You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lenya.apache.org by Thorsten Scherler <th...@wyona.com> on 2006/08/04 17:07:42 UTC

[Rant] Behavior of TRUNK

Hi all,

I need to raise my concerns about the situation we have (again) on the
current trunk.

[Speaking purely as PMC member]
We raised the issue of branching when we will break the branch. The last
time this came up by dropping the extension of files. I thought we
agreed that if we break the trunk (like uuid is doing right now) we will
create a branch and merging back when it is stable and they are clear
instructions about updating homegrown modules to the new API. 

However we just broke the trunk again and now the quota of side effect
questions are raising again. This cost a lot of time for this community.
We cannot afford this ATM!

The discussion about branching gave the impression that we like to be
able to break trunk every now and then, but under the condition of
regular releases. May I remind that number 1 faq is still "when will be
1.4 released".

Either we release before breaking trunk again or we *have to* branch
*before* breaking the trunk. ...or we will never ever release 1.4 at all
because all the extra work is loosing too much community. 

See the revolution branch as a side effect of this situation. 

[Speaking purely as dev]
I cannot spend more hours to update my customers modules code against
the current HEAD. We far too often are breaking the current API that
custom code is not working anymore. 

As professional lenya dev with commit access I am writing custom lenya
related code and meanwhile enhancing the lenya code. When I write a
module it is normally against the current trunk back then. 

The customer reviews it and has some changes. I now have the decission
between keep on using the original revision or updating the module to
use current HEAD. If I choose old revision then I cannot incorporate any
enhancements that HEAD brought or I can do while working on the module. 

If I choose HEAD in the current situation, like I did, I am screwed.
Leaving me with a lot of more days to review all the modules I wrote. I
cannot do that anymore I have no time for it. I need a release to base
my custom modules on. 

We need a release now.

salu2 
-- 
Thorsten Scherler
COO Spain
Wyona Inc.  -  Open Source Content Management  -  Apache Lenya
http://www.wyona.com                   http://lenya.apache.org
thorsten.scherler@wyona.com                thorsten@apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: [Rant] Behavior of TRUNK

Posted by Thorsten Scherler <th...@apache.org>.
El vie, 04-08-2006 a las 13:09 -0400, Doug Chestnut escribió:
> Hi Thorsten,

Thanks Doug for this very constructive response.

> 
> Thorsten Scherler wrote:
> > Hi all,
> > 
> > I need to raise my concerns about the situation we have (again) on the
> > current trunk.
> > 
> > [Speaking purely as PMC member]
> > We raised the issue of branching when we will break the branch. The last
> > time this came up by dropping the extension of files. I thought we
> > agreed that if we break the trunk (like uuid is doing right now) we will
> > create a branch and merging back when it is stable and they are clear
> > instructions about updating homegrown modules to the new API. 
> > 
> > However we just broke the trunk again and now the quota of side effect
> > questions are raising again. This cost a lot of time for this community.
> > We cannot afford this ATM!
> > 
> > The discussion about branching gave the impression that we like to be
> > able to break trunk every now and then, but under the condition of
> > regular releases. May I remind that number 1 faq is still "when will be
> > 1.4 released".
> > 
> > Either we release before breaking trunk again or we *have to* branch
> > *before* breaking the trunk. ...or we will never ever release 1.4 at all
> > because all the extra work is loosing too much community. 
> I see your point here, but I think the urgency of a 1.4 release has 
> caused the trunk to be broken now.  I believe that UUID's are the last 
> agreed upon hurdle to a 1.4 release (after testing and bugfixes).  UUIDs 
> would break backwards compatibility and need to be in place for a release.
> 

Yes, I understand and agree why we broke trunk, but I only wished that
this would have happen in a branch. Like you say the UUID integration is
the last feature that needs to go into the release, because of
compatibility reasons. 

Meaning for me "we will release before breaking trunk again", or ? ;)


> > 
> > See the revolution branch as a side effect of this situation. 
> > 
> > [Speaking purely as dev]
> > I cannot spend more hours to update my customers modules code against
> > the current HEAD. We far too often are breaking the current API that
> > custom code is not working anymore. 
> I feel your pain Thorsten, but also sympathize with the developer trying 
> to incorporate all of the required functionality for a 1.4 release.  It 
> is an astonishing task to move 1.4 to uuid support.

Yes and I am very grateful for the work the developer are doing. I am
only wishing that some of the task would be developed in a branch.
 

> 
> This situation has come about by us using the trunk for production.  The 
> trunk is far better (my opinion) than 1.2 and probably should have been 
> released w/o uuid,jcr,observation,etc last year at this time (as was the 
> schedule).  

Yes, it is a tricky situation this days to write components for trunk.
We really need to get started releasing on a regular basis. This way we
can calculate for which version of lenya one can build ones component
for.

> There is always 1.6 which will hopefully be far better than 1.4.

I would like that people have a look as well in the revolution branch
before we start with 1.6. 1.6 should incorporate the best of all worlds
of lenya.

> > 
> > As professional lenya dev with commit access I am writing custom lenya
> > related code and meanwhile enhancing the lenya code. When I write a
> > module it is normally against the current trunk back then. 
> > 
> > The customer reviews it and has some changes. I now have the decission
> > between keep on using the original revision or updating the module to
> > use current HEAD. If I choose old revision then I cannot incorporate any
> > enhancements that HEAD brought or I can do while working on the module. 
> > 
> > If I choose HEAD in the current situation, like I did, I am screwed.
> > Leaving me with a lot of more days to review all the modules I wrote. I
> > cannot do that anymore I have no time for it. I need a release to base
> > my custom modules on. 
> > 
> > We need a release now.
> I think we are all in agreement that a release is required. ;)
> 

;)


salu2

> --Doug
> > 
> > salu2 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
> For additional commands, e-mail: dev-help@lenya.apache.org
> 
-- 
thorsten

"Together we stand, divided we fall!" 
Hey you (Pink Floyd)


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org


Re: [Rant] Behavior of TRUNK

Posted by Doug Chestnut <dh...@virginia.edu>.
Hi Thorsten,

Thorsten Scherler wrote:
> Hi all,
> 
> I need to raise my concerns about the situation we have (again) on the
> current trunk.
> 
> [Speaking purely as PMC member]
> We raised the issue of branching when we will break the branch. The last
> time this came up by dropping the extension of files. I thought we
> agreed that if we break the trunk (like uuid is doing right now) we will
> create a branch and merging back when it is stable and they are clear
> instructions about updating homegrown modules to the new API. 
> 
> However we just broke the trunk again and now the quota of side effect
> questions are raising again. This cost a lot of time for this community.
> We cannot afford this ATM!
> 
> The discussion about branching gave the impression that we like to be
> able to break trunk every now and then, but under the condition of
> regular releases. May I remind that number 1 faq is still "when will be
> 1.4 released".
> 
> Either we release before breaking trunk again or we *have to* branch
> *before* breaking the trunk. ...or we will never ever release 1.4 at all
> because all the extra work is loosing too much community. 
I see your point here, but I think the urgency of a 1.4 release has 
caused the trunk to be broken now.  I believe that UUID's are the last 
agreed upon hurdle to a 1.4 release (after testing and bugfixes).  UUIDs 
would break backwards compatibility and need to be in place for a release.

> 
> See the revolution branch as a side effect of this situation. 
> 
> [Speaking purely as dev]
> I cannot spend more hours to update my customers modules code against
> the current HEAD. We far too often are breaking the current API that
> custom code is not working anymore. 
I feel your pain Thorsten, but also sympathize with the developer trying 
to incorporate all of the required functionality for a 1.4 release.  It 
is an astonishing task to move 1.4 to uuid support.

This situation has come about by us using the trunk for production.  The 
trunk is far better (my opinion) than 1.2 and probably should have been 
released w/o uuid,jcr,observation,etc last year at this time (as was the 
schedule).  There is always 1.6 which will hopefully be far better than 1.4.
> 
> As professional lenya dev with commit access I am writing custom lenya
> related code and meanwhile enhancing the lenya code. When I write a
> module it is normally against the current trunk back then. 
> 
> The customer reviews it and has some changes. I now have the decission
> between keep on using the original revision or updating the module to
> use current HEAD. If I choose old revision then I cannot incorporate any
> enhancements that HEAD brought or I can do while working on the module. 
> 
> If I choose HEAD in the current situation, like I did, I am screwed.
> Leaving me with a lot of more days to review all the modules I wrote. I
> cannot do that anymore I have no time for it. I need a release to base
> my custom modules on. 
> 
> We need a release now.
I think we are all in agreement that a release is required. ;)

--Doug
> 
> salu2 

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lenya.apache.org
For additional commands, e-mail: dev-help@lenya.apache.org