of the auto-save actions "push" a value
into a global array when the request is sent then "pop" the value
after the request returns. This is how I can determine if there are
any outstanding AJAX requests by verifying that the array is empty.

Option #3 seems ideal but I can''t get it to work. It appears that the
javascript interpreter is single threaded. I just loop for eternity
checking to see when that global array becomes empty but it never
does. The interpreter doesn''t divide up its processing to allow the
array to get "popped" in an alternate thread.

Right now I''m thinking of going for option #1, but as stated above, it
has it''s own set of issues.

Does anyone have some pearls of wisdom to share? Inquiring minds want
to know.

Thanks in advance,
-- Aaron


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-04 20:39 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

:) You''re missing #4, the right answer...

Disable the buttons (or set a "saveInProgress" flag) prior to firing
autosave, and revert back to "enabled" state on success... easy
shmeasy.

On 10/4/07, arosenzweig
<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org>
wrote:>
>
> Hello,
>
> Here''s a mind bender. Normally, we want all our Web 2.0 actions to
> happen asynchronously. This allows for multiple actions to occur and
> finish when they are ready. We can invoke the XMLHttpRequest
> "synchronously" but there are some dangers with that. If it takes
a
> long time for the server to respond, the web browser will be
> completely frozen during that time, giving the appearance that it has
> crashed. With that background, here''s my issue...
>
> I have a form with several text fields. When an "onchange" occurs
the
> data is saved to the database via AJAX. So you can imagine entering 20
> items and after you tab off of each one that data gets immediately
> saved to the database without further action from the user. Let''s
call
> this an "auto-save" and it works great.
>
> Now imagine that in addition to the text fields, there are a few
> buttons on the page. Preferably, I''d like the clicking of a button
to
> "wait" while any pending XMLHTTPRequests are active. Please
correct me
> here if I''m wrong but I believe my options are:
>
> 1) Make all auto-save calls "synchronous".
>
> 2) Leave auto-save calls in async mode and pray that they finish
> before the "onclick" of the button. You can also show a message
"hey,
> an auto-save is busy" when you click but that is annoying. The user
> shouldn''t be blocked out and require a second click because the
auto-
> save needed 100 milliseconds to complete.
>
> 3) Do some polling in the "onclick" of the button and
don''t allow the
> button action to continue until all XMLHTTPRequests are complete.
>
> Right now I''m doing #2 above. Most of the time, users are tabbing
and
> the auto-save happens quickly so they never notice any quirkiness.
> However, when they do decide to type a value, not tab off, but
> immediately click the action button they get the warning message about
> the auto save in progress. All of the auto-save actions "push" a
value
> into a global array when the request is sent then "pop" the value
> after the request returns. This is how I can determine if there are
> any outstanding AJAX requests by verifying that the array is empty.
>
> Option #3 seems ideal but I can''t get it to work. It appears that
the
> javascript interpreter is single threaded. I just loop for eternity
> checking to see when that global array becomes empty but it never
> does. The interpreter doesn''t divide up its processing to allow
the
> array to get "popped" in an alternate thread.
>
> Right now I''m thinking of going for option #1, but as stated
above, it
> has it''s own set of issues.
>
> Does anyone have some pearls of wisdom to share? Inquiring minds want
> to know.
>
> Thanks in advance,
> -- Aaron
>
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-04 20:41 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

if you go the "setting the flag" route, you would just add logic in
your
button handlers that say "if not save in progress already, save, otherwise
don''t save"

personally I''d go with disabling the interaction temporarily (with the
button, not the entire browser)

On 10/4/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
> :) You''re missing #4, the right answer...
>
> Disable the buttons (or set a "saveInProgress" flag) prior to
firing
> autosave, and revert back to "enabled" state on success... easy
shmeasy.
>
> On 10/4/07, arosenzweig
<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org> wrote:
> >
> >
> > Hello,
> >
> > Here''s a mind bender. Normally, we want all our Web 2.0
actions to
> > happen asynchronously. This allows for multiple actions to occur and
> > finish when they are ready. We can invoke the XMLHttpRequest
> > "synchronously" but there are some dangers with that. If it
takes a
> > long time for the server to respond, the web browser will be
> > completely frozen during that time, giving the appearance that it has
> > crashed. With that background, here''s my issue...
> >
> > I have a form with several text fields. When an "onchange"
occurs the
> > data is saved to the database via AJAX. So you can imagine entering 20
> > items and after you tab off of each one that data gets immediately
> > saved to the database without further action from the user.
Let''s call
> > this an "auto-save" and it works great.
> >
> > Now imagine that in addition to the text fields, there are a few
> > buttons on the page. Preferably, I''d like the clicking of a
button to
> > "wait" while any pending XMLHTTPRequests are active. Please
correct me
> > here if I''m wrong but I believe my options are:
> >
> > 1) Make all auto-save calls "synchronous".
> >
> > 2) Leave auto-save calls in async mode and pray that they finish
> > before the "onclick" of the button. You can also show a
message "hey,
> > an auto-save is busy" when you click but that is annoying. The
user
> > shouldn''t be blocked out and require a second click because
the auto-
> > save needed 100 milliseconds to complete.
> >
> > 3) Do some polling in the "onclick" of the button and
don''t allow the
> > button action to continue until all XMLHTTPRequests are complete.
> >
> > Right now I''m doing #2 above. Most of the time, users are
tabbing and
> > the auto-save happens quickly so they never notice any quirkiness.
> > However, when they do decide to type a value, not tab off, but
> > immediately click the action button they get the warning message about
> > the auto save in progress. All of the auto-save actions
"push" a value
> > into a global array when the request is sent then "pop" the
value
> > after the request returns. This is how I can determine if there are
> > any outstanding AJAX requests by verifying that the array is empty.
> >
> > Option #3 seems ideal but I can''t get it to work. It appears
that the
> > javascript interpreter is single threaded. I just loop for eternity
> > checking to see when that global array becomes empty but it never
> > does. The interpreter doesn''t divide up its processing to
allow the
> > array to get "popped" in an alternate thread.
> >
> > Right now I''m thinking of going for option #1, but as stated
above, it
> > has it''s own set of issues.
> >
> > Does anyone have some pearls of wisdom to share? Inquiring minds want
> > to know.
> >
> > Thanks in advance,
> > -- Aaron
> >
> >
> > > >
> >
>
>
> --
> Ryan Gahl
> Manager, Senior Software Engineer
> Nth Penguin, LLC
> http://www.nthpenguin.com
> --
> Architect
> WebWidgetry.com / MashupStudio.com
> Future Home of the World''s First Complete Web Platform
> --
> Inquire: 1-262-951-6727
> Blog: http://www.someElement.com
> LinkedIn Profile: http://www.linkedin.com/in/ryangahl



-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org

2007-Oct-04 22:23 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

Ryan thanks for your fast reply but I think the root problem still 
remains. The desired effect is:

"The user types in some text then immediately clicks the button, after a 
brief pause for the auto-save to complete the button action should go."

I believe the solution you offered would allow us to either "allow or 
deny" the button press, but it won''t allow a delay while ajax
calls are
ongoing and then allow the button press.

In other words its not:

"Click, get message that you have to wait or have a grayed out button, 
then click again"

But rather:

"Click... brief pause, button action invokes."

Perhaps SJAX is the only way to do this due to the single threaded nature 
of the javascript interpreter. If it wasn''t single threaded I could 
possibly sleep or keep looping and waiting for all AJAX requests to come 
back before executing the button click logic.

-- Aaron




"Ryan Gahl" <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
Sent by: rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
04-10-2007 04:41 PM
Please respond to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org


To
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
cc

Subject
[Rails-spinoffs] Re: AJAX vs. SJAX and single threaded javascript






if you go the "setting the flag" route, you would just add logic in
your
button handlers that say "if not save in progress already, save, otherwise 
don''t save"

personally I''d go with disabling the interaction temporarily (with the 
button, not the entire browser) 

On 10/4/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
:) You''re missing #4, the right answer...

Disable the buttons (or set a "saveInProgress" flag) prior to firing 
autosave, and revert back to "enabled" state on success... easy
shmeasy.


On 10/4/07, arosenzweig <arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org
> wrote:

Hello,

Here''s a mind bender. Normally, we want all our Web 2.0 actions to
happen asynchronously. This allows for multiple actions to occur and
finish when they are ready. We can invoke the XMLHttpRequest 
"synchronously" but there are some dangers with that. If it takes a
long time for the server to respond, the web browser will be
completely frozen during that time, giving the appearance that it has
crashed. With that background, here''s my issue...

I have a form with several text fields. When an "onchange" occurs the
data is saved to the database via AJAX. So you can imagine entering 20
items and after you tab off of each one that data gets immediately 
saved to the database without further action from the user. Let''s call
this an "auto-save" and it works great.

Now imagine that in addition to the text fields, there are a few
buttons on the page. Preferably, I''d like the clicking of a button to 
"wait" while any pending XMLHTTPRequests are active. Please correct me
here if I''m wrong but I believe my options are:

1) Make all auto-save calls "synchronous".

2) Leave auto-save calls in async mode and pray that they finish 
before the "onclick" of the button. You can also show a message
"hey,
an auto-save is busy" when you click but that is annoying. The user
shouldn''t be blocked out and require a second click because the auto- 
save needed 100 milliseconds to complete.

3) Do some polling in the "onclick" of the button and don''t
allow the
button action to continue until all XMLHTTPRequests are complete.

Right now I''m doing #2 above. Most of the time, users are tabbing and 
the auto-save happens quickly so they never notice any quirkiness.
However, when they do decide to type a value, not tab off, but
immediately click the action button they get the warning message about
the auto save in progress. All of the auto-save actions "push" a value
into a global array when the request is sent then "pop" the value
after the request returns. This is how I can determine if there are
any outstanding AJAX requests by verifying that the array is empty. 

Option #3 seems ideal but I can''t get it to work. It appears that the
javascript interpreter is single threaded. I just loop for eternity
checking to see when that global array becomes empty but it never 
does. The interpreter doesn''t divide up its processing to allow the
array to get "popped" in an alternate thread.

Right now I''m thinking of going for option #1, but as stated above, it
has it''s own set of issues. 

Does anyone have some pearls of wisdom to share? Inquiring minds want
to know.

Thanks in advance,
-- Aaron


http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl 



-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Martin Bialasinski

2007-Oct-04 22:48 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

On 10/5/07, arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org
<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org> wrote:
> "The user types in some text then immediately clicks the button, after
a brief pause for the auto-save to complete the button action should go."
>
function buttonhandler(){
  if (autosave_in_progress){
    setTimeout(buttonhandler, 100);
  } else {
    button_action();
  }
}

and set this as the event handler for the button.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-04 22:51 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

Indeed... OP dude - it''s all about the closures. This is the kind of
thinking I was getting at...

On 10/4/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
>
> On 10/5/07, arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org
<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org> wrote:
>
> > "The user types in some text then immediately clicks the button,
after a
> brief pause for the auto-save to complete the button action should
go."
> >
>
> function buttonhandler(){
>   if (autosave_in_progress){
>     setTimeout(buttonhandler, 100);
>   } else {
>     button_action();
>   }
> }
>
> and set this as the event handler for the button.
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Martin Bialasinski

2007-Oct-04 23:28 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

On 10/5/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:> Indeed... OP dude - it''s all about the closures.
Except there are no closures involved :-)

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Richard Quadling

2007-Oct-05 08:34 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

On 04/10/2007, arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org
<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org>
wrote:>
>
> Ryan thanks for your fast reply but I think the root problem still
> remains. The desired effect is:
>
> "The user types in some text then immediately clicks the button, after
a
> brief pause for the auto-save to complete the button action should
go."
>
> I believe the solution you offered would allow us to either "allow or
> deny" the button press, but it won''t allow a delay while ajax
calls are
> ongoing and then allow the button press.
>
> In other words its not:
>
> "Click, get message that you have to wait or have a grayed out button,
> then click again"
>
> But rather:
>
> "Click... brief pause, button action invokes."
>
> Perhaps SJAX is the only way to do this due to the single threaded nature
> of the javascript interpreter. If it wasn''t single threaded I
could possibly
> sleep or keep looping and waiting for all AJAX requests to come back before
> executing the button click logic.
>
> -- Aaron
>
>
>
>  *"Ryan Gahl"
<ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>*
> Sent by: rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
>
> 04-10-2007 04:41 PM  Please respond to
> rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
>
>   To
> rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org  cc
>
>  Subject
> [Rails-spinoffs] Re: AJAX vs. SJAX and single threaded javascript
>
>
>
>
>
>
> if you go the "setting the flag" route, you would just add logic
in your
> button handlers that say "if not save in progress already, save,
otherwise
> don''t save"
>
> personally I''d go with disabling the interaction temporarily (with
the
> button, not the entire browser)
>
> On 10/4/07, *Ryan Gahl*
<*ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org*
<ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>>
> wrote:
> :) You''re missing #4, the right answer...
>
> Disable the buttons (or set a "saveInProgress" flag) prior to
firing
> autosave, and revert back to "enabled" state on success... easy
shmeasy.
>
>
> On 10/4/07, *arosenzweig*
<*arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org
*<arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org>>
> wrote:
>
> Hello,
>
> Here''s a mind bender. Normally, we want all our Web 2.0 actions to
> happen asynchronously. This allows for multiple actions to occur and
> finish when they are ready. We can invoke the XMLHttpRequest
> "synchronously" but there are some dangers with that. If it takes
a
> long time for the server to respond, the web browser will be
> completely frozen during that time, giving the appearance that it has
> crashed. With that background, here''s my issue...
>
> I have a form with several text fields. When an "onchange" occurs
the
> data is saved to the database via AJAX. So you can imagine entering 20
> items and after you tab off of each one that data gets immediately
> saved to the database without further action from the user. Let''s
call
> this an "auto-save" and it works great.
>
> Now imagine that in addition to the text fields, there are a few
> buttons on the page. Preferably, I''d like the clicking of a button
to
> "wait" while any pending XMLHTTPRequests are active. Please
correct me
> here if I''m wrong but I believe my options are:
>
> 1) Make all auto-save calls "synchronous".
>
> 2) Leave auto-save calls in async mode and pray that they finish
> before the "onclick" of the button. You can also show a message
"hey,
> an auto-save is busy" when you click but that is annoying. The user
> shouldn''t be blocked out and require a second click because the
auto-
> save needed 100 milliseconds to complete.
>
> 3) Do some polling in the "onclick" of the button and
don''t allow the
> button action to continue until all XMLHTTPRequests are complete.
>
> Right now I''m doing #2 above. Most of the time, users are tabbing
and
> the auto-save happens quickly so they never notice any quirkiness.
> However, when they do decide to type a value, not tab off, but
> immediately click the action button they get the warning message about
> the auto save in progress. All of the auto-save actions "push" a
value
> into a global array when the request is sent then "pop" the value
> after the request returns. This is how I can determine if there are
> any outstanding AJAX requests by verifying that the array is empty.
>
> Option #3 seems ideal but I can''t get it to work. It appears that
the
> javascript interpreter is single threaded. I just loop for eternity
> checking to see when that global array becomes empty but it never
> does. The interpreter doesn''t divide up its processing to allow
the
> array to get "popped" in an alternate thread.
>
> Right now I''m thinking of going for option #1, but as stated
above, it
> has it''s own set of issues.
>
> Does anyone have some pearls of wisdom to share? Inquiring minds want
> to know.
>
> Thanks in advance,
> -- Aaron
>
>
> http://www.nthpenguin.com
> --
> Architect
> WebWidgetry.com / MashupStudio.com
> Future Home of the World''s First Complete Web Platform
> --
> Inquire: 1-262-951-6727
> Blog: *http://www.someElement.com* <http://www.someelement.com/>
> LinkedIn Profile:
*http://www.linkedin.com/in/ryangahl*<http://www.linkedin.com/in/ryangahl>
>
>
>
> --
> Ryan Gahl
> Manager, Senior Software Engineer
> Nth Penguin, LLC*
> **http://www.nthpenguin.com* <http://www.nthpenguin.com/>
> --
> Architect
> WebWidgetry.com / MashupStudio.com
> Future Home of the World''s First Complete Web Platform
> --
> Inquire: 1-262-951-6727
> Blog: *http://www.someElement.com* <http://www.someelement.com/>
> LinkedIn Profile:
*http://www.linkedin.com/in/ryangahl*<http://www.linkedin.com/in/ryangahl>
>
> >
>
Maybe something like this ...

function buttonhandler(){
  if (0 != Ajax.activeRequestCount){
    setTimeout(buttonhandler, 100);
  } else {
    button_action();
  }
}

The activeRequestCount property is auto-maintained by using an
onCreate/onComplete Responder (In V1.6.0_RC0 look at lines 1182 - 1189).

You may need to also add your own onException handler to deal with things
going wrong within an Ajax callback and maybe resulting in the
activeRequestCount NOT being zero due to an exception - not sure how to
generate an exception to test this.




-- 
-----
Richard Quadling
Zend Certified Engineer : http://zend.com/zce.php?c=ZEND002498&r=213474731
"Standing on the shoulders of some very clever giants!"

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-05 13:37 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

When you call setTimout the buttonhandler reference becomes a closure :)

On 10/4/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
>
> On 10/5/07, Ryan Gahl
<ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > Indeed... OP dude - it''s all about the closures.
>
> Except there are no closures involved :-)
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Martin Bialasinski

2007-Oct-05 16:46 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

On 10/5/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:> When you call setTimout the buttonhandler reference becomes a closure :)
I don''t quite understand what you mean. A variable is not a closure.
When the function gets executed by the timeout, it does not share
neither scope nor execution context with the first invocation. There
is no forming of a closure anywhere.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-05 17:00 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

Yes it does...

autosave_in_progress is closed within the function reference you pass in.
Just because you don''t have to use an anonymous function does not mean
it''s
not a closure. setTimeout is based on closures -- you pass in a function
reference and it can be executed any time in the future (even well after the
initial scope where you defined buttonhandler or the variable
autosave_in_progress is long gone therefore making that uncallable
elsewhere). It would not work if it wasn''t using a closure.

function foo() {

var autosave_in_progress;

function buttonhandler(){
 if (autosave_in_progress){
   setTimeout(buttonhandler, 100);
 } else {
   button_action();
 }
}

}();

Can you see how it''s a closure now? In the above, foo is executed, and
the
buttonhandler function is defined within it, but is executed only _after_
that scope is gone. Both the variable autosave_in_progress and the function
buttonhandler as closed over.

On 10/5/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
>
> On 10/5/07, Ryan Gahl
<ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > When you call setTimout the buttonhandler reference becomes a closure
:)
>
> I don''t quite understand what you mean. A variable is not a
closure.
> When the function gets executed by the timeout, it does not share
> neither scope nor execution context with the first invocation. There
> is no forming of a closure anywhere.
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

arosenzweig-fbjq1H7RxnhWk0Htik3J/w@public.gmane.org

2007-Oct-05 19:42 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

Thank you very much - Ryan Gahl, Martin Bialasinski, and Richard Quadling.

That''s a neat design pattern you got there. I like it a lot! I wanted
to
take a moment to tell you what I discovered today.

I found that for our app SJAX (Synchronous AJAX) calls were very quick and 
actually afforded a better user experience. It''s easier to chain the
order
of updates on the screen and the "pausing" effect I was looking for
just
happened without additional work. We may change our mind later and come up 
with something else related to your design pattern but for right now,
it''s
good. If there is something that would cause the reply to take a long 
time, there is probably something wrong with the server. For hard core 
data entry (which is what our app is about), it''s not good to have you 
think that a lot of asynchronous saves got saved when they didn''t...
it''s
better to see the spinning pinwheel as soon as there is a problem with the 
server.

I tweaked your design pattern slightly and experimented with it:

function buttonhandler(){
  if (autosave_in_progress()){
    setTimeout(buttonhandler, 100);
  } else {
    button_action();
  }
  return false; //This is the small one line tweak
}

The "autosave_in_progress" is a function that could be defined in any 
particular way to return a "true" or a "false". It might
just look at "0
!= Ajax.activeRequestCount" but it could do something else.

This is neat in that it gives that kind of "polling" idea which allows
for
ajax calls to do their thing and for you to keep checking for their 
completion before you perform another action. You still have to be careful 
you hit all the possiblities like Hrefs, Submit buttons, etc. You also 
have to be careful that action doesn''t actually continue down
it''s path
that''s why I put in a "return false" there at the end. Think
of an
href.onclick where you don''t want the clicking of the link to
immediately
invoke, but only after "auto_save_in_progress()" evaluates to
"true". In
the case of a "onbeforeunload" (when you are leaving or refreshing the
page) you might have to use SJAX, it''s probably the only way. Without
SJAX
for the onbeforeunload, you''d have to make an immediate decision if you
want stop the user or let them move on.

Cheers,
--Aaron




"Ryan Gahl" <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 
Sent by: rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
05-10-2007 01:00 PM
Please respond to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org


To
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
cc

Subject
[Rails-spinoffs] Re: AJAX vs. SJAX and single threaded javascript






Yes it does... 

autosave_in_progress is closed within the function reference you pass in. 
Just because you don''t have to use an anonymous function does not mean 
it''s not a closure. setTimeout is based on closures -- you pass in a 
function reference and it can be executed any time in the future (even 
well after the initial scope where you defined buttonhandler or the 
variable autosave_in_progress is long gone therefore making that 
uncallable elsewhere). It would not work if it wasn''t using a closure. 

function foo() {

var autosave_in_progress;

function buttonhandler(){
 if (autosave_in_progress){
   setTimeout(buttonhandler, 100);
 } else {
   button_action();
 }
}

}(); 


Can you see how it''s a closure now? In the above, foo is executed, and
the
buttonhandler function is defined within it, but is executed only _after_ 
that scope is gone. Both the variable autosave_in_progress and the 
function buttonhandler as closed over. 










On 10/5/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

On 10/5/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote: > When you call setTimout the buttonhandler reference becomes a closure :)
I don''t quite understand what you mean. A variable is not a closure.
When the function gets executed by the timeout, it does not share 
neither scope nor execution context with the first invocation. There
is no forming of a closure anywhere.

http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Martin Bialasinski

2007-Oct-05 19:51 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

We might be speaking about different things here.

When calling setTimeout() with a function reference, the JS engine
does not form a new closure as part of the setTimeout inner workings.

If setTimeout would create a closure, this example sould log "1" and
not "undefined":

function foo(){
  console.log(test);
}

function bar(){
  var test = 1;
  setTimeout(foo, 100);
}

bar();

What you see in your example is that buttonhandler itself forms a
closure as it accesses a variable in its outer scope. There is a
closure because of the way the function is defined, not because
setTimeout got a function reference.

Please correct me anyone if my understanding of the situation is off.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-05 20:00 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

It''s still creating a closure, but now it''s not closing over
the variable
inside bar (obvisouly since you are not passing bar as the function
reference). Any local vars that are available within or defined by foo will
have been closed over though.

function blah() {
  var hooblyHoo = 1;

  function foo() {
    alert(hooblyHoo);
  }

  setTimeout(foo, 100);
}();

I made this one simpler... it''s a closure... no ifs ands or buts about
it.
So is the original one.

When calling setTimeout() with a function reference, the JS
engine> does not form a new closure as part of the setTimeout inner workings.

Incorrect.




On 10/5/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
>
> We might be speaking about different things here.
>
> When calling setTimeout() with a function reference, the JS engine
> does not form a new closure as part of the setTimeout inner workings.
>
> If setTimeout would create a closure, this example sould log "1"
and
> not "undefined":
>
> function foo(){
>   console.log(test);
> }
>
> function bar(){
>   var test = 1;
>   setTimeout(foo, 100);
> }
>
> bar();
>
> What you see in your example is that buttonhandler itself forms a
> closure as it accesses a variable in its outer scope. There is a
> closure because of the way the function is defined, not because
> setTimeout got a function reference.
>
> Please correct me anyone if my understanding of the situation is off.
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-05 20:04 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

function blah() {
  var auto_save_in_progress = true;

  function buttonhandler() {
    if (auto_save_in_progress)
      setTimeout(buttonhandler, 100);
  }

  buttonhandler()
}();

''tis a closure just like the previous foo example, only difference is
it''s
harder to see because of its recursive nature.



On 10/5/07, Ryan Gahl <ryan.gahl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
> It''s still creating a closure, but now it''s not closing
over the variable
> inside bar (obvisouly since you are not passing bar as the function
> reference). Any local vars that are available within or defined by foo will
> have been closed over though.
>
> function blah() {
>   var hooblyHoo = 1;
>
>   function foo() {
>     alert(hooblyHoo);
>   }
>
>   setTimeout(foo, 100);
> }();
>
> I made this one simpler... it''s a closure... no ifs ands or buts
about it.
> So is the original one.
>
> When calling setTimeout() with a function reference, the JS engine
> > does not form a new closure as part of the setTimeout inner workings.
>
>
> Incorrect.
>
>
>
>
> On 10/5/07, Martin Bialasinski
<klingeling-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> >
> >
> > We might be speaking about different things here.
> >
> > When calling setTimeout() with a function reference, the JS engine
> > does not form a new closure as part of the setTimeout inner workings.
> >
> > If setTimeout would create a closure, this example sould log
"1" and
> > not "undefined":
> >
> > function foo(){
> >   console.log (test);
> > }
> >
> > function bar(){
> >   var test = 1;
> >   setTimeout(foo, 100);
> > }
> >
> > bar();
> >
> > What you see in your example is that buttonhandler itself forms a
> > closure as it accesses a variable in its outer scope. There is a
> > closure because of the way the function is defined, not because
> > setTimeout got a function reference.
> >
> > Please correct me anyone if my understanding of the situation is off.
> >
> > > >
> >
>
>
> --
> Ryan Gahl
> Manager, Senior Software Engineer
> Nth Penguin, LLC
> http://www.nthpenguin.com
> --
> Architect
> WebWidgetry.com / MashupStudio.com
> Future Home of the World''s First Complete Web Platform
> --
> Inquire: 1-262-951-6727
> Blog: http://www.someElement.com
> LinkedIn Profile: http://www.linkedin.com/in/ryangahl
>


-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Lisa B

2007-Oct-06 20:23 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

you are not going to be able to use a global array to track requests,
which is why you can''t get #3 to work.  The array will stay the same
as when you sent it..  but if you wrap your array in a class it will
accept changes while requests are in progress, it will behave in the
global sense that you want.  This is how Prototype already provides a
means to tell if there are outstanding requests.  You can inspect the
variable Ajax.activeRequestCount to see if requests are still active.
You can also make your own that are arrays and are more customized,
but if all you want to know is ''are there requests out'' then
that
answer is already available to your script you are just not using it
because you are trying to use javascript global vars, which won''t
work.   Open up prototype and see how it''s handled..
activeRequestCount



--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Lisa B

2007-Oct-06 20:27 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

I think you should accept the button click rather than fade the
button, but then hold off on sending it until Ajax.activeRequestCount
= 0..  this way you can process that last entry that the user made
just before the button click.. and they won''t know anything but that a
slight hang is happening on button click..  which is no surprise to
most users..


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Ryan Gahl

2007-Oct-06 20:29 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

Except that some users will undoubtedly click multiple times because of the
hang. Disabling interaction immediately following said interaction is a
_very_ common UI development practice, for good reason.

On 10/6/07, Lisa B
<beforeamillionuniverses-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:>
>
> I think you should accept the button click rather than fade the
> button, but then hold off on sending it until Ajax.activeRequestCount
> = 0..  this way you can process that last entry that the user made
> just before the button click.. and they won''t know anything but
that a
> slight hang is happening on button click..  which is no surprise to
> most users..
>
>
> >
>

-- 
Ryan Gahl
Manager, Senior Software Engineer
Nth Penguin, LLC
http://www.nthpenguin.com
--
Architect
WebWidgetry.com / MashupStudio.com
Future Home of the World''s First Complete Web Platform
--
Inquire: 1-262-951-6727
Blog: http://www.someElement.com
LinkedIn Profile: http://www.linkedin.com/in/ryangahl

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Lisa B

2007-Oct-07 02:30 UTC

head link

Re: AJAX vs. SJAX and single threaded javascript

yes disabling the button on its own click is necessary any time you
take over its default behavior and there could be a lag..  I was
referring to disabling the button on each and every request triggered
as other form fields are activated.. would be nicer to avoid that if
possible..


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---

Rails spinoffs - Oct 2007 - AJAX vs. SJAX and single threaded javascript

AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript

Re: AJAX vs. SJAX and single threaded javascript