You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jcp-open@apache.org by "Geir Magnusson Jr." <ge...@pobox.com> on 2007/02/16 00:34:59 UTC
JSR-311 update
I just got off the phone with Sun, and they will be making the
changes Roy requested. The JSR itself can't be changed during the
voting period, but they will be advising the EC of the change in a
note soon-ish. We'll reference that in our vote, which will be yes.
In terms of organizing who wants to participate, I want to come up
with a list of volunteers. So far, I have :
Justin
Craig - wants to participate, not lead
Dims
Steve (?)
Did I miss anyone?
geir
Re: JSR-311 update
Posted by Dan Diephouse <da...@envoisolutions.com>.
OK, well if I can at least be on the list on the Apache end I will be happy.
:-)
- Dan
On 2/18/07, Justin Erenkrantz <ju...@erenkrantz.com> wrote:
>
> On Sun, Feb 18, 2007 at 08:29:46PM -0500, Dan Diephouse wrote:
> > I would be interested in participating.
> >
> > Can someone clarify one thing for me: can multiple Apache people
> > participate? Or is there one liaison between the JSR committee and a
> mailing
> > list of Apache folks?
>
> I believe it depends on how the spec lead wants to run it. At their
> choice, they can limit to one EG per group. If they do that, we can
> always
> create a list on our end. -- justin
>
--
Dan Diephouse
Envoi Solutions
http://envoisolutions.com | http://netzooid.com/blog
Re: JSR-311 update
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On Sun, Feb 18, 2007 at 08:29:46PM -0500, Dan Diephouse wrote:
> I would be interested in participating.
>
> Can someone clarify one thing for me: can multiple Apache people
> participate? Or is there one liaison between the JSR committee and a mailing
> list of Apache folks?
I believe it depends on how the spec lead wants to run it. At their
choice, they can limit to one EG per group. If they do that, we can always
create a list on our end. -- justin
Re: JSR-311 update
Posted by "Geir Magnusson Jr." <ge...@pobox.com>.
On Feb 18, 2007, at 8:29 PM, Dan Diephouse wrote:
> I would be interested in participating.
>
> Can someone clarify one thing for me: can multiple Apache people
> participate? Or is there one liaison between the JSR committee and
> a mailing list of Apache folks?
Depends on the spec lead, and what they allow. I try to push for at
least 2.
geir
>
> - Dan
>
> On 2/16/07, Steve Loughran <st...@apache.org> wrote: Geir
> Magnusson Jr. wrote:
> > I just got off the phone with Sun, and they will be making the
> changes
> > Roy requested. The JSR itself can't be changed during the voting
> > period, but they will be advising the EC of the change in a note
> > soon-ish. We'll reference that in our vote, which will be yes.
> >
> > In terms of organizing who wants to participate, I want to come
> up with
> > a list of volunteers. So far, I have :
> >
> > Justin
> > Craig - wants to participate, not lead
> > Dims
> > Steve (?)
>
> I may participate, but am currently uncommitted, Realistically, I
> would
> like the whole thing to be an OSS project, including the TCK, so
> anyone
> can do it. But I see that as unlikely.
>
> if there is an apache list on the topic, yes, I'd subscribe.
>
> to be honest, I'm more interested in what makes a good client API for
> restiness, something which models long haul communications properly,
> something built on httpclient. That is definately something we can
> do in
> house.
>
>
> >
> > Did I miss anyone?
>
> Dan Diephouse?
>
> Anyone from Cocoon?
>
>
>
> --
> Dan Diephouse
> Envoi Solutions
> http://envoisolutions.com | http://netzooid.com/blog
Re: JSR-311 update
Posted by Dan Diephouse <da...@envoisolutions.com>.
I would be interested in participating.
Can someone clarify one thing for me: can multiple Apache people
participate? Or is there one liaison between the JSR committee and a mailing
list of Apache folks?
- Dan
On 2/16/07, Steve Loughran <st...@apache.org> wrote:
>
> Geir Magnusson Jr. wrote:
> > I just got off the phone with Sun, and they will be making the changes
> > Roy requested. The JSR itself can't be changed during the voting
> > period, but they will be advising the EC of the change in a note
> > soon-ish. We'll reference that in our vote, which will be yes.
> >
> > In terms of organizing who wants to participate, I want to come up with
> > a list of volunteers. So far, I have :
> >
> > Justin
> > Craig - wants to participate, not lead
> > Dims
> > Steve (?)
>
> I may participate, but am currently uncommitted, Realistically, I would
> like the whole thing to be an OSS project, including the TCK, so anyone
> can do it. But I see that as unlikely.
>
> if there is an apache list on the topic, yes, I'd subscribe.
>
> to be honest, I'm more interested in what makes a good client API for
> restiness, something which models long haul communications properly,
> something built on httpclient. That is definately something we can do in
> house.
>
>
> >
> > Did I miss anyone?
>
> Dan Diephouse?
>
> Anyone from Cocoon?
>
--
Dan Diephouse
Envoi Solutions
http://envoisolutions.com | http://netzooid.com/blog
Re: JSR-311 update
Posted by Steve Loughran <st...@apache.org>.
Geir Magnusson Jr. wrote:
> I just got off the phone with Sun, and they will be making the changes
> Roy requested. The JSR itself can't be changed during the voting
> period, but they will be advising the EC of the change in a note
> soon-ish. We'll reference that in our vote, which will be yes.
>
> In terms of organizing who wants to participate, I want to come up with
> a list of volunteers. So far, I have :
>
> Justin
> Craig - wants to participate, not lead
> Dims
> Steve (?)
I may participate, but am currently uncommitted, Realistically, I would
like the whole thing to be an OSS project, including the TCK, so anyone
can do it. But I see that as unlikely.
if there is an apache list on the topic, yes, I'd subscribe.
to be honest, I'm more interested in what makes a good client API for
restiness, something which models long haul communications properly,
something built on httpclient. That is definately something we can do in
house.
>
> Did I miss anyone?
Dan Diephouse?
Anyone from Cocoon?
Re: JSR-311 update
Posted by Steve Loughran <st...@apache.org>.
Geir Magnusson Jr. wrote:
> <sigh>
>
> Thanks for the info...
>
> THis may or may not be the same
I feel like we are part of the 1957 congress of the soviet union got to
vote in favour of their leaders of actions. "The representatives of
Hungary would like to thank their fraternal brothers for fullfilling
their obligations under the Warsaw Pact and rescuing our country from
the counter-revolutionaries"
-steve
Re: JSR-311 update
Posted by "Geir Magnusson Jr." <ge...@pobox.com>.
<sigh>
Thanks for the info...
THis may or may not be the same
On Mar 13, 2007, at 6:25 AM, Steve Loughran wrote:
> Justin Erenkrantz wrote:
>> On 2/15/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
>>> I just got off the phone with Sun, and they will be making the
>>> changes Roy requested. The JSR itself can't be changed during the
>>> voting period, but they will be advising the EC of the change in a
>>> note soon-ish. We'll reference that in our vote, which will be yes.
>>>
>>> In terms of organizing who wants to participate, I want to come up
>>> with a list of volunteers. So far, I have :
>>>
>>> Justin
>>> Craig - wants to participate, not lead
>>> Dims
>>> Steve (?)
>>>
>>> Did I miss anyone?
>> Any updates on this? Any magic voodoo I have to perform? -- justin
>
>
> sun have just released a 'Sun Web Developer Pack": http://
> developers.sun.com/web/swdp/
>
> one of its features is a: RESTful Web Services API (Early Access)
>
> http://developers.sun.com/web/swdp/docs/OnePagerREST_APIs.html
>
> The development of RESTful Web services using the current APIs
> (Servlet or JAX-WS) requires a great deal of boiler-plate code that
> could usefully be provided by a container. In addition developers
> are required to understand a great deal of information related to
> HTTP in order to built services that work correctly with the Web
> architecture. Sun Web Developer Pack ships with an Early Access set
> of APIs and a runtime for components build using those APIs to
> address this problem. The runtime undertake much of the HTTP-
> centric functionality and only delegate to components as necessary.
> This greatly simplify the development of RESTful Web services on
> the Java platform by allowing the developer to concentrate on the
> functionality of their service rather than having to deal with HTTP
> minutiae.
>
> The APIs enshrine concepts from REST and promote a RESTful approach
> to building services:
>
> * Developers will write POJOs that represent a Web resource.
> * Ease of development features are provided by use of Java
> annotations that allow declarative specification of resource metadata.
> * Input to, and output from, resource methods is in the form of
> representations which encapsulate arbitrary Java data types
> * The URI at which a resource is published/deployed is set
> declaratively using a Java annotation
> * Additional helper classes are provided as deemed necessary
> during development of sample services
>
> The APIs is independent of deployment technology and will allow
> services to be hosted as Servlets, JAX-WS endpoints and in the
> lightweight Java SE HTTP server. Code will be provided to adapt
> these containers to hosting of RESTful services developed using the
> new APIs.
>
> ...
>
> I am very confused. Surely the purpose of JSR-311 was to come up
> with these APIs? But Sun have just released an early access version
> of the same APIs? How can that be, unless someone has worked out a
> way to send emails backwards in time?
>
> -steve
Re: JSR-311 update
Posted by "Geir Magnusson Jr." <ge...@pobox.com>.
On Mar 13, 2007, at 11:15 AM, Justin Erenkrantz wrote:
> On 3/13/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
>> I'm was hoping they'd follow the convention set w/ java.util.logging
>> and call it javas.xml.ws.resting or javax.xml.webservicing.resting
>
> I thought we agreed on the JCP proposal that the API shouldn't have
> 'REST' in the name? Not to mention that 'RESTful' is a complete joke.
> If nothing else, I hope to minimize such foolishness from the JSR.
IIRC, Sun has stated that they will use javax.ws.rs as the package name.
geir
Re: JSR-311 update
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On 3/13/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
> I'm was hoping they'd follow the convention set w/ java.util.logging
> and call it javas.xml.ws.resting or javax.xml.webservicing.resting
I thought we agreed on the JCP proposal that the API shouldn't have
'REST' in the name? Not to mention that 'RESTful' is a complete joke.
If nothing else, I hope to minimize such foolishness from the JSR.
-- justin
Re: JSR-311 update
Posted by "Geir Magnusson Jr." <ge...@pobox.com>.
I'm was hoping they'd follow the convention set w/ java.util.logging
and call it javas.xml.ws.resting or javax.xml.webservicing.resting
geir
On Mar 13, 2007, at 11:06 AM, Steve Loughran wrote:
> Geir Magnusson Jr. wrote:
>> that's the conclusion I came to... that they did something as part
>> of the toolkit.
>
> as is their right.
>
>> What will be interesting is to see if they want to use the JSR to
>> rubberstamp what they concocted so their toolkit isn't instantly
>> obsolete.
>
>
> Yes, the warning signs there are
> 1. the EA tag, "Early Access". Early access to what?
> 2. the packaging com.sun.ws.rest.api.* is not far off
> javax.xml.ws.rest, is it?
>
> -steve
Re: JSR-311 update
Posted by Steve Loughran <st...@apache.org>.
Geir Magnusson Jr. wrote:
> that's the conclusion I came to... that they did something as part of
> the toolkit.
>
as is their right.
> What will be interesting is to see if they want to use the JSR to
> rubberstamp what they concocted so their toolkit isn't instantly obsolete.
Yes, the warning signs there are
1. the EA tag, "Early Access". Early access to what?
2. the packaging com.sun.ws.rest.api.* is not far off
javax.xml.ws.rest, is it?
-steve
Re: JSR-311 update
Posted by "Geir Magnusson Jr." <ge...@pobox.com>.
that's the conclusion I came to... that they did something as part of
the toolkit.
What will be interesting is to see if they want to use the JSR to
rubberstamp what they concocted so their toolkit isn't instantly
obsolete.
geir
On Mar 13, 2007, at 10:13 AM, Dan Diephouse wrote:
> More technical details are here:
>
> http://developers.sun.com/web/swdp/docs/tutorial/doc/p35.html
>
> I think its what one of the members was thinking for JSR 311.
> However, they make no claims about 311, so it seems entirely
> reasonable to me that they can create their own proprietary APIs.
>
> - Dan
>
> On 3/13/07, Steve Loughran <st...@apache.org> wrote: Justin
> Erenkrantz wrote:
> > On 2/15/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
> >> I just got off the phone with Sun, and they will be making the
> >> changes Roy requested. The JSR itself can't be changed during the
> >> voting period, but they will be advising the EC of the change in a
> >> note soon-ish. We'll reference that in our vote, which will be
> yes.
> >>
> >> In terms of organizing who wants to participate, I want to come up
> >> with a list of volunteers. So far, I have :
> >>
> >> Justin
> >> Craig - wants to participate, not lead
> >> Dims
> >> Steve (?)
> >>
> >> Did I miss anyone?
> >
> > Any updates on this? Any magic voodoo I have to perform? -- justin
>
>
> sun have just released a 'Sun Web Developer Pack":
> http://developers.sun.com/web/swdp/
>
> one of its features is a: RESTful Web Services API (Early Access)
>
> http://developers.sun.com/web/swdp/docs/OnePagerREST_APIs.html
>
> The development of RESTful Web services using the current APIs
> (Servlet
> or JAX-WS) requires a great deal of boiler-plate code that could
> usefully be provided by a container. In addition developers are
> required
> to understand a great deal of information related to HTTP in order to
> built services that work correctly with the Web architecture. Sun Web
> Developer Pack ships with an Early Access set of APIs and a runtime
> for
> components build using those APIs to address this problem. The runtime
> undertake much of the HTTP-centric functionality and only delegate to
> components as necessary. This greatly simplify the development of
> RESTful Web services on the Java platform by allowing the developer to
> concentrate on the functionality of their service rather than
> having to
> deal with HTTP minutiae.
>
> The APIs enshrine concepts from REST and promote a RESTful approach to
> building services:
>
> * Developers will write POJOs that represent a Web resource.
> * Ease of development features are provided by use of Java
> annotations that allow declarative specification of resource metadata.
> * Input to, and output from, resource methods is in the form of
> representations which encapsulate arbitrary Java data types
> * The URI at which a resource is published/deployed is set
> declaratively using a Java annotation
> * Additional helper classes are provided as deemed necessary
> during
> development of sample services
>
> The APIs is independent of deployment technology and will allow
> services
> to be hosted as Servlets, JAX-WS endpoints and in the lightweight Java
> SE HTTP server. Code will be provided to adapt these containers to
> hosting of RESTful services developed using the new APIs.
>
> ...
>
> I am very confused. Surely the purpose of JSR-311 was to come up with
> these APIs? But Sun have just released an early access version of the
> same APIs? How can that be, unless someone has worked out a way to
> send
> emails backwards in time?
>
> -steve
>
>
>
> --
> Dan Diephouse
> Envoi Solutions
> http://envoisolutions.com | http://netzooid.com/blog
Re: JSR-311 update
Posted by Dan Diephouse <da...@envoisolutions.com>.
More technical details are here:
http://developers.sun.com/web/swdp/docs/tutorial/doc/p35.html
I think its what one of the members was thinking for JSR 311. However, they
make no claims about 311, so it seems entirely reasonable to me that they
can create their own proprietary APIs.
- Dan
On 3/13/07, Steve Loughran <st...@apache.org> wrote:
>
> Justin Erenkrantz wrote:
> > On 2/15/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
> >> I just got off the phone with Sun, and they will be making the
> >> changes Roy requested. The JSR itself can't be changed during the
> >> voting period, but they will be advising the EC of the change in a
> >> note soon-ish. We'll reference that in our vote, which will be yes.
> >>
> >> In terms of organizing who wants to participate, I want to come up
> >> with a list of volunteers. So far, I have :
> >>
> >> Justin
> >> Craig - wants to participate, not lead
> >> Dims
> >> Steve (?)
> >>
> >> Did I miss anyone?
> >
> > Any updates on this? Any magic voodoo I have to perform? -- justin
>
>
> sun have just released a 'Sun Web Developer Pack":
> http://developers.sun.com/web/swdp/
>
> one of its features is a: RESTful Web Services API (Early Access)
>
> http://developers.sun.com/web/swdp/docs/OnePagerREST_APIs.html
>
> The development of RESTful Web services using the current APIs (Servlet
> or JAX-WS) requires a great deal of boiler-plate code that could
> usefully be provided by a container. In addition developers are required
> to understand a great deal of information related to HTTP in order to
> built services that work correctly with the Web architecture. Sun Web
> Developer Pack ships with an Early Access set of APIs and a runtime for
> components build using those APIs to address this problem. The runtime
> undertake much of the HTTP-centric functionality and only delegate to
> components as necessary. This greatly simplify the development of
> RESTful Web services on the Java platform by allowing the developer to
> concentrate on the functionality of their service rather than having to
> deal with HTTP minutiae.
>
> The APIs enshrine concepts from REST and promote a RESTful approach to
> building services:
>
> * Developers will write POJOs that represent a Web resource.
> * Ease of development features are provided by use of Java
> annotations that allow declarative specification of resource metadata.
> * Input to, and output from, resource methods is in the form of
> representations which encapsulate arbitrary Java data types
> * The URI at which a resource is published/deployed is set
> declaratively using a Java annotation
> * Additional helper classes are provided as deemed necessary during
> development of sample services
>
> The APIs is independent of deployment technology and will allow services
> to be hosted as Servlets, JAX-WS endpoints and in the lightweight Java
> SE HTTP server. Code will be provided to adapt these containers to
> hosting of RESTful services developed using the new APIs.
>
> ...
>
> I am very confused. Surely the purpose of JSR-311 was to come up with
> these APIs? But Sun have just released an early access version of the
> same APIs? How can that be, unless someone has worked out a way to send
> emails backwards in time?
>
> -steve
>
--
Dan Diephouse
Envoi Solutions
http://envoisolutions.com | http://netzooid.com/blog
Re: JSR-311 update
Posted by Steve Loughran <st...@apache.org>.
Roy T. Fielding wrote:
> On Mar 13, 2007, at 3:25 AM, Steve Loughran wrote:
>> I am very confused. Surely the purpose of JSR-311 was to come up with
>> these APIs? But Sun have just released an early access version of the
>> same APIs? How can that be, unless someone has worked out a way to
>> send emails backwards in time?
>
> Nobody in their right mind starts a standards committee without already
> having implemented something that works as they intend the first draft
> of the standard to work. It is very rare for such standards to be
> changed much by the committee, because design-by-committee always sucks.
> Usually, either the committee focuses on specifying the details of
> what works, or a second party brings their own implementation into
> the committee and everyone changes course to specify that.
>
> ....Roy
I agree, protos are essential. but at the same time, calling it "Early
Access"? That implies that it is already seen as the way forward, when
there were other approaches -especially restlet- on the table.
Re: JSR-311 update
Posted by "Roy T. Fielding" <fi...@gbiv.com>.
On Mar 13, 2007, at 3:25 AM, Steve Loughran wrote:
> I am very confused. Surely the purpose of JSR-311 was to come up
> with these APIs? But Sun have just released an early access version
> of the same APIs? How can that be, unless someone has worked out a
> way to send emails backwards in time?
Nobody in their right mind starts a standards committee without already
having implemented something that works as they intend the first draft
of the standard to work. It is very rare for such standards to be
changed much by the committee, because design-by-committee always sucks.
Usually, either the committee focuses on specifying the details of
what works, or a second party brings their own implementation into
the committee and everyone changes course to specify that.
....Roy
Re: JSR-311 update
Posted by Steve Loughran <st...@apache.org>.
Justin Erenkrantz wrote:
> On 2/15/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
>> I just got off the phone with Sun, and they will be making the
>> changes Roy requested. The JSR itself can't be changed during the
>> voting period, but they will be advising the EC of the change in a
>> note soon-ish. We'll reference that in our vote, which will be yes.
>>
>> In terms of organizing who wants to participate, I want to come up
>> with a list of volunteers. So far, I have :
>>
>> Justin
>> Craig - wants to participate, not lead
>> Dims
>> Steve (?)
>>
>> Did I miss anyone?
>
> Any updates on this? Any magic voodoo I have to perform? -- justin
sun have just released a 'Sun Web Developer Pack":
http://developers.sun.com/web/swdp/
one of its features is a: RESTful Web Services API (Early Access)
http://developers.sun.com/web/swdp/docs/OnePagerREST_APIs.html
The development of RESTful Web services using the current APIs (Servlet
or JAX-WS) requires a great deal of boiler-plate code that could
usefully be provided by a container. In addition developers are required
to understand a great deal of information related to HTTP in order to
built services that work correctly with the Web architecture. Sun Web
Developer Pack ships with an Early Access set of APIs and a runtime for
components build using those APIs to address this problem. The runtime
undertake much of the HTTP-centric functionality and only delegate to
components as necessary. This greatly simplify the development of
RESTful Web services on the Java platform by allowing the developer to
concentrate on the functionality of their service rather than having to
deal with HTTP minutiae.
The APIs enshrine concepts from REST and promote a RESTful approach to
building services:
* Developers will write POJOs that represent a Web resource.
* Ease of development features are provided by use of Java
annotations that allow declarative specification of resource metadata.
* Input to, and output from, resource methods is in the form of
representations which encapsulate arbitrary Java data types
* The URI at which a resource is published/deployed is set
declaratively using a Java annotation
* Additional helper classes are provided as deemed necessary during
development of sample services
The APIs is independent of deployment technology and will allow services
to be hosted as Servlets, JAX-WS endpoints and in the lightweight Java
SE HTTP server. Code will be provided to adapt these containers to
hosting of RESTful services developed using the new APIs.
...
I am very confused. Surely the purpose of JSR-311 was to come up with
these APIs? But Sun have just released an early access version of the
same APIs? How can that be, unless someone has worked out a way to send
emails backwards in time?
-steve
Re: JSR-311 update
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On 2/15/07, Geir Magnusson Jr. <ge...@pobox.com> wrote:
> I just got off the phone with Sun, and they will be making the
> changes Roy requested. The JSR itself can't be changed during the
> voting period, but they will be advising the EC of the change in a
> note soon-ish. We'll reference that in our vote, which will be yes.
>
> In terms of organizing who wants to participate, I want to come up
> with a list of volunteers. So far, I have :
>
> Justin
> Craig - wants to participate, not lead
> Dims
> Steve (?)
>
> Did I miss anyone?
Any updates on this? Any magic voodoo I have to perform? -- justin