Re: [Zope3-dev] RFC: Reunite Zope 2 and Zope 3 in the source code repository
On Wednesday 23 November 2005 10:16, Philipp von Weitershausen wrote:
Sounds crazy, I know. But I'm serious. Looking for your comments at: http://dev.zope.org/Zope3/ReuniteZope2AndZope3InTheSourceCodeRepository
I am -1. If I could I would veto this proposal. Here is why: To be totally honest, I really, really don't care about Zope 2! I am a Zope 3 developer. If Zope 2 code is in the Zope 3 code base, I have to relearn it again and additionally learn Five. Why? Just so I can keep developing Zope 3. This may raise the contribution bar too high for me and I would consider stopping to contribute. If the bar is too high for me, what do you expect from other people? Next, there are several third party applications that do not care about Zope 2 either, but that use the trunk to do their development with. One example is SchoolTool. Having to checkout both, Zope 3 and 2 would just be ridiculous, to say the least! (Note that several contributions of mine during the last weeks were due to my work on SchoolTool using a writeable Zope 3 trunk checkout.) The proposal only benefits Zope 2 people, really. Sure, some of the stuff in Zope 2 that should be forward-ported, but that's minimal. Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
Stephan Richter wrote:
On Wednesday 23 November 2005 10:16, Philipp von Weitershausen wrote:
Sounds crazy, I know. But I'm serious. Looking for your comments at: http://dev.zope.org/Zope3/ReuniteZope2AndZope3InTheSourceCodeRepository
I am -1. If I could I would veto this proposal. Here is why:
To be totally honest, I really, really don't care about Zope 2!
I'll debate with you this reason. I don't think that this changes your dislike of merging the repositories and this argument is on a side-track and not intended to convince you of this. What my point is here is that your attitude about Zope 2 is wrong: as a pure-play Zope 3 developer you *should* care about Zope 2. Some of us have been doing quite a bit of work of bringing Zope 3 to the Zope 2 world. I believe that at least partially as a result of this, Zope 3 is getting a lot more attention from Zope 2 developers. I think that this attention is extremely valuable to the Zope 3 project. There is an awful lot of experience, skills and knowledge in the Zope 2 world that is immensely valuable to Zope 3 developers. We *don't* have a full respresentation of these extremely valuable perspectives in the Zope 3 development community right now. If Zope 2 developers get the impression that core Zope 3 developers don't give a shit about Zope 2, they may not be so likely to actually come on board. That would be a disastrous development indeed. We really need an increased connection between the Zope 2 world and the Zope 3 world. Regards, Martijn
What my point is here is that your attitude about Zope 2 is wrong: as a pure-play Zope 3 developer you *should* care about Zope 2.
Some of us have been doing quite a bit of work of bringing Zope 3 to the Zope 2 world. I believe that at least partially as a result of this, Zope 3 is getting a lot more attention from Zope 2 developers. I think that this attention is extremely valuable to the Zope 3 project. There is an awful lot of experience, skills and knowledge in the Zope 2 world that is immensely valuable to Zope 3 developers. We *don't* have a full respresentation of these extremely valuable perspectives in the Zope 3 development community right now.
If Zope 2 developers get the impression that core Zope 3 developers don't give a shit about Zope 2, they may not be so likely to actually come on board. That would be a disastrous development indeed. We really need an increased connection between the Zope 2 world and the Zope 3 world.
... and I'm one of those developers. I care, for the moment, about one thing: Plone. I want Plone to move to use more Zope 3 technology, sure, because it's clearly superior. But we're not going to re-write it from scratch, and neither, I suspect, will most large Z2 applications. So my path to Z3 is through Z2, by way of Five. Frankly, without that migration path, Z3 is nice and pretty and architecturally very cool and totally irrelevant to me. And judging from the size of the Plone community, if nothing else, there's quite a few people like me. Martin (about to buy Phil's book, honest) -- (muted)
Martin Aspeli wrote:
... and I'm one of those developers. I care, for the moment, about one thing: Plone. I want Plone to move to use more Zope 3 technology, sure, because it's clearly superior. But we're not going to re-write it from scratch, and neither, I suspect, will most large Z2 applications.
That I don't know about. Plone probably should, for a number of reasons, but I guess it's quite a hard problem to tackle...
So my path to Z3 is through Z2, by way of Five. Frankly, without that migration path, Z3 is nice and pretty and architecturally very cool and totally irrelevant to me.
Indeed, but why does that mean people in Z3-land who feel the same way as you about Z2 and Plone should have to put up with looking after all of Z2? ;) cheers, Chris -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk
On 23 Nov 2005, at 20:10, Martijn Faassen wrote:
If Zope 2 developers get the impression that core Zope 3 developers don't give a shit about Zope 2, they may not be so likely to actually come on board. That would be a disastrous development indeed. We really need an increased connection between the Zope 2 world and the Zope 3 world.
Amen to that. I don't see how this "only benefits Zope2 developers". Zope2 developers already get the whole bag, lock, stock and barrel. There is no change at all for them. People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style. Ok, I accept that, no problem at all. But why should this be any different for Zope3 developers, obviously including Zope2 code would mean exactly the same thing for them. Come on now. jens
On Wednesday 23 November 2005 18:49, Jens Vagelpohl wrote:
People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style. Ok, I accept that, no problem at all. But why should this be any different for Zope3 developers, obviously including Zope2 code would mean exactly the same thing for them. Come on now.
Personally, I have never advocated inserting Zope 3 into Zope 2. Some people really wanted Zope 3 in Zope 2, so that they could use the new technology. So they added it. That's fine by me. But if they then turn around and say, "Look we have Zope 3 in Zope 2, so you should also have Zope 2 in Zope 3.", then I am complaining loudly, because I do not want to have anything to do with Zope 2. And it just means that I am becoming a Zope 2 developer again. Forget that! I'd rather fork Zope 3, then work on a version that has Zope 2 in it. It is just too much overhead for me to know all the involved technologies (Zope 2 and Five). I have barely time to keep up with Zope 3 and stay on top of it. Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
On 24 Nov 2005, at 00:09, Stephan Richter wrote:
On Wednesday 23 November 2005 18:49, Jens Vagelpohl wrote:
People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style. Ok, I accept that, no problem at all. But why should this be any different for Zope3 developers, obviously including Zope2 code would mean exactly the same thing for them. Come on now.
Personally, I have never advocated inserting Zope 3 into Zope 2. Some people really wanted Zope 3 in Zope 2, so that they could use the new technology. So they added it. That's fine by me. But if they then turn around and say, "Look we have Zope 3 in Zope 2, so you should also have Zope 2 in Zope 3.", then I am complaining loudly, because I do not want to have anything to do with Zope 2. And it just means that I am becoming a Zope 2 developer again. Forget that! I'd rather fork Zope 3, then work on a version that has Zope 2 in it. It is just too much overhead for me to know all the involved technologies (Zope 2 and Five). I have barely time to keep up with Zope 3 and stay on top of it.
I believe your assertion that you have to deal with Zope2 and become "a Zope 2 developer again" is just plain wrong. As wrong as Zope2 developers have to be Zope3 developers. jens
On Wednesday 23 November 2005 19:26, Jens Vagelpohl wrote:
I believe your assertion that you have to deal with Zope2 and become "a Zope 2 developer again" is just plain wrong. As wrong as Zope2 developers have to be Zope3 developers.
How so? Let's say I refactor something in the component architecture. In order to make the tests pass, I have to at least fix Five failures as well. In order to fix Five, I have to understand Zope 2. Thus I am effectively a Zope 2 developer. Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
While I don't agree with the +1 voters, I understand and appreciate their arguments. That said... On Nov 23, 2005, at 6:49 PM, Jens Vagelpohl wrote:
People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style. Ok, I accept that, no problem at all. But why should this be any different for Zope3 developers, obviously including Zope2 code would mean exactly the same thing for them. Come on now.
...this is not true. Zope 2 depends on Zope 3, via Five. Zope 3 does not depend on Zope 2. Therefore, making a change in Zope 2 cannot affect functionality in the slightest, let alone break a test, in Zope 3. The same cannot be said of the reverse. Zope 2 devs don't have to touch Zope 3 unless they want to leverage some cool new feature--in which case they are Zope Five devs, probably. Zope 3 devs must touch Zope 2, in this new world order, whether they want to or not, when changes break the stuff that Zope 2 has leveraged. To grant a point to Philipp's argument, it's possible that changes that break Zope 2 are non-backwards-compatible changes in Zope 3 that should have been caught. But consider this story: a Zope 3 dev changes something and deprecates an API. As part of the dev's responsibility, the checkin also makes all code in Zope 3 use the replacement API. Now Zope 2 works, but is generating deprecation warnings whenever the deprecated API is called. Is it the Zope 3 dev's responsibility to change Zope 2 to eliminate the deprecation warnings? What about in the following release when the old Zope 3 API is eliminated--whose responsibility is it then to fix Zope 2? If you view Zope 2 as a downstream client of Zope 3, you probably give one answer; if you view the two projects as a mingled whole, you probably give another. The question here is effectively whether all Zope 3 developers must become Zope 'Five' developers. As you said, Zope 2 developers can choose to proceed essentially unaffected. Zope 3 devs could not. Gary
On Wednesday 23 November 2005 22:14, Gary Poster wrote:
The question here is effectively whether all Zope 3 developers must become Zope 'Five' developers. As you said, Zope 2 developers can choose to proceed essentially unaffected. Zope 3 devs could not.
Amen. Regards, Stephan -- Stephan Richter CBU Physics & Chemistry (B.S.) / Tufts Physics (Ph.D. student) Web2k - Web Software Design, Development and Training
Gary Poster wrote:
Zope 2 depends on Zope 3, via Five. Zope 3 does not depend on Zope 2.
A very good point, but one which makes me feel that Zope 2 shouldn't be merged in with Zope 3 ;-) Put differently, if we're merging in Zope 2 into the repository, then why not SchoolTool, or any of the other projects that _use_ Zope 3?
Zope 2 devs don't have to touch Zope 3 unless they want to leverage some cool new feature--in which case they are Zope Five devs, probably. Zope 3 devs must touch Zope 2, in this new world order, whether they want to or not, when changes break the stuff that Zope 2 has leveraged.
I don't agree. Again, if Zope 3 changes break SchoolTool, is that a Zope 3 developer problem or a SchoolTool problem? FOrcing the maintenance of Zope 2 onto the already overloaded Zope 3 devs seems a little unfair...
The question here is effectively whether all Zope 3 developers must become Zope 'Five' developers. As you said, Zope 2 developers can choose to proceed essentially unaffected. Zope 3 devs could not.
And this for me, means that even if the repos merge, so evil svn:externals can be avoided, the tests should not be run together on the Zope 3 side, and "Zope 3" should come bundled with Zope 2, even if Zope 2 comes bundled with Zope 3 and Five! cheers, Chris -- Simplistix - Content Management, Zope & Python Consulting - http://www.simplistix.co.uk
On Nov 30, 2005, at 2:18 PM, Chris Withers wrote:
Gary Poster wrote:
Zope 2 depends on Zope 3, via Five. Zope 3 does not depend on Zope 2.
A very good point, but one which makes me feel that Zope 2 shouldn't be merged in with Zope 3 ;-)
Actually, yes, all of my points were made to that end--so AFAICT you are agreeing with me, not disagreeing. :-) Gary
Jens Vagelpohl wrote:
...
People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style.
Hee hee. And they believed it? Do they wanna buy a bridge? ;) This is an age old argument made when someone wants to add a new feature to a development system. It is patently false. Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
Jim Fulton wrote:
Jens Vagelpohl wrote:
...
People keep telling Zope2 developers that the inclusion of Zope3 doesn't mean you have to touch it, if you don't use it it is just inert code that won't cause any change in your Zope2 development style.
Hee hee. And they believed it? Do they wanna buy a bridge? ;)
This is an age old argument made when someone wants to add a new feature to a development system. It is patently false.
I recall a slightly different discussion I was involved in. I remember Zope 2 core developers worrying about the inclusion of Five in Zope 2.8; they were worried they'd need to maintain its codebase. The arguments against this were: * there's a lively development community around Five, don't worry * Five has a minimal impact on Zope 2; Zope 2 sources itself weren't changing. Both were true. I don't think it was claimed that your development style wouldn't be affected, as obviously we hope people will actually *use* Five in Zope 2 development. With Zope 2.9, this story is starting to change, as Zope 3 technology is making it deeper into Zope 2. Then again, I think the people who worried then have been becoming more familiar with Five since then, so hopefully appreciate it more now as a feature, not just as a potential maintenance burden. Regards, Martijn
On Nov 24, 2005, at 8:37 AM, Martijn Faassen wrote:
I recall a slightly different discussion I was involved in. I remember Zope 2 core developers worrying about the inclusion of Five in Zope 2.8; they were worried they'd need to maintain its codebase.
I was one of these people. Since then, I've completely changed my mind; it was a pure win. - C
Chris McDonough wrote:
On Nov 24, 2005, at 8:37 AM, Martijn Faassen wrote:
I recall a slightly different discussion I was involved in. I remember Zope 2 core developers worrying about the inclusion of Five in Zope 2.8; they were worried they'd need to maintain its codebase.
I was one of these people. Since then, I've completely changed my mind; it was a pure win.
It makes me happy to hear that. Thanks! Regards, Martijn
On 11/24/05 8:54 AM, "Chris McDonough" <chrism@plope.com> wrote:
On Nov 24, 2005, at 8:37 AM, Martijn Faassen wrote:
I recall a slightly different discussion I was involved in. I remember Zope 2 core developers worrying about the inclusion of Five in Zope 2.8; they were worried they'd need to maintain its codebase.
I was one of these people. Since then, I've completely changed my mind; it was a pure win.
- C
As was I. Five has brought me and a few others I know closer to Zope 3. This is a huge win. Many of us have lots of experience to bring to the benefit of the Z3 community. As a result, I am excited to take even more strides to Z3. Phillips goal is a good one. The challenge is if there is resistance, how to smooth the bridge. Rather then Z3 developers shoot it down outright, they should provide reasonable alternatives (other then doomsday scenarios). It is a good thing to bring more developers into the Z3 community, many of us Z2 developers are hungry for this. I don't know the answer for every developer on either side of this argument, but there must be an effective compromise out there. The two groups need to work to come to that solution and not alienate one or the other or blindly shoot it down and hope it will go away. It's unreasonable to do so. Taking my ball and heading homely, Andrew
Martijn Faassen wrote:
Stephan Richter wrote:
On Wednesday 23 November 2005 10:16, Philipp von Weitershausen wrote:
Sounds crazy, I know. But I'm serious. Looking for your comments at: http://dev.zope.org/Zope3/ReuniteZope2AndZope3InTheSourceCodeRepository
I am -1. If I could I would veto this proposal. Here is why:
To be totally honest, I really, really don't care about Zope 2!
I'll debate with you this reason. I don't think that this changes your dislike of merging the repositories and this argument is on a side-track and not intended to convince you of this.
What my point is here is that your attitude about Zope 2 is wrong: as a pure-play Zope 3 developer you *should* care about Zope 2.
Some of us have been doing quite a bit of work of bringing Zope 3 to the Zope 2 world. I believe that at least partially as a result of this, Zope 3 is getting a lot more attention from Zope 2 developers. I think that this attention is extremely valuable to the Zope 3 project. There is an awful lot of experience, skills and knowledge in the Zope 2 world that is immensely valuable to Zope 3 developers. We *don't* have a full respresentation of these extremely valuable perspectives in the Zope 3 development community right now.
If Zope 2 developers get the impression that core Zope 3 developers don't give a shit about Zope 2, they may not be so likely to actually come on board. That would be a disastrous development indeed. We really need an increased connection between the Zope 2 world and the Zope 3 world.
I couldn't have said it better, Martijn. Stephan might not care about the Zope 2 codebase (frankly, I'm mostly with him on that, which is why I'm working to improve it), but let's not forget that Zope 3 is currently actively managed by only 10 or so people. Everytime we make a release there are heroic efforts involved, mostly by Stephan himself. How long are we supposed to continue like this? Like Martijn, I strongly believe reuniting efforts will eventually mean *more* resources for Zope 3, not less (this is also a point where I disagre with Gary). As Martin pointed out with his own example, the reunification will tremendously lower the bar for more Zope 3 contributors which, given the time and resources other projects have and are willing to spend on the framework (e.g. thanks to Goldegg), should not be ignored. To give you another, much better example: Florent recently brought Zope 3 events ot Zope 2 and made a great effort in doing so. In return, this work now made him think about improving the Zope 3 object event hiearchy, the post to the zope3-dev list was even sent today. What better example of an improvement of Zope 3 due to Zope 2 integration can there be? Philipp ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.
Martijn Faassen wrote:
Stephan Richter wrote:
On Wednesday 23 November 2005 10:16, Philipp von Weitershausen wrote:
Sounds crazy, I know. But I'm serious. Looking for your comments at: http://dev.zope.org/Zope3/ReuniteZope2AndZope3InTheSourceCodeRepository
I am -1. If I could I would veto this proposal. Here is why:
To be totally honest, I really, really don't care about Zope 2!
I'll debate with you this reason. I don't think that this changes your dislike of merging the repositories and this argument is on a side-track and not intended to convince you of this.
What my point is here is that your attitude about Zope 2 is wrong: as a pure-play Zope 3 developer you *should* care about Zope 2.
Some of us have been doing quite a bit of work of bringing Zope 3 to the Zope 2 world. I believe that at least partially as a result of this, Zope 3 is getting a lot more attention from Zope 2 developers. I think that this attention is extremely valuable to the Zope 3 project. There is an awful lot of experience, skills and knowledge in the Zope 2 world that is immensely valuable to Zope 3 developers. We *don't* have a full respresentation of these extremely valuable perspectives in the Zope 3 development community right now.
If Zope 2 developers get the impression that core Zope 3 developers don't give a shit about Zope 2, they may not be so likely to actually come on board. That would be a disastrous development indeed. We really need an increased connection between the Zope 2 world and the Zope 3 world.
+100 Jim -- Jim Fulton mailto:jim@zope.com Python Powered! CTO (540) 361-1714 http://www.python.org Zope Corporation http://www.zope.com http://www.zope.org
participants (10)
-
Andrew Sawyers -
Chris McDonough -
Chris Withers -
Gary Poster -
Jens Vagelpohl -
Jim Fulton -
Martijn Faassen -
Martin Aspeli -
Philipp von Weitershausen -
Stephan Richter