From: Jody G. <jga...@re...> - 2007-10-17 16:43:36
|
I have tried talking to a few other board members (Chris and Paul) and they do not seem to be decisive as Arnulf Cristl. If we can review the following proposal and vote on it by Nov 2nd, we would at least go into this meeting in an organized fashion: - http://docs.codehaus.org/display/GEOTOOLS/Update+GeoTools+Headers As usual David Zwiers sends me a private email saying he wont hand over (c); while Refractions can do that on his behalf for our XML parsing code chances are we should start to make other plans. I went through the OSGeo projects and about half are (c) as individuals and the other half (c) as an organization. Jody |
From: Martin D. <mar...@ge...> - 2007-10-17 17:29:04
|
Jody Garnett a écrit : > If we can review the following proposal and vote on it by Nov 2nd, we > would at least go into this meeting in an organized fashion: > - http://docs.codehaus.org/display/GEOTOOLS/Update+GeoTools+Headers I tried twice today and the web site seems to be down. Will try again tomorrow. Thanks for pushing that Martin |
From: Jody G. <jga...@re...> - 2007-10-17 17:35:14
|
Jesse also reported problems; it is working now for us. Remember you can always jump on irc.codehaus.org#codehaus and let them know when things are slow (often we are the first people to notice). Jody > Jody Garnett a écrit : > >> If we can review the following proposal and vote on it by Nov 2nd, we >> would at least go into this meeting in an organized fashion: >> - http://docs.codehaus.org/display/GEOTOOLS/Update+GeoTools+Headers >> > > I tried twice today and the web site seems to be down. Will try again tomorrow. > > Thanks for pushing that > > Martin > |
From: Jody G. <jga...@re...> - 2007-10-17 17:41:00
|
It seems that we are having some trouble over at codehaus; here is the point of the exercise: > /* > * GeoTools - The Open Source Java GIS Tookit > * http://geotools.org > * (C) 2007, Open Source Geospatial Foundation (OSGeo) > * (C) 2005-2006, GeoTools Project Management Committee (PMC) > * (C) 2004, Refractions Research Inc (RRI) > * > * This library is free software; you can redistribute it and/or > * modify it under the terms of the GNU Lesser General Public > * License as published by the Free Software Foundation; > * version 2.1 of the License. > * > * This library is distributed in the hope that it will be useful, > * but WITHOUT ANY WARRANTY; without even the implied warranty of > * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > * Lesser General Public License for more details. > */ Things to note: - (c) Open Source Geospatial Foundation - leaving (c) GeoTools PMC in there even if it has no legal weight - previous organizations (or individuals) are still listed - license is LGPL 2.1 - Management is spelled correctly The interesting part is putting this header on our files without doing a code contribution agreement. Apparently there is an understanding that commit history is good enough; when the board reached this understanding is an open question. Cheers, Jody |
From: Andrea A. <aa...@op...> - 2007-10-17 17:43:43
|
Jody Garnett ha scritto: > It seems that we are having some trouble over at codehaus; here is the > point of the exercise: >> /* >> * GeoTools - The Open Source Java GIS Tookit >> * http://geotools.org >> * (C) 2007, Open Source Geospatial Foundation (OSGeo) >> * (C) 2005-2006, GeoTools Project Management Committee (PMC) >> * (C) 2004, Refractions Research Inc (RRI) >> * >> * This library is free software; you can redistribute it and/or >> * modify it under the terms of the GNU Lesser General Public >> * License as published by the Free Software Foundation; >> * version 2.1 of the License. >> * >> * This library is distributed in the hope that it will be useful, >> * but WITHOUT ANY WARRANTY; without even the implied warranty of >> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU >> * Lesser General Public License for more details. >> */ > Things to note: > - (c) Open Source Geospatial Foundation > - leaving (c) GeoTools PMC in there even if it has no legal weight > - previous organizations (or individuals) are still listed > - license is LGPL 2.1 > - Management is spelled correctly > > The interesting part is putting this header on our files without doing a > code contribution agreement. Apparently there is an understanding that > commit history is good enough; when the board reached this understanding > is an open question. I'm not against doing that work, provided it's wort something? Are we sure the board thinks it's ok and also sufficient from a legal point of view? Cheers Andrea |
From: Chris H. <ch...@op...> - 2007-10-17 20:43:32
Attachments:
cholmes.vcf
|
Jody Garnett wrote: > Andrea Aime wrote: >> Jody Garnett ha scritto: >>> The interesting part is putting this header on our files without >>> doing a code contribution agreement. Apparently there is an >>> understanding that commit history is good enough; when the board >>> reached this understanding is an open question. >> I'm not against doing that work, provided it's wort something? >> Are we sure the board thinks it's ok and also sufficient from a legal >> point of view? > That is what Arnulf Cristl indicated in that IRC chat; which made us sad > as it would mean that we wasted a lot of our time due to bad communication. The board was never requiring copyright assignment, we just thought it would be a nice thing to do, and didn't seem like it'd be all that hard and would make us feel better that we were on better legal ground. And it makes it easier to potentially change the license in the future - it felt like we could do the work and set us up to be in good shape if we wanted to change anything. But I feel comfortable from a legal stand point with just changing headers. I think we also need to post some notice publicly about changing things. But most open source legal advice points to things being fine if individual contributors retain copyright, or put it in to some vague PMC. C > > Asking on the incubator list I get the following response from Frank: >> All legal advice we have received indicates that a cvs commit action >> does not constitute "assignment of copyright". Though we have taken >> the position that it is sufficient to declare a change of license >> as long as it is done by the copyright holder, or with their agreement. >> >> So, are you talking about a commit operation for copyright assignment >> or for licensing? > Cameron chimes in with: >> Regarding assignment of copywrite. At the following meeting, it was >> argued by Eric Raymond, then passed that Contributor License >> Agreements (CLA)s are not required by projects. >> http://wiki.osgeo.org/index.php/IncCom_Meeting2 >> >> I think this is different to geotools' situation which requires >> retrospective re-assignment of copywrite to OSGeo > Frank clarifies with: >> Correct. That discussion was whether a contributor license agreement >> is necessary for projects where the original developer retains the >> copyright (as in most projects). The answer from Rich Steele was that >> having a CLA puts the organization in a stronger legal position but is >> not strictly necessary. We decided (within the incubator) not to require >> projects (like GDAL for instance) to require a CLA if they do not desire >> one. >> >> Since then we have set forth a policy in the incubator that projects >> without a CLA need to at least have committers agree to (not on paper) >> to some sort of committer guidelines that make their responsibilities >> reasonably clear, such as those at: >> >> http://wiki.osgeo.org/index.php/Commiter_Responsibilities_Guidelines >> >> This leaves us with three likely models for contributors in OSGeo: >> >> 1) copyright holder retains copyright, but signs a strong CLA (MapGuide, >> FDO, OpenLayers?) >> >> 2) copyright holder retains copyright, but only is only expected to >> follow >> reasonable contribution guidelines to keep IP clean. (GDAL, MapServer) >> >> 3) copyright holder turns over copyright to the foundation which >> requires a signed legal assignment on paper. (proposed for GeoTools) > Frank has the impression that contributors maintains (c) is the usual > way of things; doing some research shows that is not the case. But if > that is the only thing the board can get their head around perhaps that > is the step we will have to take? > > Jody > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > _______________________________________________ > Geotools-administration mailing list > Geo...@li... > https://lists.sourceforge.net/lists/listinfo/geotools-administration > > !DSPAM:4005,471667d9323412090977483! > |
From: Cameron S. <cam...@gm...> - 2007-10-18 01:20:35
|
Jody, thanks for setting up http://docs.codehaus.org/display/GEOTOOLS/Update+GeoTools+Headers Does this list ALL the remaining steps to graduation? The impression I got from IRC and personal email threads was that there was more. I'm hoping to see this, or another wiki list all your actions, especially any blocking issues so that we can focus our energy. Eg: If David Zwiers will not hand over (c), and this is considered an issue, then is should be listed. -- Regarding copywrite assignment, from a non-lawyer's perspective, I suspect that assigning copywrite to the PMC when you know it is not a legal entity is not a good idea. (I find it ironic that I'm involved in the mess created by assigning copywrite to the PMC when I flippantly helped create the (c) notice for geotools years ago. I appologize for my part in it.) What we did in Mapbuilder (where I created the same problem) was we listed all the PMC in the copywrite notice. Each file header is listed as: (c) as per mapbuilder/LICENSE And LICENCE lists all current and past PMC members. (actually, we use committers instead of PMC members but the same thing holds). We then got an archived email from all past contributors confirming that this was ok. (We did something similar when we switched from GPL to LGPL). This path might not be of interest to Geotools members and I don't want to push it upon you. I'm just suggesting it as an alternative which might help. -- Chris Holmes wrote: > > > Jody Garnett wrote: >> Andrea Aime wrote: >>> Jody Garnett ha scritto: >>>> The interesting part is putting this header on our files without >>>> doing a code contribution agreement. Apparently there is an >>>> understanding that commit history is good enough; when the board >>>> reached this understanding is an open question. >>> I'm not against doing that work, provided it's wort something? >>> Are we sure the board thinks it's ok and also sufficient from a legal >>> point of view? >> That is what Arnulf Cristl indicated in that IRC chat; which made us >> sad as it would mean that we wasted a lot of our time due to bad >> communication. > > The board was never requiring copyright assignment, we just thought it > would be a nice thing to do, and didn't seem like it'd be all that > hard and would make us feel better that we were on better legal > ground. And it makes it easier to potentially change the license in > the future - it felt like we could do the work and set us up to be in > good shape if we wanted to change anything. > > But I feel comfortable from a legal stand point with just changing > headers. I think we also need to post some notice publicly about > changing things. But most open source legal advice points to things > being fine if individual contributors retain copyright, or put it in > to some vague PMC. > > C > >> >> Asking on the incubator list I get the following response from Frank: >>> All legal advice we have received indicates that a cvs commit action >>> does not constitute "assignment of copyright". Though we have taken >>> the position that it is sufficient to declare a change of license >>> as long as it is done by the copyright holder, or with their agreement. >>> >>> So, are you talking about a commit operation for copyright assignment >>> or for licensing? >> Cameron chimes in with: >>> Regarding assignment of copywrite. At the following meeting, it was >>> argued by Eric Raymond, then passed that Contributor License >>> Agreements (CLA)s are not required by projects. >>> http://wiki.osgeo.org/index.php/IncCom_Meeting2 >>> >>> I think this is different to geotools' situation which requires >>> retrospective re-assignment of copywrite to OSGeo >> Frank clarifies with: >>> Correct. That discussion was whether a contributor license agreement >>> is necessary for projects where the original developer retains the >>> copyright (as in most projects). The answer from Rich Steele was that >>> having a CLA puts the organization in a stronger legal position but is >>> not strictly necessary. We decided (within the incubator) not to >>> require >>> projects (like GDAL for instance) to require a CLA if they do not >>> desire >>> one. >>> >>> Since then we have set forth a policy in the incubator that projects >>> without a CLA need to at least have committers agree to (not on paper) >>> to some sort of committer guidelines that make their responsibilities >>> reasonably clear, such as those at: >>> >>> http://wiki.osgeo.org/index.php/Commiter_Responsibilities_Guidelines >>> >>> This leaves us with three likely models for contributors in OSGeo: >>> >>> 1) copyright holder retains copyright, but signs a strong CLA >>> (MapGuide, >>> FDO, OpenLayers?) >>> >>> 2) copyright holder retains copyright, but only is only expected to >>> follow >>> reasonable contribution guidelines to keep IP clean. (GDAL, MapServer) >>> >>> 3) copyright holder turns over copyright to the foundation which >>> requires a signed legal assignment on paper. (proposed for GeoTools) >> Frank has the impression that contributors maintains (c) is the usual >> way of things; doing some research shows that is not the case. But if >> that is the only thing the board can get their head around perhaps >> that is the step we will have to take? >> >> Jody >> >> ------------------------------------------------------------------------- >> >> This SF.net email is sponsored by: Splunk Inc. >> Still grepping through log files to find problems? Stop. >> Now Search log events and configuration files using AJAX and a browser. >> Download your FREE copy of Splunk now >> http://get.splunk.com/ >> _______________________________________________ >> Geotools-administration mailing list >> Geo...@li... >> https://lists.sourceforge.net/lists/listinfo/geotools-administration >> >> !DSPAM:4005,471667d9323412090977483! >> > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > ------------------------------------------------------------------------ > > _______________________________________________ > Geotools-administration mailing list > Geo...@li... > https://lists.sourceforge.net/lists/listinfo/geotools-administration > -- Cameron Shorter Systems Architect, http://lisasoft.com.au Tel: +61 (0)2 8570 5050 Mob: +61 (0)419 142 254 |
From: Jody G. <jga...@re...> - 2007-10-18 00:14:26
|
Chris Holmes wrote: > The board was never requiring copyright assignment, we just thought it > would be a nice thing to do, and didn't seem like it'd be all that > hard and would make us feel better that we were on better legal > ground. And it makes it easier to potentially change the license in > the future - it felt like we could do the work and set us up to be in > good shape if we wanted to change anything. > > But I feel comfortable from a legal stand point with just changing > headers. I think we also need to post some notice publicly about > changing things. But most open source legal advice points to things > being fine if individual contributors retain copyright, or put it in > to some vague PMC. Hrm; so what happens now? Do we need to make a decision of any form? You left two options on the table. Is there anything we can do to that page to make things clear before the meeting? Jody |
From: Cameron S. <cam...@gm...> - 2007-10-18 11:18:41
|
Jody Garnett wrote: > Is there anything we can do to that page to make things clear before the > meeting? > Jody > I'd find it very useful to know: 1. Have developers been asked about assigning copywrite to OSGeo (or another body)? 2. What was asked? 3. What were the responses? 4. If there are negative responses, what were their reasons? Can they be addressed? 5. Were alternatives suggested. Hopefully this should make it clearer as to what options are available and the consequences of each option. -- Cameron Shorter Systems Architect, http://lisasoft.com.au Tel: +61 (0)2 8570 5050 Mob: +61 (0)419 142 254 |
From: Jody G. <jga...@re...> - 2007-10-18 16:22:37
|
Cameron Shorter wrote: > Jody Garnett wrote: >> Is there anything we can do to that page to make things clear before >> the meeting? >> Jody >> > I'd find it very useful to know: > 1. Have developers been asked about assigning copywrite to OSGeo (or > another body)? Yes; but until we know what they need to sign that is all we have done - ask. > 2. What was asked? It was more of a warning; we are working with OSGeo and they may want us to sign something. > 3. What were the responses? Sign what? We are not sure yet. > 4. If there are negative responses, what were their reasons? Can they > be addressed? I am not going to assign (c) over; we can go for coffee and talk about it. > 5. Were alternatives suggested. Dropping the module back to "unsupported" and not including it in the official release. > Hopefully this should make it clearer as to what options are available > and the consequences of each option. Jody |
From: Cameron S. <cam...@gm...> - 2007-10-18 22:39:01
|
Ok, The key message I see for a hasty solution is that we need to have someone define what needs doing as a list of action steps, then review the plan by all stakeholders, adjust issues in the plan as required, then work off the issues. This plan needs to be at a fairly high level. It seems that only a couple of people know all the details and they don't have the authority to resolve the issues. So lets create a concise plan, which can be viewed quickly, and easily identify where the blocking issues are. So would suitable steps be: 0. Create a straw man plan. Jody has started that. 1. Write out a straw man license 2a i. Ask developers if they will commit to signing up to license conditions 2a ii. Action developer concerns. license, go back to 1. 2b i. Ask OSGeo Board / Legal if they accept the license. 2b ii. Refine as required. (One needs to come before the other, probably Geotools first) 3. ... Incorporate above into the plan. Jody Garnett wrote: > Cameron Shorter wrote: >> Jody Garnett wrote: >>> Is there anything we can do to that page to make things clear before >>> the meeting? >>> Jody >>> >> I'd find it very useful to know: >> 1. Have developers been asked about assigning copywrite to OSGeo (or >> another body)? > Yes; but until we know what they need to sign that is all we have done > - ask. >> 2. What was asked? > It was more of a warning; we are working with OSGeo and they may want > us to sign something. >> 3. What were the responses? > Sign what? We are not sure yet. >> 4. If there are negative responses, what were their reasons? Can they >> be addressed? > I am not going to assign (c) over; we can go for coffee and talk about > it. >> 5. Were alternatives suggested. > Dropping the module back to "unsupported" and not including it in the > official release. >> Hopefully this should make it clearer as to what options are >> available and the consequences of each option. > Jody > > -- Cameron Shorter Systems Architect, http://lisasoft.com.au Tel: +61 (0)2 8570 5050 Mob: +61 (0)419 142 254 |
From: Jody G. <jga...@re...> - 2007-10-18 22:46:10
|
Thanks for prodding this along; > The key message I see for a hasty solution is that we need to have > someone define what needs doing as a list of action steps, then review > the plan by all stakeholders, adjust issues in the plan as required, > then work off the issues. I think we have two plans: 1. Origional OSGeo board plan with a contribution agreement for developers to sign, open source license, and .... nothing else? 2. Adrian's plan with a contribution agreement for developers and the osgeo foundation to sign, open source license Perhaps we need to write up two plans and figure out which wolf the PMC / OSGeo board wants feed. > This plan needs to be at a fairly high level. It seems that only a > couple of people know all the details and they don't have the > authority to resolve the issues. So lets create a concise plan, which > can be viewed quickly, and easily identify where the blocking issues are. I think part of the problem is nobody quite has the full picture anymore :-) Adrian or Chris can you hunt down the previous contributor agreements? > So would suitable steps be: > > 0. Create a straw man plan. Jody has started that. > 1. Write out a straw man license I think you mean "Contributors Agreement", license agreement is LGPL and everyone is happy with that. > 2a i. Ask developers if they will commit to signing up to license > conditions > 2a ii. Action developer concerns. license, go back to 1. > 2b i. Ask OSGeo Board / Legal if they accept the license. > 2b ii. Refine as required. > (One needs to come before the other, probably Geotools first) > 3. ... Graduate! Jody |
From: Jody G. <jga...@re...> - 2007-10-17 19:51:34
|
Andrea Aime wrote: > Jody Garnett ha scritto: >> The interesting part is putting this header on our files without >> doing a code contribution agreement. Apparently there is an >> understanding that commit history is good enough; when the board >> reached this understanding is an open question. > > I'm not against doing that work, provided it's wort something? > Are we sure the board thinks it's ok and also sufficient from a legal > point of view? That is what Arnulf Cristl indicated in that IRC chat; which made us sad as it would mean that we wasted a lot of our time due to bad communication. Asking on the incubator list I get the following response from Frank: > All legal advice we have received indicates that a cvs commit action > does not constitute "assignment of copyright". Though we have taken > the position that it is sufficient to declare a change of license > as long as it is done by the copyright holder, or with their agreement. > > So, are you talking about a commit operation for copyright assignment > or for licensing? Cameron chimes in with: > Regarding assignment of copywrite. At the following meeting, it was > argued by Eric Raymond, then passed that Contributor License > Agreements (CLA)s are not required by projects. > http://wiki.osgeo.org/index.php/IncCom_Meeting2 > > I think this is different to geotools' situation which requires > retrospective re-assignment of copywrite to OSGeo Frank clarifies with: > Correct. That discussion was whether a contributor license agreement > is necessary for projects where the original developer retains the > copyright (as in most projects). The answer from Rich Steele was that > having a CLA puts the organization in a stronger legal position but is > not strictly necessary. We decided (within the incubator) not to require > projects (like GDAL for instance) to require a CLA if they do not desire > one. > > Since then we have set forth a policy in the incubator that projects > without a CLA need to at least have committers agree to (not on paper) > to some sort of committer guidelines that make their responsibilities > reasonably clear, such as those at: > > http://wiki.osgeo.org/index.php/Commiter_Responsibilities_Guidelines > > This leaves us with three likely models for contributors in OSGeo: > > 1) copyright holder retains copyright, but signs a strong CLA (MapGuide, > FDO, OpenLayers?) > > 2) copyright holder retains copyright, but only is only expected to > follow > reasonable contribution guidelines to keep IP clean. (GDAL, MapServer) > > 3) copyright holder turns over copyright to the foundation which > requires a signed legal assignment on paper. (proposed for GeoTools) Frank has the impression that contributors maintains (c) is the usual way of things; doing some research shows that is not the case. But if that is the only thing the board can get their head around perhaps that is the step we will have to take? Jody |
From: Martin D. <mar...@ge...> - 2007-10-18 10:47:24
|
Jody Garnett a écrit : > Things to note: > - (c) Open Source Geospatial Foundation > - leaving (c) GeoTools PMC in there even if it has no legal weight What is the advantage of keeping PMC if we list OSGEO (and maybe individuals)? Especially since PMC fluctuate with time. I would be tempted to remove PMC if we put OSGEO. > - previous organizations (or individuals) are still listed If we can not resolve the legal issue otherwise, thats fine. But if we can, my preference would be to sign whatever is needed for giving the copyright to OSGEO, provided that: * OSGEO accept it. * We can be confident that OSGEO will protect the open source nature of the project. In my understanding, the last point was an Adrian's concern a few months ago: he wanted that the legal paper contains some text that explicitly state that GeoTools must stay under an OSI approved license. He wanted to make sure that the sad Mathworld story don't happen to us. If we can not get that, should be put a dual OSGEO & FSF license (if FSF accept)? > - license is LGPL 2.1 Just curious: is LGPL 3.0 an option? Advantages? Inconvenients? Martin |
From: Jody G. <jga...@re...> - 2007-10-18 16:27:42
|
Martin Desruisseaux wrote: > Jody Garnett a écrit : > >> Things to note: >> - (c) Open Source Geospatial Foundation >> - leaving (c) GeoTools PMC in there even if it has no legal weight >> > > What is the advantage of keeping PMC if we list OSGEO (and maybe individuals)? > None; perhaps it would be better to list the individuals? Actually it probably would be better. Good suggestion. The point is to document the (c) between 2004 up until it was handed over to OSGeo. > Especially since PMC fluctuate with time. I would be tempted to remove PMC if we put OSGEO. > Yes; that is the point :-) >> - previous organizations (or individuals) are still listed >> > If we can not resolve the legal issue otherwise, thats fine. But if we can, my > preference would be to sign whatever is needed for giving the copyright to > OSGEO, provided that: > > * OSGEO accept it. > * We can be confident that OSGEO will protect the open source nature of the project. > I am confident in that; but so far there was nothing on the books that forces this issue. They have an organizational charter which we can decide to trust; or we can ask them to assign a document stating down exactly what we expect (I think this was an alternative that Adrian Custer was writing up). > In my understanding, the last point was an Adrian's concern a few months ago: he > wanted that the legal paper contains some text that explicitly state that > GeoTools must stay under an OSI approved license. He wanted to make sure that > the sad Mathworld story don't happen to us. If we can not get that, should be > put a dual OSGEO & FSF license (if FSF accept)? > Well we could also just do the FSF license and then just let OSGeo support us as a promotional organization. This would represent a failure on part of the value they were bringing to the table when we joined. But after two years; most of them deadlocked on this issue; perhaps failure is an option. >> - license is LGPL 2.1 >> > > Just curious: is LGPL 3.0 an option? Advantages? Inconvenients? > I have not looked at it; I did consider GPL + Classpath Exception (because now that Java does it all biz minded people tend to like that option). But we had so much fun with (c) I did not want to talk about license unless we had to. Jody |