Zope Corporation's Initial Reaction on the ZF Comments
My first attempt to post to this list bounced, because I'm not a subscriber. Jim enabled me to post, so I'm resending, without cc'ing the z3lab list again. If you hit reply-all, please add z3lab@lists.nuxeo.com to the cc list (if you're allowed to post there as well :-) Hi all. Whew, lots of traffic, with good ideas and comments made by all. While all of us at Zope Corp appreciate the input, we can't lose sight of two points: 1) Even though lots of the "Rock Stars" of the Zope Community are on one or both of these lists, not all are, and certainly not the entire Zope Community (especially _customers_), so these lists cannot substitute for the entire input stream. 2) We have called for an International IRC chat to discuss this next Tuesday, and tried to pick a time that could work for people from the West Coast of the US all the way to hardy souls in Asia, but at least Eastern Europe. Until people can weigh in and get a sense of everyone's responses, this list is just fodder for that discussion. As such, it is highly unlikely that I will post again on this specific thread to these lists before the IRC, so _please_ don't be offended if you have a fantastic rebuttal to a point that I try to make here, and don't get a response. I just subscribed to the z3lab list (I'm not on dev), and will see your response, and hopefully prepare a ZC response for the IRC. OK, enough with the background, on to make some points :-) I found all of the discussion interesting, but I am also confused by some of it. Specifically, the use of the Plone Foundation as the model that we should all aspire to. If I understand my facts correctly, the Plone Foundation was kicked off (and likely funded by) Computer Associates (CA). They still have 2 board seats as far as I can see. In fact, for all the rhetoric about "individuals", each board member has their company named after them, which implies to me that people looking at that list should assume that they vote the way their company would want them to, not the way they feel about specific issues. Specifically, if Norm Patriquin of CA leaves CA, will he remain a board member, or does CA have some right to appoint another director in his place? If the answer is that CA controls the board seat, then please let's stop pretending that this is all about "individuals". It's obvious that companies do not vote, individuals vote. It is also obvious that individuals who represent companies are more likely to vote in a direction that is good for their company. Nothing wrong with that (IMHO) as long they can't force something on the rest of the members. Second, if we had adopted the Plone Foundation organization verbatim, just changing the word Plone to Zope, would that have been 100% satisfactory to everyone in the Zope world? If so, that would surprise me, but more importantly, it would still have been a "unilateral" move on our part, not to even allow potential dissenters a say. In other words, there is no one model that will work for everyone, and we are being careful not to set _anything_ in stone until we hear everyone's thoughts. If we intended to act unilaterally, and in only our interests, we would have announced a completed Foundation with a "take it or leave it" attitude, or we would have put a very short date on getting it done. Instead, we announced that it would be done by the end of October 2005, so that _this_ process could have a real chance to succeed in an open manner. No one has a gun to our head to do this, and in fact, no one has the slightest leverage on us to do this. We are doing it because we _want_ to, because we think it's the _right thing do_, and because we think the timing is right with Zope 3 ready for prime time, and ready to explode. If we wanted to try and retain the maximum benefit from that explosion, we would probably just keep it all to ourselves. We are not, and we would like at least the benefit of the doubt as to our motives, if not an actual "Thank You" :-) Like Stefane, we too are slightly leaning towards an Eclipse model. In that model, committers are first-class members, and do _not_ pay dues! Companies and Customers (in their term "Consumers") are first-class members too, but not only pay dues (don't worry, we won't charge what they do ;-), but also _have to commit development resources_. No one vendor has _any_ control of _anything_ in the Eclipse Foundation, but they don't apologize for the fact that the underlying software is _strategic_ to the Vendor organizations in their attempt to make a profit. Stefane Fermigier wrote:
IMHO, "vendor-neutral" means, in this context, that the Foundation must take into account the interests of all the stakeholders (individual hackers, vendors, customers), and shouldn't be interpreted as "vendor-free".
I agreed, and would add that "vendor-neutral" can also (and IMHO should) be "vendor-friendly". Let's not forget that ZPL is not GPL. We chose a commercially friendly license 7 years ago, and have only made it "friendlier" to people who want to profit from Zope each time we changed it. Why would we spin off the copyright to the code in a _less_ commercially friendly way. Our version of commercial friendliness is equally friendly to individuals and companies, both of who peacefully co-exist in the Zope ecosystem. To summarize, we are doing this for the benefit of all. No matter how bad the Foundation ends up (in the opinion of any individual or company), if they have an interest in Zope, they will already be better off. Under all circumstances, the copyright will be held by the Foundation, and the Foundation's right to call the software "Zope" will exist forever, so all of the previous doomsday predictions about Zope Corp being bought and the Zope trademark used to stop people from working on software called Zope will be over. No to some specific points... Paul Everitt
Can you assure us regarding the most fundamental points:
1) The Zope Foundation will not have any special terms for any companies, unless those terms are approved by the membership.
Correct. There will be no special terms. It is possible that there will be one exception to this, but this too hasn't been decided yet. Perhaps, as the "reward" for Zope Corporation donating all of the current copyright to the Foundation, Zope Corporation will be have a permanent board seat. That won't give them any special powers, but will be a recognition of their original contribution.
2) Voting, bylaws, policies, and *ownership of assets* will be in the hands of the developers acting as members. Minor issues are delegated to a freely elected board.
Yes. The Foundation (and therefore its members) own all of the assets. As for whether only "Minor" issues get raised to the board, I'm not sure what the intent of that is. In general, even major issues (not code check-ins, which could be construed as a "Minor" issue) can and should be discussed by the board, but then should be presented to the membership for a vote, according to the bylaws.
3) There are no special terms or negotiation points that favor one group over another. For example, CPS-the-product will not be shipped with Z3ECM unless we agree.
Obviously! Paul Everitt wrote:
The ASF has companies like IBM that put legions of developers on the code. They can live with this policy. And Apache seems to flourish.
Correct. Of course, when the body of work known as Eclipse was first developed at IBM, which Apache was _not_, somehow they chose to create a new structure (The Eclipse Foundation), rather than simply choosing the ASF as the model. So, while they are obviously fine with ASF as members, when their own original contribution was at stake, they went another route.
I can't think of any open source foundation that has company voting in the governance. There is a role for companies, as sponsors. But, not in the governance.
Stefane pointed out the Eclipse Foundation already, so it's a little surprising that you are still looking for an example of another successful open source Foundation that is also openly vendor friendly, if not vendor driven. *********************************************************** Again, it is highly unlikely that I will respond again before the IRC, even if you are purposely "provocative", so please don't take any personal offense! Thanks for reading this far!
I found this interesting enough to look into anyway... for anybody who is interested, here's the scoop. CA has 2 seats out of 9 on the Plone Foundation board. Apparently there's special treatment of these seats via http://plone.org/foundation/about/board/special_seats which is mostly a perk to CA for providing $100,000 (!) in seed funding. There is a special provision in place for allowing CA to keep these board seats. That candiacy provision can apparently be voted out only by the Board itself (these seats cannot be filled with non-CA people via a normal general election). Beyond this rule, there don't seem to be any bylaws to keep this ratio of individuals-to-companies intact. An advisory nonvoting seat on the board was created for Porter-Novelli due to their willingness to provide pro-bono marketing. The rest of the board seats are filled with independents and people representing smaller companies, all of whom are coders and could have only been elected on merit (I suspect there's just not enough at stake for it to have happened any other way ;-). The Eclipse foundation has 19 total board members, 6 of whom appear to have been elected based on some definition of merit. Of the remaining 13 seats, 10 seats are filled with representatives of "Strategic Developer" companies (the companies "own" the seats presumably as long as they keep funding the foundation), and 3 are filled with "Add-In Provider" company seats (these seats can be kept by their original owners if they move from company to company as long as they fit the requirements of a "Committer Member"). There are rules in their bylaws that serve to keep this relative funder-to-individual Board member ratio intact. The Apache Software Foundation has 9 members (http://www.apache.org/foundation/board/). I don't recognize all the names, but at least three are definitely coders, and companies are not mentioned at all. Their bylaws have no math in them for retaining board seats based on payment into the foundation, but maybe there is some informal agreement. <research'ly y'rs> - C On Fri, 2005-06-17 at 12:32 -0400, Hadar Pedhazur wrote:
My first attempt to post to this list bounced, because I'm not a subscriber. Jim enabled me to post, so I'm resending, without cc'ing the z3lab list again. If you hit reply-all, please add z3lab@lists.nuxeo.com to the cc list (if you're allowed to post there as well :-)
Hi all. Whew, lots of traffic, with good ideas and comments made by all. While all of us at Zope Corp appreciate the input, we can't lose sight of two points:
1) Even though lots of the "Rock Stars" of the Zope Community are on one or both of these lists, not all are, and certainly not the entire Zope Community (especially _customers_), so these lists cannot substitute for the entire input stream.
2) We have called for an International IRC chat to discuss this next Tuesday, and tried to pick a time that could work for people from the West Coast of the US all the way to hardy souls in Asia, but at least Eastern Europe. Until people can weigh in and get a sense of everyone's responses, this list is just fodder for that discussion.
As such, it is highly unlikely that I will post again on this specific thread to these lists before the IRC, so _please_ don't be offended if you have a fantastic rebuttal to a point that I try to make here, and don't get a response. I just subscribed to the z3lab list (I'm not on dev), and will see your response, and hopefully prepare a ZC response for the IRC.
OK, enough with the background, on to make some points :-)
I found all of the discussion interesting, but I am also confused by some of it. Specifically, the use of the Plone Foundation as the model that we should all aspire to.
If I understand my facts correctly, the Plone Foundation was kicked off (and likely funded by) Computer Associates (CA). They still have 2 board seats as far as I can see. In fact, for all the rhetoric about "individuals", each board member has their company named after them, which implies to me that people looking at that list should assume that they vote the way their company would want them to, not the way they feel about specific issues.
Specifically, if Norm Patriquin of CA leaves CA, will he remain a board member, or does CA have some right to appoint another director in his place? If the answer is that CA controls the board seat, then please let's stop pretending that this is all about "individuals".
It's obvious that companies do not vote, individuals vote. It is also obvious that individuals who represent companies are more likely to vote in a direction that is good for their company. Nothing wrong with that (IMHO) as long they can't force something on the rest of the members.
Second, if we had adopted the Plone Foundation organization verbatim, just changing the word Plone to Zope, would that have been 100% satisfactory to everyone in the Zope world? If so, that would surprise me, but more importantly, it would still have been a "unilateral" move on our part, not to even allow potential dissenters a say. In other words, there is no one model that will work for everyone, and we are being careful not to set _anything_ in stone until we hear everyone's thoughts.
If we intended to act unilaterally, and in only our interests, we would have announced a completed Foundation with a "take it or leave it" attitude, or we would have put a very short date on getting it done. Instead, we announced that it would be done by the end of October 2005, so that _this_ process could have a real chance to succeed in an open manner.
No one has a gun to our head to do this, and in fact, no one has the slightest leverage on us to do this. We are doing it because we _want_ to, because we think it's the _right thing do_, and because we think the timing is right with Zope 3 ready for prime time, and ready to explode. If we wanted to try and retain the maximum benefit from that explosion, we would probably just keep it all to ourselves. We are not, and we would like at least the benefit of the doubt as to our motives, if not an actual "Thank You" :-)
Like Stefane, we too are slightly leaning towards an Eclipse model. In that model, committers are first-class members, and do _not_ pay dues! Companies and Customers (in their term "Consumers") are first-class members too, but not only pay dues (don't worry, we won't charge what they do ;-), but also _have to commit development resources_. No one vendor has _any_ control of _anything_ in the Eclipse Foundation, but they don't apologize for the fact that the underlying software is _strategic_ to the Vendor organizations in their attempt to make a profit.
Stefane Fermigier wrote:
IMHO, "vendor-neutral" means, in this context, that the Foundation must take into account the interests of all the stakeholders (individual hackers, vendors, customers), and shouldn't be interpreted as "vendor-free".
I agreed, and would add that "vendor-neutral" can also (and IMHO should) be "vendor-friendly". Let's not forget that ZPL is not GPL. We chose a commercially friendly license 7 years ago, and have only made it "friendlier" to people who want to profit from Zope each time we changed it. Why would we spin off the copyright to the code in a _less_ commercially friendly way. Our version of commercial friendliness is equally friendly to individuals and companies, both of who peacefully co-exist in the Zope ecosystem.
To summarize, we are doing this for the benefit of all. No matter how bad the Foundation ends up (in the opinion of any individual or company), if they have an interest in Zope, they will already be better off. Under all circumstances, the copyright will be held by the Foundation, and the Foundation's right to call the software "Zope" will exist forever, so all of the previous doomsday predictions about Zope Corp being bought and the Zope trademark used to stop people from working on software called Zope will be over.
No to some specific points...
Paul Everitt
Can you assure us regarding the most fundamental points:
1) The Zope Foundation will not have any special terms for any companies, unless those terms are approved by the membership.
Correct. There will be no special terms. It is possible that there will be one exception to this, but this too hasn't been decided yet. Perhaps, as the "reward" for Zope Corporation donating all of the current copyright to the Foundation, Zope Corporation will be have a permanent board seat. That won't give them any special powers, but will be a recognition of their original contribution.
2) Voting, bylaws, policies, and *ownership of assets* will be in the hands of the developers acting as members. Minor issues are delegated to a freely elected board.
Yes. The Foundation (and therefore its members) own all of the assets. As for whether only "Minor" issues get raised to the board, I'm not sure what the intent of that is. In general, even major issues (not code check-ins, which could be construed as a "Minor" issue) can and should be discussed by the board, but then should be presented to the membership for a vote, according to the bylaws.
3) There are no special terms or negotiation points that favor one group over another. For example, CPS-the-product will not be shipped with Z3ECM unless we agree.
Obviously!
Paul Everitt wrote:
The ASF has companies like IBM that put legions of developers on the code. They can live with this policy. And Apache seems to flourish.
Correct. Of course, when the body of work known as Eclipse was first developed at IBM, which Apache was _not_, somehow they chose to create a new structure (The Eclipse Foundation), rather than simply choosing the ASF as the model. So, while they are obviously fine with ASF as members, when their own original contribution was at stake, they went another route.
I can't think of any open source foundation that has company voting in the governance. There is a role for companies, as sponsors. But, not in the governance.
Stefane pointed out the Eclipse Foundation already, so it's a little surprising that you are still looking for an example of another successful open source Foundation that is also openly vendor friendly, if not vendor driven.
***********************************************************
Again, it is highly unlikely that I will respond again before the IRC, even if you are purposely "provocative", so please don't take any personal offense!
Thanks for reading this far!
_______________________________________________ Zope-Dev maillist - Zope-Dev@zope.org http://mail.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope )
[Chris McDonough]
... The Apache Software Foundation has 9 members (http://www.apache.org/foundation/board/).
Their board of directors has 9 members, but the ASF has many more members than that: http://www.apache.org/foundation/members.html
I don't recognize all the names, but at least three are definitely coders, and companies are not mentioned at all. Their bylaws have no math in them for retaining board seats based on payment into the foundation, but maybe there is some informal agreement.
I doubt it, because the Python Software Foundation's (of which I'm a director) bylaws were modeled on the ASF's. For both, you can become a member only by being nominated by an existing member, and then voted in by a majority of existing members. Anyone (regardless of whether they're a member) can sit on the board, and the board is determined solely by membership vote. So any informal agreement would have to be embraced by a majority of the members to be effective, and that's just unlikely. For the PSF, a sure way to get on the Board is to say you're willing to do it <0.3 wink>; I imagine it's similar with the ASF. We should note that the ASF and PSF are both 501(c)(3) non-profits under US tax law, which puts imprecise but serious legal bounds on how much they _can_ be "controlled by", or benefit, a small group without getting into deep legal doo doo. In effect, we endure those restrictions so that US contributors can get a tax deduction, and to make legal actions against us especially unattractive (this one's complicated, but it's not a coincidence that you don't hear about many people suing the Red Cross <wink>). I don't know which part of the non-profit spectrum the Zope Foundation is aiming at, but I'd guess that the ZF wouldn't want to hassle with a charity's legal restrictions (yup, the PSF is a "public charity" in US legal jargon, same as the Red Cross).
participants (3)
-
Chris McDonough -
Hadar Pedhazur -
Tim Peters