Is dyu an OpenID Provider?

[Alex Karasulu]

Hi all,

I looked around and just saw support for a RP. Was wondering if dyu
can also act as an OpenID Provider?

If not, how much of an effort would it require to get there? I can
contribute towards this end if there is an interest.

Are there other OpenID provider libraries out there?

Thanks,
Alex

[David Yu]

On Mon, Feb 28, 2011 at 2:27 AM, Alex Karasulu <akar...@gmail.com> wrote:

Hi all,

I looked around and just saw support for a RP. Was wondering if dyu
can also act as an OpenID Provider?

Currently its only an RP. 

If not, how much of an effort would it require to get there? I can
contribute towards this end if there is an interest.

It would be great if you can.  I'm sure a lot of ppl will find it useful.

P.S

The openid module probably needs to be another project on its own.

I've been planning to slim down the project (strip down the dependencies) and refactor a bit to 

avoid a hard dependency on the servlet api. (So it can still work when deployed on non-servlet-based http servers)

Sadly I've been pre-occupied with other stuff.  If an interested developer out there wants to commit to the project, let me know.

Are there other OpenID provider libraries out there?

Thanks,
Alex

--
You received this message because you are subscribed to the Google Groups "dyuproject" group.
To post to this group, send email to dyupr...@googlegroups.com.
To unsubscribe from this group, send email to dyuproject+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/dyuproject?hl=en.

--
When the cat is away, the mouse is alone.
- David Yu

[Alex Karasulu]

On Mon, Feb 28, 2011 at 4:03 AM, David Yu <david....@gmail.com> wrote:
>
>
> On Mon, Feb 28, 2011 at 2:27 AM, Alex Karasulu <akar...@gmail.com> wrote:
>>
>> Hi all,
>>
>> I looked around and just saw support for a RP. Was wondering if dyu
>> can also act as an OpenID Provider?
>
> Currently its only an RP.

OK thanks for the heads up.

>>
>> If not, how much of an effort would it require to get there? I can
>> contribute towards this end if there is an interest.
>
> It would be great if you can.  I'm sure a lot of ppl will find it useful.

I'm just trying to understand the amount of effort it would take to
pull this off. You have any rough estimate of what it could take?

Also I found JOID which has OP functionality but the project seems
much less active. Perhaps, if you don't have any objections an inquiry
can be made about merging both implementations and communities. I'd be
glad to put in extensive time, whatever it may take, to push this
through. I just want to make sure there's critical community mass to
the project before depending on it all over the place.

If you want to try to grow more community I can gladly help with that.
I'm heavily involved with a few projects at the Apache Software
Foundation. I'm the original author of the Apache Directory Server and
the founder of the project. I also co-founded a few other projects at
the ASF having mentored some of them through the Apache Incubator. If
you and the community here are interested we can also talk about
incubation to get more people involved and contributing.

> P.S
> The openid module probably needs to be another project on its own.

I hate name dropping projects but I was also involved with AsyncWeb
based on Apache MINA. It could be a great basis for an asynchronous
non-servlet based HTTP transport for the OpenID protocol.

> I've been planning to slim down the project (strip down the dependencies)
> and refactor a bit to
> avoid a hard dependency on the servlet api. (So it can still work when
> deployed on non-servlet-based http servers)
> Sadly I've been pre-occupied with other stuff.

I know how it is. Luckily I have an itch to scratch these days around OpenID.

> If an interested developer
> out there wants to commit to the project, let me know.

Well I am interested. I'm going to start taking a closer look over the
next week or two both at JOID and this code base.

Thanks,
Alex

[David Yu]

On Mon, Feb 28, 2011 at 11:37 AM, Alex Karasulu <akar...@apache.org> wrote:

On Mon, Feb 28, 2011 at 4:03 AM, David Yu <david....@gmail.com> wrote:
>
>
> On Mon, Feb 28, 2011 at 2:27 AM, Alex Karasulu <akar...@gmail.com> wrote:
>>
>> Hi all,
>>
>> I looked around and just saw support for a RP. Was wondering if dyu
>> can also act as an OpenID Provider?
>
> Currently its only an RP.

OK thanks for the heads up.

>>
>> If not, how much of an effort would it require to get there? I can
>> contribute towards this end if there is an interest.
>
> It would be great if you can.  I'm sure a lot of ppl will find it useful.

I'm just trying to understand the amount of effort it would take to
pull this off. You have any rough estimate of what it could take?

If you're basing it on other OP's impl (like joid as you mentioned), it would probably take less than a week.

It would certainly be an advantage If you're already familiar with the openid protocol.

Also I found JOID which has OP functionality but the project seems
much less active. Perhaps, if you don't have any objections an inquiry
can be made about merging both implementations and communities. I'd be
glad to put in extensive time, whatever it may take, to push this
through. I just want to make sure there's critical community mass to
the project before depending on it all over the place.

Sure.  Btw have you looked at simply using openid4java?  Their OP impl has been there for a long time.

If you want to try to grow more community I can gladly help with that.
I'm heavily involved with a few projects at the Apache Software
Foundation. I'm the original author of the Apache Directory Server and
the founder of the project. I also co-founded a few other projects at
the ASF having mentored some of them through the Apache Incubator. If
you and the community here are interested we can also talk about
incubation to get more people involved and contributing.

The community (if that exists) here is not active either ... being in an apache incubator might help change that.

> P.S
> The openid module probably needs to be another project on its own.

I hate name dropping projects but I was also involved with AsyncWeb
based on Apache MINA. It could be a great basis for an asynchronous
non-servlet based HTTP transport for the OpenID protocol.

> I've been planning to slim down the project (strip down the dependencies)
> and refactor a bit to
> avoid a hard dependency on the servlet api. (So it can still work when
> deployed on non-servlet-based http servers)
> Sadly I've been pre-occupied with other stuff.

I know how it is. Luckily I have an itch to scratch these days around OpenID.

> If an interested developer
> out there wants to commit to the project, let me know.

Well I am interested. I'm going to start taking a closer look over the
next week or two both at JOID and this code base.

Cool 

Thanks,
Alex

--
You received this message because you are subscribed to the Google Groups "dyuproject" group.
To post to this group, send email to dyupr...@googlegroups.com.
To unsubscribe from this group, send email to dyuproject+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/dyuproject?hl=en.

[Alex Karasulu]

David thanks so much for you feedback. I'm going to look into
openid4java as well and get back to you.

Best Regards,
Alex

[Andreas Kuckartz]

Maybe some kind of collaboration with Apache Rave makes sense:
http://wiki.apache.org/incubator/RaveProposal

That project is currently on its way to the Apache Incubator.

BTW: What about collaborating with SocialAuth
(http://code.google.com/p/socialauth/)?

Cheers,
Andreas
---