2 for October? Re: [triangle-zpug] September meeting

Chris Calloway cbc at unc.edu
Tue Sep 27 16:01:33 CEST 2005


T. Bryan wrote:
> It might make the most sense just to cancel the meeting.

So, here's the deal. The meeting is "canceled." That is, most of the
regulars aren't showing up. I'll be at Phillips 328 until 7:30 to see if
anybody shows up. If someone does, well, we can chat. But we'll put off
the main PloneCon recap until October, which might be better, seeing as
how none of the PloneCon presentations have been posted yet, as usual.
So nothing of great import/decision making will occur this evening, even
if anyone shows up. If anyone does show up, I'm going to suggest 
adjourning to a local pub and having informal discussion.

The next meeting should fall on Tuesday, October 25. Unfortunately, that
falls in the middle of an ocean tech conference in Baltimore at which I
have some presentation coordination responsibilities (Ben are you going 
to OOSTech?). So if the October meeting is at UNC, someone else will 
have to take the lead in hosting it, or it can follow normal rotation to 
NCSU. Takers?

One thing I really need to report to you about PloneCon, however, is the
GoldEgg Initiative:

http://www.goldeggstack.org/

GoldEgg is both a funding initiative and project guidance. It is seeking
to coordinate the Plone "stack" development. That is, time releases of
Plone with releases of CMF and Zope and maybe even Python. There is to
be a coordinated release of Plone, CMF, and Zope every six months hence.
The idea is that if Plone needs certain functions in the CMF, then
coordinate to put them in the CMF instead of in Plone. If Zope3 is
bringing interfaces, views, and events to the table, then use those in
the CMF instead of reinventing the wheel. If there is Zope3 technology
that can be backported to Zope2, whether through the Five project or
through the CMFonFive project, or through CMF2 development, then let's
coordinate that. As well, making this happen in open source requires
money. In order to get CMF2 out the door, maybe some key developers need
funding outside of their usual client business. Cignex has put $50K into
GoldEgg One, which is funding three sprints for a November minor dot
release of Zope2/Zope3/CMF/Plone. GoldEgg Two, of which TriZPUG may be a
part, needs $10K of funding.

Now, when I say funding, I mean to say *raising* funds, not going out
with a tin cup and asking for funds. That is, ask not what Plone can do
for you, ask what you can do for Plone. In the process, you can stay
right on the leading edge of Zope/CMF/Plone development and get access
to the very top developers like Jim Fulton and Tres Seaver at your sprints.

The reason for all this is that many big enterprises are wanting to bet
on Plone and Zope3. When I say enterprises, I may not necessarily mean
for-profit businesses, either. I may mean, the UN wants to deploy 1200
websites on Plone, but needs some quality assurance and roadmap
assurance in doing so. They can get that by funding GoldEgg.

I want to raise awareness of this initiative, so that if your enterprise
needs such assurances, you'll know how to advise your enterprise on how
to proceed.

For instance, PloneBoard was just revived and ported to Plone 2.1 with
the use of Five and Zope3 interfaces and views. It's just a BBS at this
point. No email in or out. In order to do email with PloneBoard, we need
proper portal_subscriptions (i.e., not the portal_subscriptions tool in
CMFBoard, which tries to implement a whole new framework in Plone
instead of the CMF). Portal_subscriptions properly belongs in the CMF,
not Plone. So there is some CMF work to do to get email subscriptions
into PloneBoard. In order for the CMF to properly implement
portal_subscriptions, we need to make use of Zope3 Events. GoldEgg One
is polishing Zope3 Events and bringing Zope3 Events into CMF2, so
portal_subscriptions can work properly. So you can see how this all
works together. The end customer simply know that they want PloneBoard
to have email subscriptions. They can fund GoldEgg, which will insure
that the development is coordinated properly at the correct layer in the
stack, and quality assurance is the result (and if you've ever tried to
use CMFBoard, then you know how badly this quality assurance is needed).
And what you can end up with is a generic portal_subscriptions tool that
can work for all kinds of content and not just PloneBoard.

If you are in a large enterprise that is using Plone, you need to be
figuring out how your future needs are going to be met by open source,
and thereby how to fund open source. I know at least for UNC, the Plone
people at UNC need to be putting heads together to see how this fits
into the Renaissance Computing Initiative (with Duke and NCSU) and if
the proper eyeballs can be directed to take a look at this. Which is all
goodness as far as I'm concerned because I'm tired of there not being
anything at UNC like the Oregon State University Open Source Labs while
UNC funds the largest open source repository in the world. It's like
saying, well, you can produce the code and we'll hold it for you. No, I
think we need to be involved in the production of code as well. Open
source doesn't have to be a charity event. In order to develop quality
in the Python based code stacks, we have to promote a meritocracy, with
resources flowing to quality producers, instead of leaving it up to
chance and hobbyism and the random client need being serviced at the
wrong layer in the stack. If you remember all the bugs in CMF which were
fixed by Plone developent, you'll know what I'm taking about.

-- 
Sincerely,

Chris Calloway
http://www.seacoos.org
office: 17-6 Venable Hall   phone: (919) 962-4323
mail: Campus Box #3300, UNC-CH, Chapel Hill, NC 27599





More information about the triangle-zpug mailing list