Security Redesign and Release 10.x Branch

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
70 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
Today I learnt that Command + Shift + D means Send Message and doesn't activate the mouse over dictionary in OSX.

Regards
Scott

On 7/04/2010, at 4:57 PM, Scott Gray wrote:

> I would suggest to:
> 1) release 10.04 before the merge is done
> 2) merge the code to the trunk, switch to it, fix any possible issue
> 3) do another release (10.06?)
>
> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>
> Jacopo
>
> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>
>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>
>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>
>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>
>> -Adrian
>>
>>
>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Adrian Crum
In reply to this post by Scott Gray-2
If we wait, then we're waiting for evaluation and testing of the branch.
I've done all I can do - the code is written, I suggested we do the
merge before the release branch, and I gave my reasons for suggesting it.

At this point in time I have stepped out of the discussion (in a
positive way) to give others a chance to look at the design and the code
and decide for themselves if it should be included. In other words, I
don't want to be in a position where I have to convince the community
what it should do. If the design and the implementation are good, then
there will be no need to convince anyone, right?

I'll answer questions about the executioncontext branch, and I'll
continue to work on it here and there when I have the time. If the
release branch is created without it, then that will be fine with me.

:-)

-Adrian


Scott Gray wrote:

> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  
>
> If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
> HotWax Media
> http://www.hotwaxmedia.com
>
>
> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>
>> I would suggest to:
>> 1) release 10.04 before the merge is done
>> 2) merge the code to the trunk, switch to it, fix any possible issue
>> 3) do another release (10.06?)
>>
>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>
>> Jacopo
>>
>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>
>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>
>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>
>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>
>>> -Adrian
>>>
>>>
>>>
>>>
>
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
Question:
What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?

I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.

Regards
Scott

On 7/04/2010, at 5:14 PM, Adrian Crum wrote:

> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>
> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>
> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>
> :-)
>
> -Adrian
>
>
> Scott Gray wrote:
>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>> HotWax Media
>> http://www.hotwaxmedia.com
>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>> I would suggest to:
>>> 1) release 10.04 before the merge is done
>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>> 3) do another release (10.06?)
>>>
>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>
>>> Jacopo
>>>
>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>
>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>
>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>
>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>
>>>> -Adrian
>>>>
>>>>
>>>>
>>>>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Adrian Crum
The security redesign implementation itself is mostly finished. There
are a few TODOs and they can be found in the BranchReadMe.txt file.

I recently synchronized the branch with the trunk and there is a remote
chance something in the design might have broken in the process. I need
to run some tests and review the code to see if that happened.

The Example component has been switched over to the new design.

There is a user login called "artifact-user" that demonstrates the new
design. That user login is restricted to using the Example component.

If the branch was merged back to the trunk and the new security design
was enabled, the Example component would use the new design and the
remaining components would still use the current security design. The
two can co-exist.

I imagine the process after that would be similar to when we introduced
the permission checking services - contributors can contribute code that
converts parts of the project over to the new security design.
Conversion involves removing hard-coded permission checks and creating
seed data to grant permission to component artifacts.

As I mentioned before, switching a component over to the new design can
create some unexpected problems. That's because our existing code has
security holes in it, and the new design plugs those holes - making
parts of the component unreachable. In other words, parts of code that
happily allow you to do things you don't have permission to do will
start to throw exceptions in the new design.

-Adrian


Scott Gray wrote:

> Question:
> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>
> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>
> Regards
> Scott
>
> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>
>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>
>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>
>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>
>> :-)
>>
>> -Adrian
>>
>>
>> Scott Gray wrote:
>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>> HotWax Media
>>> http://www.hotwaxmedia.com
>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>> I would suggest to:
>>>> 1) release 10.04 before the merge is done
>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>> 3) do another release (10.06?)
>>>>
>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>
>>>> Jacopo
>>>>
>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>
>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>
>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>
>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>
>>>>> -Adrian
>>>>>
>>>>>
>>>>>
>>>>>
>
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.

Regards
Scott

On 7/04/2010, at 5:35 PM, Adrian Crum wrote:

> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>
> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>
> The Example component has been switched over to the new design.
>
> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>
> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>
> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>
> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>
> -Adrian
>
>
> Scott Gray wrote:
>> Question:
>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>> Regards
>> Scott
>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>
>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>
>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>
>>> :-)
>>>
>>> -Adrian
>>>
>>>
>>> Scott Gray wrote:
>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>> HotWax Media
>>>> http://www.hotwaxmedia.com
>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>> I would suggest to:
>>>>> 1) release 10.04 before the merge is done
>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>> 3) do another release (10.06?)
>>>>>
>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>
>>>>> Jacopo
>>>>>
>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>
>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>
>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>
>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>
>>>>>> -Adrian
>>>>>>
>>>>>>
>>>>>>
>>>>>>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Anil Patel-3
I am yet to see contents of BranchReadMe.txt. In case it has some todo that are not classified as bug fix but are instead classified as enhancements, I will like it better if we officially allowed back porting such code to release branch.

Similarly, Jacques will like to get layered lookup part also included in release branch. Lets say if we allowed back porting of that feature as well. It will make him happy :) I also see it as major upgrade and will like to see it in release.

If community does not officially do it, then I will have to create vendor branch in my private repository of community release branch anyway :)

Other then this, Most of the code changes these days are in ofbiz trunk are at framework level. I am interested in branch so I get isolation from continues changes in framework. Most of those changes increase risk for delivery of project. I found myself in tough spot because of those and finally decide to not use trunk for projects.

All these little things are my reason for pushing for release branch sooner then later.

Thanks and Regards
Anil Patel
HotWax Media Inc
Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"

On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:

> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>
> Regards
> Scott
>
> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>
>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>
>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>
>> The Example component has been switched over to the new design.
>>
>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>
>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>
>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>
>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>
>> -Adrian
>>
>>
>> Scott Gray wrote:
>>> Question:
>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>> Regards
>>> Scott
>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>
>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>
>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>
>>>> :-)
>>>>
>>>> -Adrian
>>>>
>>>>
>>>> Scott Gray wrote:
>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>> HotWax Media
>>>>> http://www.hotwaxmedia.com
>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>> I would suggest to:
>>>>>> 1) release 10.04 before the merge is done
>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>> 3) do another release (10.06?)
>>>>>>
>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>
>>>>>> Jacopo
>>>>>>
>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>
>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>
>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>
>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
That is a slippery slope to walk on and as soon as we begin backporting features we just end up with another trunk.  Why not just wait until the features we deem important are ready before branching?

Regards
Scott

On 7/04/2010, at 6:06 PM, Anil Patel wrote:

> I am yet to see contents of BranchReadMe.txt. In case it has some todo that are not classified as bug fix but are instead classified as enhancements, I will like it better if we officially allowed back porting such code to release branch.
>
> Similarly, Jacques will like to get layered lookup part also included in release branch. Lets say if we allowed back porting of that feature as well. It will make him happy :) I also see it as major upgrade and will like to see it in release.
>
> If community does not officially do it, then I will have to create vendor branch in my private repository of community release branch anyway :)
>
> Other then this, Most of the code changes these days are in ofbiz trunk are at framework level. I am interested in branch so I get isolation from continues changes in framework. Most of those changes increase risk for delivery of project. I found myself in tough spot because of those and finally decide to not use trunk for projects.
>
> All these little things are my reason for pushing for release branch sooner then later.
>
> Thanks and Regards
> Anil Patel
> HotWax Media Inc
> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>
> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>
>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>
>> Regards
>> Scott
>>
>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>
>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>
>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>
>>> The Example component has been switched over to the new design.
>>>
>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>
>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>
>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>
>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>
>>> -Adrian
>>>
>>>
>>> Scott Gray wrote:
>>>> Question:
>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>> Regards
>>>> Scott
>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>
>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>
>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>
>>>>> :-)
>>>>>
>>>>> -Adrian
>>>>>
>>>>>
>>>>> Scott Gray wrote:
>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>> HotWax Media
>>>>>> http://www.hotwaxmedia.com
>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>> I would suggest to:
>>>>>>> 1) release 10.04 before the merge is done
>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>> 3) do another release (10.06?)
>>>>>>>
>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>
>>>>>>> Jacopo
>>>>>>>
>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>
>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>
>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>
>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>
>>>>>>>> -Adrian
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Adrian Crum-2
In reply to this post by Anil Patel-3
--- On Wed, 4/7/10, Anil Patel <[hidden email]> wrote:
> I am yet to see contents of
> BranchReadMe.txt.

The file is in the root folder of the branch.



     
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-4
In reply to this post by Scott Gray-2
On Apr 8, 2010, at 12:57 AM, Scott Gray wrote:

> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.

This would be part of a change in our release strategy to be more in line with what ASF likes, but we don't have to discuss this now; whenever I (or anyone else of course) will think that it is a good time for a release then we will discuss the specific situation and the vote will tell what is the general consensus about it; I don't think that it is important to discuss this in general now.

Jacopo

>
> If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
> HotWax Media
> http://www.hotwaxmedia.com
>
>
> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>
>> I would suggest to:
>> 1) release 10.04 before the merge is done
>> 2) merge the code to the trunk, switch to it, fix any possible issue
>> 3) do another release (10.06?)
>>
>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>
>> Jacopo
>>
>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>
>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>
>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>
>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>
>>> -Adrian
>>>
>>>
>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Anil Patel-3
In reply to this post by Scott Gray-2
Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.

Thanks and Regards
Anil Patel
HotWax Media Inc
Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"

On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:

> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>
> Regards
> Scott
>
> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>
>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>
>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>
>> The Example component has been switched over to the new design.
>>
>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>
>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>
>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>
>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>
>> -Adrian
>>
>>
>> Scott Gray wrote:
>>> Question:
>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>> Regards
>>> Scott
>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>
>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>
>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>
>>>> :-)
>>>>
>>>> -Adrian
>>>>
>>>>
>>>> Scott Gray wrote:
>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>> HotWax Media
>>>>> http://www.hotwaxmedia.com
>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>> I would suggest to:
>>>>>> 1) release 10.04 before the merge is done
>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>> 3) do another release (10.06?)
>>>>>>
>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>
>>>>>> Jacopo
>>>>>>
>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>
>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>
>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>
>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.

Regards
Scott

On 8/04/2010, at 12:06 PM, Anil Patel wrote:

> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>
> Thanks and Regards
> Anil Patel
> HotWax Media Inc
> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>
> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>
>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>
>> Regards
>> Scott
>>
>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>
>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>
>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>
>>> The Example component has been switched over to the new design.
>>>
>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>
>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>
>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>
>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>
>>> -Adrian
>>>
>>>
>>> Scott Gray wrote:
>>>> Question:
>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>> Regards
>>>> Scott
>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>
>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>
>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>
>>>>> :-)
>>>>>
>>>>> -Adrian
>>>>>
>>>>>
>>>>> Scott Gray wrote:
>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>> HotWax Media
>>>>>> http://www.hotwaxmedia.com
>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>> I would suggest to:
>>>>>>> 1) release 10.04 before the merge is done
>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>> 3) do another release (10.06?)
>>>>>>>
>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>
>>>>>>> Jacopo
>>>>>>>
>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>
>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>
>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>
>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>
>>>>>>>> -Adrian
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-3
Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
In this way we will have two official current releases:
09.04 Stable Release
10.04 Alpha Release

Intended audiences:
09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release

If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).

Jacopo

On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:

> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>
> Regards
> Scott
>
> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>
>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>
>> Thanks and Regards
>> Anil Patel
>> HotWax Media Inc
>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>
>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>
>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>
>>> Regards
>>> Scott
>>>
>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>
>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>
>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>
>>>> The Example component has been switched over to the new design.
>>>>
>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>
>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>
>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>
>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>
>>>> -Adrian
>>>>
>>>>
>>>> Scott Gray wrote:
>>>>> Question:
>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>> Regards
>>>>> Scott
>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>
>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>
>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>
>>>>>> :-)
>>>>>>
>>>>>> -Adrian
>>>>>>
>>>>>>
>>>>>> Scott Gray wrote:
>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>> HotWax Media
>>>>>>> http://www.hotwaxmedia.com
>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>> I would suggest to:
>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>> 3) do another release (10.06?)
>>>>>>>>
>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>
>>>>>>>> Jacopo
>>>>>>>>
>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>
>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>
>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>
>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>
>>>>>>>>> -Adrian
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-3

On Apr 13, 2010, at 10:19 AM, Jacopo Cappellato wrote:

> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
> In this way we will have two official current releases:
> 09.04 Stable Release
> 10.04 Alpha Release
>
> Intended audiences:
> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>
> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).

From a technical point of view 09.04, 10.04 etc will be svn branches, while 10.04.1, 10.04.2 etc... will be svn tags associated to the branch.

Jacopo

>
> Jacopo
>
> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>
>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>
>> Regards
>> Scott
>>
>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>
>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>
>>> Thanks and Regards
>>> Anil Patel
>>> HotWax Media Inc
>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>
>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>
>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>
>>>> Regards
>>>> Scott
>>>>
>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>
>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>
>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>
>>>>> The Example component has been switched over to the new design.
>>>>>
>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>
>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>
>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>
>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>
>>>>> -Adrian
>>>>>
>>>>>
>>>>> Scott Gray wrote:
>>>>>> Question:
>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>> Regards
>>>>>> Scott
>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>
>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>
>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>
>>>>>>> :-)
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>> Scott Gray wrote:
>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>> HotWax Media
>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>> I would suggest to:
>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>
>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>
>>>>>>>>> Jacopo
>>>>>>>>>
>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>
>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>
>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>
>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>
>>>>>>>>>> -Adrian
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
In reply to this post by Jacopo Cappellato-3
Hi Jacopo,

What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?

Thanks
Scott

On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:

> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
> In this way we will have two official current releases:
> 09.04 Stable Release
> 10.04 Alpha Release
>
> Intended audiences:
> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>
> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>
> Jacopo
>
> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>
>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>
>> Regards
>> Scott
>>
>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>
>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>
>>> Thanks and Regards
>>> Anil Patel
>>> HotWax Media Inc
>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>
>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>
>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>
>>>> Regards
>>>> Scott
>>>>
>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>
>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>
>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>
>>>>> The Example component has been switched over to the new design.
>>>>>
>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>
>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>
>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>
>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>
>>>>> -Adrian
>>>>>
>>>>>
>>>>> Scott Gray wrote:
>>>>>> Question:
>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>> Regards
>>>>>> Scott
>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>
>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>
>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>
>>>>>>> :-)
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>> Scott Gray wrote:
>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>> HotWax Media
>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>> I would suggest to:
>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>
>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>
>>>>>>>>> Jacopo
>>>>>>>>>
>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>
>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>
>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>
>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>
>>>>>>>>>> -Adrian
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>
>>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-3
On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:

> Hi Jacopo,
>
> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?

It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.

Kind regards,

Jacopo

>
> Thanks
> Scott
>
> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>
>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>> In this way we will have two official current releases:
>> 09.04 Stable Release
>> 10.04 Alpha Release
>>
>> Intended audiences:
>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>
>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>
>> Jacopo
>>
>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>
>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>
>>> Regards
>>> Scott
>>>
>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>
>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>
>>>> Thanks and Regards
>>>> Anil Patel
>>>> HotWax Media Inc
>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>
>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>
>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>
>>>>> Regards
>>>>> Scott
>>>>>
>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>
>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>
>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>
>>>>>> The Example component has been switched over to the new design.
>>>>>>
>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>
>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>
>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>
>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>
>>>>>> -Adrian
>>>>>>
>>>>>>
>>>>>> Scott Gray wrote:
>>>>>>> Question:
>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>> Regards
>>>>>>> Scott
>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>
>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>
>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>
>>>>>>>> :-)
>>>>>>>>
>>>>>>>> -Adrian
>>>>>>>>
>>>>>>>>
>>>>>>>> Scott Gray wrote:
>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>> HotWax Media
>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>> I would suggest to:
>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>
>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>
>>>>>>>>>> Jacopo
>>>>>>>>>>
>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>
>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>
>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>
>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>
>>>>>>>>>>> -Adrian
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>
>>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-3

On Apr 13, 2010, at 11:36 AM, Jacopo Cappellato wrote:

> On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:
>
>> Hi Jacopo,
>>
>> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?
>
> It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.
>

But apart from this (important, in my opinion) difference the "alpha" release will be used exactly as we have used the release candidate: to build (with the help of the community) a "stable" release.
Having the ability of distribute it will increase our chances to get more help (tests and bug fixes but also documentation) from the community and the "stable" release will be an effort of the community (instead of the result of backported patches done by committers only).

Jacopo



> Kind regards,
>
> Jacopo
>
>>
>> Thanks
>> Scott
>>
>> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>>
>>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>>> In this way we will have two official current releases:
>>> 09.04 Stable Release
>>> 10.04 Alpha Release
>>>
>>> Intended audiences:
>>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>>
>>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>>
>>> Jacopo
>>>
>>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>>
>>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>>
>>>> Regards
>>>> Scott
>>>>
>>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>>
>>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>>
>>>>> Thanks and Regards
>>>>> Anil Patel
>>>>> HotWax Media Inc
>>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>>
>>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>>
>>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>>
>>>>>> Regards
>>>>>> Scott
>>>>>>
>>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>>
>>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>>
>>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>>
>>>>>>> The Example component has been switched over to the new design.
>>>>>>>
>>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>>
>>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>>
>>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>>
>>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>> Scott Gray wrote:
>>>>>>>> Question:
>>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>>> Regards
>>>>>>>> Scott
>>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>>
>>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>>
>>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>>
>>>>>>>>> :-)
>>>>>>>>>
>>>>>>>>> -Adrian
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Scott Gray wrote:
>>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>>> HotWax Media
>>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>>> I would suggest to:
>>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>>
>>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>>
>>>>>>>>>>> Jacopo
>>>>>>>>>>>
>>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>>
>>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>>
>>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>>
>>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>>
>>>>>>>>>>>> -Adrian
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
In reply to this post by Jacopo Cappellato-3
On 13/04/2010, at 9:36 PM, Jacopo Cappellato wrote:

> On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:
>
>> Hi Jacopo,
>>
>> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?
>
> It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.

Ah okay I see what you mean and that sounds fine to me.  I'm not entirely clear on the version numbering though, 10.04a, 10.04b, 10.04 (this is the stable one), 10.04.1 (post stable bug fix release?)

>
> Kind regards,
>
> Jacopo
>
>>
>> Thanks
>> Scott
>>
>> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>>
>>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>>> In this way we will have two official current releases:
>>> 09.04 Stable Release
>>> 10.04 Alpha Release
>>>
>>> Intended audiences:
>>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>>
>>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>>
>>> Jacopo
>>>
>>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>>
>>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>>
>>>> Regards
>>>> Scott
>>>>
>>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>>
>>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>>
>>>>> Thanks and Regards
>>>>> Anil Patel
>>>>> HotWax Media Inc
>>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>>
>>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>>
>>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>>
>>>>>> Regards
>>>>>> Scott
>>>>>>
>>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>>
>>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>>
>>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>>
>>>>>>> The Example component has been switched over to the new design.
>>>>>>>
>>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>>
>>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>>
>>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>>
>>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>>
>>>>>>> -Adrian
>>>>>>>
>>>>>>>
>>>>>>> Scott Gray wrote:
>>>>>>>> Question:
>>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>>> Regards
>>>>>>>> Scott
>>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>>
>>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>>
>>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>>
>>>>>>>>> :-)
>>>>>>>>>
>>>>>>>>> -Adrian
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Scott Gray wrote:
>>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>>> HotWax Media
>>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>>> I would suggest to:
>>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>>
>>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>>
>>>>>>>>>>> Jacopo
>>>>>>>>>>>
>>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>>
>>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>>
>>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>>
>>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>>
>>>>>>>>>>>> -Adrian
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Scott Gray-2
In reply to this post by Jacopo Cappellato-3
On 13/04/2010, at 9:41 PM, Jacopo Cappellato wrote:

>
> On Apr 13, 2010, at 11:36 AM, Jacopo Cappellato wrote:
>
>> On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:
>>
>>> Hi Jacopo,
>>>
>>> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?
>>
>> It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.
>>
>
> But apart from this (important, in my opinion) difference the "alpha" release will be used exactly as we have used the release candidate: to build (with the help of the community) a "stable" release.
> Having the ability of distribute it will increase our chances to get more help (tests and bug fixes but also documentation) from the community and the "stable" release will be an effort of the community (instead of the result of backported patches done by committers only).
>
> Jacopo
I'm very much in favor of moving away from svn as our core distribution method.  I believe that doing so would encourage service providers (or even the community) to build and offer migration tools and services.

Regards
Scott

>> Kind regards,
>>
>> Jacopo
>>
>>>
>>> Thanks
>>> Scott
>>>
>>> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>>>
>>>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>>>> In this way we will have two official current releases:
>>>> 09.04 Stable Release
>>>> 10.04 Alpha Release
>>>>
>>>> Intended audiences:
>>>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>>>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>>>
>>>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>>>
>>>> Jacopo
>>>>
>>>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>>>
>>>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>>>
>>>>> Regards
>>>>> Scott
>>>>>
>>>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>>>
>>>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>>>
>>>>>> Thanks and Regards
>>>>>> Anil Patel
>>>>>> HotWax Media Inc
>>>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>>>
>>>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>>>
>>>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>>>
>>>>>>> Regards
>>>>>>> Scott
>>>>>>>
>>>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>>>
>>>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>>>
>>>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>>>
>>>>>>>> The Example component has been switched over to the new design.
>>>>>>>>
>>>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>>>
>>>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>>>
>>>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>>>
>>>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>>>
>>>>>>>> -Adrian
>>>>>>>>
>>>>>>>>
>>>>>>>> Scott Gray wrote:
>>>>>>>>> Question:
>>>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>>>> Regards
>>>>>>>>> Scott
>>>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>>>
>>>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>>>
>>>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>>>
>>>>>>>>>> :-)
>>>>>>>>>>
>>>>>>>>>> -Adrian
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Scott Gray wrote:
>>>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>>>> HotWax Media
>>>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>>>> I would suggest to:
>>>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>>>
>>>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>>>
>>>>>>>>>>>> Jacopo
>>>>>>>>>>>>
>>>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>>>
>>>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>>>
>>>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>>>
>>>>>>>>>>>>> -Adrian
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>


smime.p7s (3K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

Jacopo Cappellato-3
In reply to this post by Scott Gray-2

On Apr 13, 2010, at 12:00 PM, Scott Gray wrote:

> On 13/04/2010, at 9:36 PM, Jacopo Cappellato wrote:
>
>> On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:
>>
>>> Hi Jacopo,
>>>
>>> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?
>>
>> It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.
>
> Ah okay I see what you mean and that sounds fine to me.  I'm not entirely clear on the version numbering though, 10.04a, 10.04b, 10.04 (this is the stable one), 10.04.1 (post stable bug fix release?)
>

Numbering is an interesting point because it is difficult to state what is "stable" from what is not; in your example, of course 10.04a is not stable; however what makes 10.04 stable? In fact it is less stable than 10.04.1.
I don't know, if we are concerned about clarifying what we consider stable we could follow the following strategy: adding the prefix "alpha-" to all the releases we feel like should not be considered "stable".
For example:
alpha-10.04.a
alpha-10.04.b
Then when we feel we can consider the release stable:
10.04 (first stable release on 10.04)
10.04.1 (latest current stable release on 10.04)
or even:
stable-10.04
stable-10.04.1

Even if it could be simpler to just start from 10.04.1 since the first alpha release and then continue increasing the suffix:
alpha-10.04.1
alpha-10.04.2
stable-10.04.3
stable-10.04.4

but I understand that this is less appealing (i.e. the "stable" release will start with 10.04.3)

Jacopo


>>
>> Kind regards,
>>
>> Jacopo
>>
>>>
>>> Thanks
>>> Scott
>>>
>>> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>>>
>>>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>>>> In this way we will have two official current releases:
>>>> 09.04 Stable Release
>>>> 10.04 Alpha Release
>>>>
>>>> Intended audiences:
>>>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>>>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>>>
>>>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>>>
>>>> Jacopo
>>>>
>>>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>>>
>>>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>>>
>>>>> Regards
>>>>> Scott
>>>>>
>>>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>>>
>>>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>>>
>>>>>> Thanks and Regards
>>>>>> Anil Patel
>>>>>> HotWax Media Inc
>>>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>>>
>>>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>>>
>>>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>>>
>>>>>>> Regards
>>>>>>> Scott
>>>>>>>
>>>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>>>
>>>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>>>
>>>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>>>
>>>>>>>> The Example component has been switched over to the new design.
>>>>>>>>
>>>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>>>
>>>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>>>
>>>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>>>
>>>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>>>
>>>>>>>> -Adrian
>>>>>>>>
>>>>>>>>
>>>>>>>> Scott Gray wrote:
>>>>>>>>> Question:
>>>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>>>> Regards
>>>>>>>>> Scott
>>>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>>>
>>>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>>>
>>>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>>>
>>>>>>>>>> :-)
>>>>>>>>>>
>>>>>>>>>> -Adrian
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Scott Gray wrote:
>>>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>>>> HotWax Media
>>>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>>>> I would suggest to:
>>>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>>>
>>>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>>>
>>>>>>>>>>>> Jacopo
>>>>>>>>>>>>
>>>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>>>
>>>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>>>
>>>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>>>
>>>>>>>>>>>>> -Adrian
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>

Reply | Threaded
Open this post in threaded view
|

Re: Security Redesign and Release 10.x Branch

BJ Freeman
In reply to this post by Scott Gray-2
alpha on other teams I have been on has also included filling out a feature.
Alpha meant changes to module already implemented for functionality of
the design that was missing.
the numbering was the level of change to the code. the further the
.x.x.x the less of a change to the code.

=========================
BJ Freeman
http://bjfreeman.elance.com
Strategic Power Office with Supplier Automation <http://www.businessesnetwork.com/automation/viewforum.php?f=93>
Specialtymarket.com <http://www.specialtymarket.com/>

Systems Integrator-- Glad to Assist

Chat  Y! messenger: bjfr33man
Linkedin
<http://www.linkedin.com/profile?viewProfile=&key=1237480&locale=en_US&trk=tab_pro>


Jacopo Cappellato sent the following on 4/13/2010 3:21 AM:

> On Apr 13, 2010, at 12:00 PM, Scott Gray wrote:
>
>> On 13/04/2010, at 9:36 PM, Jacopo Cappellato wrote:
>>
>>> On Apr 13, 2010, at 11:33 AM, Scott Gray wrote:
>>>
>>>> Hi Jacopo,
>>>>
>>>> What exactly does it mean to create an "alpha" release, compared to what we have now where we create a release branch?
>>> It fundamentally means that we can distribute it outside of the inner group of contributors because the we can guarantee that it is full compliant with ASF license requirements.
>> Ah okay I see what you mean and that sounds fine to me.  I'm not entirely clear on the version numbering though, 10.04a, 10.04b, 10.04 (this is the stable one), 10.04.1 (post stable bug fix release?)
>>
>
> Numbering is an interesting point because it is difficult to state what is "stable" from what is not; in your example, of course 10.04a is not stable; however what makes 10.04 stable? In fact it is less stable than 10.04.1.
> I don't know, if we are concerned about clarifying what we consider stable we could follow the following strategy: adding the prefix "alpha-" to all the releases we feel like should not be considered "stable".
> For example:
> alpha-10.04.a
> alpha-10.04.b
> Then when we feel we can consider the release stable:
> 10.04 (first stable release on 10.04)
> 10.04.1 (latest current stable release on 10.04)
> or even:
> stable-10.04
> stable-10.04.1
>
> Even if it could be simpler to just start from 10.04.1 since the first alpha release and then continue increasing the suffix:
> alpha-10.04.1
> alpha-10.04.2
> stable-10.04.3
> stable-10.04.4
>
> but I understand that this is less appealing (i.e. the "stable" release will start with 10.04.3)
>
> Jacopo
>
>
>>> Kind regards,
>>>
>>> Jacopo
>>>
>>>> Thanks
>>>> Scott
>>>>
>>>> On 13/04/2010, at 8:19 PM, Jacopo Cappellato wrote:
>>>>
>>>>> Sorry if I am hijacking this thread, but the more I think of it the more I believe we should officially create an "alpha" release 10.04, instead of simply creating a release candidate for 10.04.
>>>>> In this way we will have two official current releases:
>>>>> 09.04 Stable Release
>>>>> 10.04 Alpha Release
>>>>>
>>>>> Intended audiences:
>>>>> 09.04: final users with no interest (or resources) in helping the community to build and maintain stable releases
>>>>> 10.04: users (they could be service providers, end user companies with internal resources or longer term goals etc...) that are willing to help the community to build and maintain a stable release
>>>>>
>>>>> If there will be interest around the 10.04 alpha release, we will get bug fixes that will be part of a future 10.04.1 "stable" (bug fix) release (or a "beta" release), or even 10.04.2,3,4,5 etc... (each of them more stable than the predecessor).
>>>>>
>>>>> Jacopo
>>>>>
>>>>> On Apr 8, 2010, at 8:11 PM, Scott Gray wrote:
>>>>>
>>>>>> Just to be clear though, I am NOT in favor of back-porting large chunks of functionality to the release branch under the guise of bug fixes.
>>>>>>
>>>>>> Regards
>>>>>> Scott
>>>>>>
>>>>>> On 8/04/2010, at 12:06 PM, Anil Patel wrote:
>>>>>>
>>>>>>> Looks like, none who participated in this thread have objections for merging of securitycontext20091231 branch with trunk.
>>>>>>>
>>>>>>> Thanks and Regards
>>>>>>> Anil Patel
>>>>>>> HotWax Media Inc
>>>>>>> Find us on the web at www.hotwaxmedia.com or Google Keyword "ofbiz"
>>>>>>>
>>>>>>> On Apr 7, 2010, at 7:46 PM, Scott Gray wrote:
>>>>>>>
>>>>>>>> Well I don't see any problem with dropping it in right now then.  The real question will be what do people want to be able to backport once the release branch is created.
>>>>>>>>
>>>>>>>> Regards
>>>>>>>> Scott
>>>>>>>>
>>>>>>>> On 7/04/2010, at 5:35 PM, Adrian Crum wrote:
>>>>>>>>
>>>>>>>>> The security redesign implementation itself is mostly finished. There are a few TODOs and they can be found in the BranchReadMe.txt file.
>>>>>>>>>
>>>>>>>>> I recently synchronized the branch with the trunk and there is a remote chance something in the design might have broken in the process. I need to run some tests and review the code to see if that happened.
>>>>>>>>>
>>>>>>>>> The Example component has been switched over to the new design.
>>>>>>>>>
>>>>>>>>> There is a user login called "artifact-user" that demonstrates the new design. That user login is restricted to using the Example component.
>>>>>>>>>
>>>>>>>>> If the branch was merged back to the trunk and the new security design was enabled, the Example component would use the new design and the remaining components would still use the current security design. The two can co-exist.
>>>>>>>>>
>>>>>>>>> I imagine the process after that would be similar to when we introduced the permission checking services - contributors can contribute code that converts parts of the project over to the new security design. Conversion involves removing hard-coded permission checks and creating seed data to grant permission to component artifacts.
>>>>>>>>>
>>>>>>>>> As I mentioned before, switching a component over to the new design can create some unexpected problems. That's because our existing code has security holes in it, and the new design plugs those holes - making parts of the component unreachable. In other words, parts of code that happily allow you to do things you don't have permission to do will start to throw exceptions in the new design.
>>>>>>>>>
>>>>>>>>> -Adrian
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Scott Gray wrote:
>>>>>>>>>> Question:
>>>>>>>>>> What exactly is the current status of the execution branch?  What is it that needs to be done for it to be enabled in the trunk?
>>>>>>>>>> I'm sorry if you feel you've already answered that question but I'm afraid it still isn't entirely clear to me.
>>>>>>>>>> Regards
>>>>>>>>>> Scott
>>>>>>>>>> On 7/04/2010, at 5:14 PM, Adrian Crum wrote:
>>>>>>>>>>> If we wait, then we're waiting for evaluation and testing of the branch. I've done all I can do - the code is written, I suggested we do the merge before the release branch, and I gave my reasons for suggesting it.
>>>>>>>>>>>
>>>>>>>>>>> At this point in time I have stepped out of the discussion (in a positive way) to give others a chance to look at the design and the code and decide for themselves if it should be included. In other words, I don't want to be in a position where I have to convince the community what it should do. If the design and the implementation are good, then there will be no need to convince anyone, right?
>>>>>>>>>>>
>>>>>>>>>>> I'll answer questions about the executioncontext branch, and I'll continue to work on it here and there when I have the time. If the release branch is created without it, then that will be fine with me.
>>>>>>>>>>>
>>>>>>>>>>> :-)
>>>>>>>>>>>
>>>>>>>>>>> -Adrian
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Scott Gray wrote:
>>>>>>>>>>>> Considering we have yet to do an official release after 3.5 years and the lack of user interest in our release branches (partly because we recommend the trunk to everybody), I think it would be a waste of time and effort to create more than one release branch per year.  If we want the security branch in there then lets wait, there is no good reason for us to release this month, it's just an arbitrary date.
>>>>>>>>>>>> HotWax Media
>>>>>>>>>>>> http://www.hotwaxmedia.com
>>>>>>>>>>>> On 7/04/2010, at 12:07 AM, Jacopo Cappellato wrote:
>>>>>>>>>>>>> I would suggest to:
>>>>>>>>>>>>> 1) release 10.04 before the merge is done
>>>>>>>>>>>>> 2) merge the code to the trunk, switch to it, fix any possible issue
>>>>>>>>>>>>> 3) do another release (10.06?)
>>>>>>>>>>>>>
>>>>>>>>>>>>> I know this is not inline with what we currently think a release should be, but this is very inline with what the ASF practices and so I will continue to insist with the release-often practice. :-)
>>>>>>>>>>>>>
>>>>>>>>>>>>> Jacopo
>>>>>>>>>>>>>
>>>>>>>>>>>>> On Apr 4, 2010, at 8:21 PM, Adrian Crum wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>>> I would like to start bringing parts of the executioncontext20091231 branch into the trunk before we create the next release branch. The implementation of the new security design is not finished, but it will be disabled - so everything will still work the same.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> My goal is to allow users of the 10.x release to plan for the forthcoming changes, and maybe have the conversion to the new design completed by the release that follows 10.x.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> I will wait a few days, and if there are no objections I will begin merging the design into the trunk.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> -Adrian
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>>>>>>>>>>>>>>
>
>


1234