Result: DVCS options for Plone Core

[Hanno Schlichting-4]

On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting <[hidden email]> wrote:
> I'll compile a summary and post follow up on this list in a week
> (which should give everyone enough time to recover from Easter).

I'm amazed how clear the vote result is! We don't have to wait any
longer to draw a conclusion:

We want to move to Github.

If that for some reason cannot possible be made to work, we stay with
Subversion. All other options are straight out.

What we need now is a detailed plan and do the required work. I'll
volunteer to help organizing this. If you want to help out in this,
please ping me. I'll try to start a more detailed proposal and sketch
out some tasks over Easter.

Hanno

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Ross Patterson-2]

Hanno Schlichting <[hidden email]>
writes:

> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting <[hidden email]> wrote:
>> I'll compile a summary and post follow up on this list in a week
>> (which should give everyone enough time to recover from Easter).
>
> I'm amazed how clear the vote result is! We don't have to wait any
> longer to draw a conclusion:
>
> We want to move to Github.
>
> If that for some reason cannot possible be made to work, we stay with
> Subversion. All other options are straight out.
>
> What we need now is a detailed plan and do the required work. I'll
> volunteer to help organizing this. If you want to help out in this,
> please ping me. I'll try to start a more detailed proposal and sketch
> out some tasks over Easter.

What about issue tracker/VCS integration?  Last I recall, github still
doesn't have a good issue tracker.  We have a lot invested in our trac
setup.  What are the options there?

Ross


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Hanno Schlichting-4]

On Fri, Apr 22, 2011 at 9:20 PM, Ross Patterson <[hidden email]> wrote:
> What about issue tracker/VCS integration?  Last I recall, github still
> doesn't have a good issue tracker.  We have a lot invested in our trac
> setup.  What are the options there?

I'll add this to the list. I think the github Trac plugin [1] is good
enough for us, but it needs a bit more investigation.

Hanno

[1] https://github.com/davglass/github-trac/

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Elizabeth Leddy-2]

On Fri, Apr 22, 2011 at 12:25 PM, Hanno Schlichting <[hidden email]> wrote:

On Fri, Apr 22, 2011 at 9:20 PM, Ross Patterson <[hidden email]> wrote:
> What about issue tracker/VCS integration?  Last I recall, github still
> doesn't have a good issue tracker.  We have a lot invested in our trac
> setup.  What are the options there?

I'll add this to the list. I think the github Trac plugin [1] is good
enough for us, but it needs a bit more investigation.

Github updated its issue tracker and it actually is much better but I don't think logistically we could/should abandon trac. FWIW, The current git integration in trac is "OK". There are commit hooks to close tickets in trac from git commit messages but I wouldn't even try to reflect the repo from git to trac - its loaded with memory issues and rarely stays in sync.

liz

 

Hanno

[1] https://github.com/davglass/github-trac/

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Ross Patterson-2]

Hanno Schlichting <[hidden email]>
writes:

> On Fri, Apr 22, 2011 at 9:20 PM, Ross Patterson <[hidden email]> wrote:
>> What about issue tracker/VCS integration?  Last I recall, github still
>> doesn't have a good issue tracker.  We have a lot invested in our trac
>> setup.  What are the options there?
>
> I'll add this to the list. I think the github Trac plugin [1] is good
> enough for us, but it needs a bit more investigation.

Awesome, that was my last real qualm.  There will, of course, be screams
of bloody murder, but that's never stopped me, even when their my own.
:-)

Ross


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[sdouche]

On Fri, Apr 22, 2011 at 21:20, Ross Patterson <[hidden email]> wrote:

> What about issue tracker/VCS integration?  Last I recall, github still
> doesn't have a good issue tracker.  We have a lot invested in our trac
> setup.  What are the options there?

Hi Ross,
Have you seen Issue v2.0[1]? What functionalities are missing in your opinion?


[1] https://github.com/blog/831-issues-2-0-the-next-generation


--
Sebastien Douche <[hidden email]>
Twitter: @sdouche (agile, lean, python, git, open source)

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[sdouche]

On Fri, Apr 22, 2011 at 21:08, Hanno Schlichting <[hidden email]> wrote:
> I'm amazed how clear the vote result is! We don't have to wait any
> longer to draw a conclusion:
>
> We want to move to Github.

Good news, GitHub is the best (at least for now) social coding platform.

> What we need now is a detailed plan and do the required work. I'll
> volunteer to help organizing this. If you want to help out in this,
> please ping me. I'll try to start a more detailed proposal and sketch
> out some tasks over Easter.

Cool.



--
Sebastien Douche <[hidden email]>
Twitter: @sdouche (agile, lean, python, git, open source)

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Ross Patterson-2]

Sebastien Douche <[hidden email]>
writes:

> On Fri, Apr 22, 2011 at 21:20, Ross Patterson <[hidden email]> wrote:
>
>> What about issue tracker/VCS integration?  Last I recall, github still
>> doesn't have a good issue tracker.  We have a lot invested in our trac
>> setup.  What are the options there?
>
> Hi Ross,
> Have you seen Issue v2.0[1]? What functionalities are missing in your opinion?
>
>
> [1] https://github.com/blog/831-issues-2-0-the-next-generation

Did you see the "Last I recall" bit?  :-)  Good to see they've gotten
better.

Ross


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[sdouche]

On Sat, Apr 23, 2011 at 03:05, Ross Patterson <[hidden email]> wrote:
>> Have you seen Issue v2.0[1]? What functionalities are missing in your opinion?
>>
>> [1] https://github.com/blog/831-issues-2-0-the-next-generation
>
> Did you see the "Last I recall" bit?  :-)  Good to see they've gotten
> better.

I see it :). But I would to know if for you (Plone core devs), the new
Issue tracker is "usable".



--
Sebastien Douche <[hidden email]>
Twitter: @sdouche (agile, lean, python, git, open source)

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[aclark]

On 4/22/11 3:08 PM, Hanno Schlichting wrote:

> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting<[hidden email]>  wrote:
>> I'll compile a summary and post follow up on this list in a week
>> (which should give everyone enough time to recover from Easter).
>
> I'm amazed how clear the vote result is! We don't have to wait any
> longer to draw a conclusion:
>
> We want to move to Github.
>
> If that for some reason cannot possible be made to work, we stay with
> Subversion. All other options are straight out.
>
> What we need now is a detailed plan and do the required work. I'll
> volunteer to help organizing this. If you want to help out in this,
> please ping me. I'll try to start a more detailed proposal and sketch
> out some tasks over Easter.

Yay! Exciting.

Coincidentally, I just moved my professional client work there and
experienced a few (minor) gotchas FWIW:

* The "Importing from Subversion? Click here" button didn't work. I
waited upwards of 15 min to get queued, then gave up. I assume due to
overload and/or high demand.

* The "Importing from Subversion? Click here" button only handles public
repos (or doesn't handle them, as was the case for me. See previous
gotcha.)

* svn2git (the top recommended choice here:
http://help.github.com/svn-importing/) will not import the entire
contents of svn.plone.org/svn/plone in any meaningful way, you have to
make some choices first:

   * In general, it is easy to migrate repos with
     branches/tags/trunk using a command like:

     (inside an empty dir)
     $ svn2git http://svn.plone.org/svn/plone/Products.CMFPlone/\
     --no-minimize-url

   * If you want to migrate a repo like
     svn.plone.org/svn/plone/CMFPlone (i.e. a repo with branches
     and tags, but no trunk) you may need to tell it explicity:

     $ svn2git http://svn.plone.org/svn/plone/CMFPlone/ --notrunk

   * If you really do want to migrate svn.plone.org/svn/plone into a
     single git repo, do this:

     $ svn2git http://svn.plone.org/svn/plone/CMFPlone/ --rootistrunk

   * It will take a while to process all revisions.

   * Don't forget to "push --all" to get all the branches to the remote
     (IIRC, that is all you have to do).

All of these are fairly trivial concerns, AFAICT. And in less than the
time it took to write this mail, I just ran a quick import test on
Products.CMFPlone to the github collective:

- https://github.com/collective/Products.CMFPlone

One thing I didn't do was pay any attention to svn authors, which I
assume we will want to do if/when a migration is done "for real".

Anyway, due to the aforementioned constraints, I ended up importing my
entire svn repo to a single private repo called "import". Then I
(re-)imported active projects only to their own repo and added the
respective developers. And of course I moved my public repo:

- https://github.com/ACLARKNET/public (as one repo, sans branches)

Also, the Foundation will need to pay the bill, and GitHub is expensive
I'm told[1]. The current repo is 2.1GB (du -hs on
deus:/srv/svn.plone.org/svn/plone) and the nearest plans are:

- Bronze ($25/month) 10 2.40GB
- Silver ($50/month) 20 6.00GB

Assuming we want some breathing room and choose 6.00GB (with 20 private
repos which we don't really need except for the foundation repo) that's
only $600/year which seems quite reasonable.


HTH,




Alex




[1] Rok Garbas told me this, I didn't shop around because I like using
GitHub so much…


>
> Hanno
>
> ------------------------------------------------------------------------------
> Fulfilling the Lean Software Promise
> Lean software platforms are now widely adopted and the benefits have been
> demonstrated beyond question. Learn why your peers are replacing JEE
> containers with lightweight application servers - and what you can gain
> from the move. http://p.sf.net/sfu/vmware-sfemails


--
Alex Clark · http://aclark.net/training


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Balazs Ree-2]

On 2011-04-22 21:08:45 +0200, Hanno Schlichting said:

> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting
> <[hidden email]> wrote:
>> I'll compile a summary and post follow up on this list in a week
>> (which should give everyone enough time to recover from Easter).
>
> I'm amazed how clear the vote result is! We don't have to wait any
> longer to draw a conclusion:
>
> We want to move to Github.

The poll was kind of quick. I just read the news now when you announced
that you have opened it two days ago, and I see the poll is already
closed.

The results are what I expected, and my vote would not have changed it
much - but still I find important to phrase my voice - not for the
first, but probably for the last time about this topic.

I started to use DVCS around 2005. I decided to use Bazaar and I used
it quite extensively since then. Mostly for private projects, and for
some community development as well. I have been very satisfied with it.

Now, thanks to the advent of github, I have to use git as well. And
every time I use git, I cry back for my good old bazaar.

Putting it straight: all major DVCS's provide about the same
functionality. All can be learned by someone who wants to learn them.
But there are differences in ease of use and simplicity. These also
depend on personal taste of course. Since I use both Bazaar and Git, I
am in the position to compare them - and, I personally feel that git
has zero benefit over bzr, on the other hand I find it's annoyingly
more complex to use it.

You can however probably have a negative criticism about any other DVCS
provided that you are used to work with a particular one first. And the
"ease of use" argument in the end gets overwhelmed by the "ease of
share" argument.

The popularity and "ease of share" argument definitely favors for
moving to github. This is why Plone - also considering the poll result
- is predestined to choose this system. I feel however that, although
by choosing github Plone goes for the most popular solution, it does
not necessary move to the best usable solution.

--
Balazs Ree, Greenfinity, LLC



------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Wichert Akkerman]

On 4/23/11 04:01 , Sebastien Douche wrote:

> On Sat, Apr 23, 2011 at 03:05, Ross Patterson<[hidden email]>  wrote:
>>> Have you seen Issue v2.0[1]? What functionalities are missing in your opinion?
>>>
>>> [1] https://github.com/blog/831-issues-2-0-the-next-generation
>>
>> Did you see the "Last I recall" bit?  :-)  Good to see they've gotten
>> better.
>
> I see it :). But I would to know if for you (Plone core devs), the new
> Issue tracker is "usable".

The missing support for attachments to issues is a dealbreaker imho.

Wichert.

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Ed Manlove-2]

> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting <[hidden email]> wrote:
>> I'll compile a summary and post follow up on this list in a week
>> (which should give everyone enough time to recover from Easter).
>
> I'm amazed how clear the vote result is! We don't have to wait any
> longer to draw a conclusion:
>
> We want to move to Github.
>
> If that for some reason cannot possible be made to work, we stay with
> Subversion. All other options are straight out.
>
> What we need now is a detailed plan and do the required work. I'll
> volunteer to help organizing this. If you want to help out in this,
> please ping me. I'll try to start a more detailed proposal and sketch
> out some tasks over Easter.
>
> Hanno
>

I though the poll was going to remain open till AFTER Easter.  So add

Host our own Git         -1
Host our own Mercurial   -1
Host our own Bazaar      -1
Move to Github           -1
Move to Bitbucket        -1
Move to Launchpad        -1
Keep our own Subversion  +1


for what it is worth.  My issue (other that the fact I don't like
change) is that no one except Liz has given a reason for change, as far
as I can see.  Liz reasons are good ones; social interaction of
developers and the "cool" factor which I don't downplay and has it
merits.  My response to her reasons would focus around how is our Plone
community not working?  This applies to both "social interaction" factor
and "cool" factor.  What problems does she, or anyone else, see with our
community that exhibits broken communication and/or is not helping to
attract new people to Plone?  How we operate on a daily basis has a lot
more to do with the simple repository we use.

I am also confused by, for example the past writings of Martin,
http://www.martinaspeli.net/articles/developer-centric-version-control-considered-harmful,
  and his recent vote.  Are you recanting this opinion of github?

So let me be the ass or the clog in the wheel and ask...

For what reason does each of us see as a reason to change?

Ed

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Dylan Jay]

Better merging, quick local branching, and offline use are main draw
cards of dvcs. But you can search the web plenty of discussion on
this. I don't think there is much disagreement that dvcs is better,
but just a question of when, which one and which repo.

Dylan Jay
Technical solution manager
PretaWeb 99552830

On 23/04/2011, at 1:37 PM, Ed Manlove <[hidden email]> wrote:

>
>> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting <[hidden email]> wrote:
>>> I'll compile a summary and post follow up on this list in a week
>>> (which should give everyone enough time to recover from Easter).
>>
>> I'm amazed how clear the vote result is! We don't have to wait any
>> longer to draw a conclusion:
>>
>> We want to move to Github.
>>
>> If that for some reason cannot possible be made to work, we stay with
>> Subversion. All other options are straight out.
>>
>> What we need now is a detailed plan and do the required work. I'll
>> volunteer to help organizing this. If you want to help out in this,
>> please ping me. I'll try to start a more detailed proposal and sketch
>> out some tasks over Easter.
>>
>> Hanno
>>
>
> I though the poll was going to remain open till AFTER Easter.  So add
>
> Host our own Git         -1
> Host our own Mercurial   -1
> Host our own Bazaar      -1
> Move to Github           -1
> Move to Bitbucket        -1
> Move to Launchpad        -1
> Keep our own Subversion  +1
>
>
> for what it is worth.  My issue (other that the fact I don't like
> change) is that no one except Liz has given a reason for change, as far
> as I can see.  Liz reasons are good ones; social interaction of
> developers and the "cool" factor which I don't downplay and has it
> merits.  My response to her reasons would focus around how is our Plone
> community not working?  This applies to both "social interaction" factor
> and "cool" factor.  What problems does she, or anyone else, see with our
> community that exhibits broken communication and/or is not helping to
> attract new people to Plone?  How we operate on a daily basis has a lot
> more to do with the simple repository we use.
>
> I am also confused by, for example the past writings of Martin,
> http://www.martinaspeli.net/articles/developer-centric-version-control-considered-harmful,
>  and his recent vote.  Are you recanting this opinion of github?
>
> So let me be the ass or the clog in the wheel and ask...
>
> For what reason does each of us see as a reason to change?
>
> Ed
>
> ------------------------------------------------------------------------------
> Fulfilling the Lean Software Promise
> Lean software platforms are now widely adopted and the benefits have been
> demonstrated beyond question. Learn why your peers are replacing JEE
> containers with lightweight application servers - and what you can gain
> from the move. http://p.sf.net/sfu/vmware-sfemails
> _______________________________________________
> Plone-developers mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/plone-developers

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Wouter Vanden Hove-2]

Ed Manlove wrote:

> So let me be the ass or the clog in the wheel and ask...
> For what reason does each of us see as a reason to change?

I would like to add another question: why is this not a PLIP?

Is this a less important decision than let's say have "Improved search
results" in Plone 4.2?


--
Greets,
WouterVH


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[aclark]

On 4/23/11 7:36 AM, Ed Manlove wrote:

>
>> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting<[hidden email]>  wrote:
>>> I'll compile a summary and post follow up on this list in a week
>>> (which should give everyone enough time to recover from Easter).
>>
>> I'm amazed how clear the vote result is! We don't have to wait any
>> longer to draw a conclusion:
>>
>> We want to move to Github.
>>
>> If that for some reason cannot possible be made to work, we stay with
>> Subversion. All other options are straight out.
>>
>> What we need now is a detailed plan and do the required work. I'll
>> volunteer to help organizing this. If you want to help out in this,
>> please ping me. I'll try to start a more detailed proposal and sketch
>> out some tasks over Easter.
>>
>> Hanno
>>
>
> I though the poll was going to remain open till AFTER Easter.  So add
>
> Host our own Git         -1
> Host our own Mercurial   -1
> Host our own Bazaar      -1
> Move to Github           -1
> Move to Bitbucket        -1
> Move to Launchpad        -1
> Keep our own Subversion  +1
>
>
> for what it is worth.  My issue (other that the fact I don't like
> change) is that no one except Liz has given a reason for change, as far
> as I can see.  Liz reasons are good ones; social interaction of
> developers and the "cool" factor which I don't downplay and has it
> merits.  My response to her reasons would focus around how is our Plone
> community not working?  This applies to both "social interaction" factor
> and "cool" factor.  What problems does she, or anyone else, see with our
> community that exhibits broken communication and/or is not helping to
> attract new people to Plone?  How we operate on a daily basis has a lot
> more to do with the simple repository we use.
>
> I am also confused by, for example the past writings of Martin,
> http://www.martinaspeli.net/articles/developer-centric-version-control-considered-harmful,
>    and his recent vote.  Are you recanting this opinion of github?
>
> So let me be the ass or the clog in the wheel and ask...
>
> For what reason does each of us see as a reason to change?

One reason is the "hosted vs self-managed" issue. I have every
confidence that volunteers can do a good job at maintaining critical
plone.org resources, but every so often you have to take a hard look at
your infrastructure and re-evaluate what is best done internally vs.
what is best outsourced.

In this case, we want Plone to continue to grow and thrive over the
course of the next 5 years at least, and github provides an amazing
growth opportunity (through the use of their tools to provide social
interaction and collaboration, we can expect a faster rate of development).

I would still recommend that the community rely on OSU and the new svn
server that is being installed there. Primarily because I don't expect
we'll move the "svn collective" anywhere yet, if ever, just the Plone core.


Alex



>
> Ed
>
> ------------------------------------------------------------------------------
> Fulfilling the Lean Software Promise
> Lean software platforms are now widely adopted and the benefits have been
> demonstrated beyond question. Learn why your peers are replacing JEE
> containers with lightweight application servers - and what you can gain
> from the move. http://p.sf.net/sfu/vmware-sfemails


--
Alex Clark · http://aclark.net/training


------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Ed Manlove-2]

Alex,

There are three items I want to address for which I will do interspersed
below as compared to three separate emails.  Please anyone feel free to
break out any of these to focus on if you like.

Alex Clark wrote:

> On 4/23/11 7:36 AM, Ed Manlove wrote:
>  
>>
>> So let me be the ass or the clog in the wheel and ask...
>>
>> For what reason does each of us see as a reason to change?
>>    
>
> One reason is the "hosted vs self-managed" issue. I have every
> confidence that volunteers can do a good job at maintaining critical
> plone.org resources, but every so often you have to take a hard look at
> your infrastructure and re-evaluate what is best done internally vs.
> what is best outsourced.
>  

Ignore for a moment the social tools and extra features provided by a
hosted solution, should we not go with a hosted svn solution instead of
a hosted git solution? What benefits or disadvantages does git have over
svn?  I know we have discussed  some technical administration issues and
one or two operational differences but are there any roadblocks with svn
that git provides a solution to?
> In this case, we want Plone to continue to grow and thrive over the
> course of the next 5 years at least, and github provides an amazing
> growth opportunity (through the use of their tools to provide social
> interaction and collaboration, we can expect a faster rate of development).
>  
Not saying that they will not but what functionality do you see within
their social tools that we currently don't have and will guarantee a
faster rate of development?
> I would still recommend that the community rely on OSU and the new svn
> server that is being installed there. Primarily because I don't expect
> we'll move the "svn collective" anywhere yet, if ever, just the Plone core.
>
>  
Won't having a part svn, part git source greater complicate development?
or buildout? How will non-technical users be affected by this change?
> Alex
>
>  

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[sdouche]

On Sat, Apr 23, 2011 at 04:30, Alex Clark <[hidden email]> wrote:
> Also, the Foundation will need to pay the bill, and GitHub is expensive
> I'm told[1].

Why do you want pay for public repositories?

> The current repo is 2.1GB

Git's repository is small when compared to SVN (especially with many branches) .



--
Sebastien Douche <[hidden email]>
Twitter: @sdouche (agile, lean, python, git, open source)

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Steve McMahon]

Wyn raises some excellent points in his note. I just wanted to address the financial point.

If there is strong agreement among core developers on a direction, I can assure everyone that the PF will find the financial resources to make it happen. That's what it's for. So, at this point, I'm more interested in making sure that we really think through the issues of control, security, flexibility, backup, and intellectual property rights.

On Sat, Apr 23, 2011 at 8:57 AM, w.p.w <[hidden email]> wrote:

...

If we change to GitHUB how do we pay for it ? PF depends on fund raising and other small incomes, it is a charity, do we start charging a nominal amount for access to the core and collective ? insist that only Plone core is hosted by PF and the whole of the collective must move to there own public repos ?

...

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers

[Martin Aspeli-3]

Hi

On 23 Apr 2011, at 12:36, Ed Manlove <[hidden email]> wrote:

>
>> On Thu, Apr 21, 2011 at 11:12 PM, Hanno Schlichting <[hidden email]> wrote:
>>> I'll compile a summary and post follow up on this list in a week
>>> (which should give everyone enough time to recover from Easter).
>>
>> I'm amazed how clear the vote result is! We don't have to wait any
>> longer to draw a conclusion:
>>
>> We want to move to Github.
>>
>> If that for some reason cannot possible be made to work, we stay with
>> Subversion. All other options are straight out.
>>
>> What we need now is a detailed plan and do the required work. I'll
>> volunteer to help organizing this. If you want to help out in this,
>> please ping me. I'll try to start a more detailed proposal and sketch
>> out some tasks over Easter.
>>
>> Hanno
>>
>
> I though the poll was going to remain open till AFTER Easter.  So add
>
> Host our own Git         -1
> Host our own Mercurial   -1
> Host our own Bazaar      -1
> Move to Github           -1
> Move to Bitbucket        -1
> Move to Launchpad        -1
> Keep our own Subversion  +1
>
>
> for what it is worth.  My issue (other that the fact I don't like
> change) is that no one except Liz has given a reason for change, as far
> as I can see.  Liz reasons are good ones; social interaction of
> developers and the "cool" factor which I don't downplay and has it
> merits.  My response to her reasons would focus around how is our Plone
> community not working?  This applies to both "social interaction" factor
> and "cool" factor.  What problems does she, or anyone else, see with our
> community that exhibits broken communication and/or is not helping to
> attract new people to Plone?  How we operate on a daily basis has a lot
> more to do with the simple repository we use.
>
> I am also confused by, for example the past writings of Martin,
> http://www.martinaspeli.net/articles/developer-centric-version-control-considered-harmful,
>  and his recent vote.  Are you recanting this opinion of github?

With shared organisational hosting, the problem is smaller. I still have some reservation about the intellectual complexity of dvcs, but my personal vote is based on what I prefer and I se the attraction of github. Git without github is less interesting to me.

Note that I'm still -1 on moving the collective to github, mainly because I fear putting people off, but it does seem a bit like the writing is on the wall and a lot of people are preferring git.

In the end,the vcs is just a tool. I think it's important not to get too anxious over the choice. We'll be checking in python regardless.

Martin

> So let me be the ass or the clog in the wheel and ask...
>
> For what reason does each of us see as a reason to change?
>
> Ed
>
> ------------------------------------------------------------------------------
> Fulfilling the Lean Software Promise
> Lean software platforms are now widely adopted and the benefits have been
> demonstrated beyond question. Learn why your peers are replacing JEE
> containers with lightweight application servers - and what you can gain
> from the move. http://p.sf.net/sfu/vmware-sfemails
> _______________________________________________
> Plone-developers mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/plone-developers

------------------------------------------------------------------------------
Fulfilling the Lean Software Promise
Lean software platforms are now widely adopted and the benefits have been
demonstrated beyond question. Learn why your peers are replacing JEE
containers with lightweight application servers - and what you can gain
from the move. http://p.sf.net/sfu/vmware-sfemails
_______________________________________________
Plone-developers mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/plone-developers