Thread: Re: pgAdmin licence
Dave Page wrote: > This email has been sent to you because our records indicate that at > some point in the past you have contributed to the pgAdmin III project > (http://www.pgadmin.org). It this is not the case, please let me know as > soon as possible. > > Since it was first released, pgAdmin III has been distributed under > version 1.0 of the Artistic licence. I am writing to request your > permission to distribute the code you have submitted under version 2.0 > of the Artistic licence, and any future versions that might be published. > > There are a number of reasons why we would like to change the licence > used by pgAdmin (per Tom Callaway of Redhat): > > - The FSF says it isn't free. They say that the text is vague, and that > it is open to misinterpretation. > > - The perl community agrees with this assessment. They went so far as > to rewrite the Artistic license to resolve all the identified problems > (see http://www.perlfoundation.org/artistic_license_2_0). > > - The Artistic License (1.0) recently went to trial in the US and lost. > The judge interpreted it in a very negative way, and that worries me. > (http://lawandlifesiliconvalley.blogspot.com/2007/08/new-open-source-legal-decision-jacobsen.html) > > I'm sure you will agree that it would be sensible for us to use the > newer version of the licence in the future and would therefore > appreciate it if you could reply to this email as soon as possible > stating either: > > * I agree for all contributions I have made to the pgAdmin III project > to be distributed under version 2.0 or any future versions of the > Artistic licence. > > or: > > * I do not agree to any licence change for contributions I have made to > the pgAdmin III project. > > Dave, for me the worst case already happened last year, when I had to fork my own code to continue my job. As a consequence, I won't give away control over sources any more for tools I release to the public, such as www.pse-consulting.de/pg/pgImport.zip. I do NOT agree to change licensing on my prior code. Regards, Andreas
Andreas Pflug wrote: > for me the worst case already happened last year, when I had to fork my > own code to continue my job. As a consequence, I won't give away control > over sources any more for tools I release to the public, such as > www.pse-consulting.de/pg/pgImport.zip. I find that an incredible claim. You stopped working on pgAdmin because a majority of the developers at the time thought there were tangible benefits from the use of wxGrid in the Query Tool over wxListCtrl. You couldn't produce a single provable technical reason why that was the case, except to say you didn't think it would work properly. Since then we haven't had a single bug reported that could be attributed to the use of wxGrid. In the UK we refer to that as throwing the toys out of the pram. I assume from what you say about your job however was that the real reason you didn't want those changes made was because you relied on the existing code for some extensions you had written for your own private use - much as you once admitted was the case for some of the base/top class splits that you implemented. No matter who the developer, that *can never* be a factor in any design decision made in any self-respecting Open Source project. > I do NOT agree to change licensing on my prior code. So noted. Thank you for replying. Regards, Dave.
Dave Page wrote: > Andreas Pflug wrote: > >> for me the worst case already happened last year, when I had to fork my >> own code to continue my job. As a consequence, I won't give away control >> over sources any more for tools I release to the public, such as >> www.pse-consulting.de/pg/pgImport.zip. >> > > I find that an incredible claim. You stopped working on pgAdmin because > a majority of the developers at the time thought there were tangible > benefits from the use of wxGrid in the Query Tool over wxListCtrl. No. You removed VC6 support, which was absolutely vital for me, without ever asking whether that was viable. Since my development was tightly integrated, I wasn't able to switch one of projects separately (at least in the timeframe available). > You > couldn't produce a single provable technical reason why that was the > case, except to say you didn't think it would work properly. Since then > we haven't had a single bug reported that could be attributed to the use > of wxGrid. In the UK we refer to that as throwing the toys out of the pram. > > I assume from what you say about your job however was that the real > reason you didn't want those changes made was because you relied on the > existing code for some extensions you had written for your own private > use Yes, but not the query tool nor any of its helper classes. The base classes however are reused by several other projects, including pgImport. Regards, Andreas
Andreas Pflug wrote: > No. You removed VC6 support, which was absolutely vital for me, without > ever asking whether that was viable. Since my development was tightly > integrated, I wasn't able to switch one of projects separately (at least > in the timeframe available). Oh, I'm sorry - I thought it was the wxGrid thing you objected to. WRT to moving to VC++ 2005, I *did* ask your opinion (and even stated 'and drop VC++6 support'), and the only objection you raised was that you thought we'd have to link with the .NET runtimes which I investigated and found was not the case. The thread is here: http://archives.postgresql.org/pgadmin-hackers/2006-04/msg00054.php Regardless, if you needed to retain VC++ 6.0 support for compatibility with your private work, surely it would have been infinitely easier to just maintain your own project file, rather than forking and having to duplicate all the effort to keep up to date with new PostgreSQL releases? > Yes, but not the query tool nor any of its helper classes. The base > classes however are reused by several other projects, including pgImport. As far as I can see, that puts you in breach of our licence as the distribution you posted a link to on your PSE Consulting website (www.pse-consulting.de/pg/pgImport.zip) includes *only* a zipped executable and no sign of any of the requirements of the pgAdmin licence. Much of the base class code in pgAdmin was written or modified by me (particularly the connection and recordset classes) and therefore cannot be claimed as your own original work. Please rectify this oversight immediately. Regards, Dave
Dave Page wrote: > Andreas Pflug wrote: > > >> No. You removed VC6 support, which was absolutely vital for me, without >> ever asking whether that was viable. Since my development was tightly >> integrated, I wasn't able to switch one of projects separately (at least >> in the timeframe available). >> > > Oh, I'm sorry - I thought it was the wxGrid thing you objected to. > > WRT to moving to VC++ 2005, I *did* ask your opinion (and even stated > 'and drop VC++6 support'), and the only objection you raised was that > you thought we'd have to link with the .NET runtimes which I > investigated and found was not the case. > > The thread is here: > http://archives.postgresql.org/pgadmin-hackers/2006-04/msg00054.php > > Regardless, if you needed to retain VC++ 6.0 support for compatibility > with your private work, surely it would have been infinitely easier to > just maintain your own project file, rather than forking and having to > duplicate all the effort to keep up to date with new PostgreSQL releases? > > >> Yes, but not the query tool nor any of its helper classes. The base >> classes however are reused by several other projects, including pgImport. >> > > As far as I can see, that puts you in breach of our licence as the > distribution you posted a link to on your PSE Consulting website > (www.pse-consulting.de/pg/pgImport.zip) includes *only* a zipped > executable and no sign of any of the requirements of the pgAdmin > licence. Much of the base class code in pgAdmin was written or modified > by me (particularly the connection and recordset classes) and therefore > cannot be claimed as your own original work. > I don't think so. It was me who invented the db base classes, besides they have undergone some major rewrite (ever heard of pgSetIterator?) I checked quite thoroughly, including the icon (which I'd replace if I had another one at hand). Regards, Andreas
Andreas Pflug wrote: >> > I don't think so. It was me who invented the db base classes, besides > they have undergone some major rewrite (ever heard of pgSetIterator?) > I checked quite thoroughly, including the icon (which I'd replace if I > had another one at hand). You invented the abstraction, but taking pgConnBase as an example, pgConnBase::pgConnBase, pgConnBase::ExecuteVoid, pgConnBase::ExecuteScalar, pgConnBase::ExecuteSet were all copied from pgConn and were the key member functions originally written by Mark Yeatman and myself. Similarly, pgSetBase was largely ripped from pgSet, base.cpp includes numerous functions from misc.cpp, appbase.cpp includes a small amount of code from pgAdmin3.cpp. I can easily trace back significant amounts of code in those base classes to times before you submitted *any* patches, to commits by Mark and I (as can anyone will to spend a few minutes at http://svn.pgadmin.org/). The only base class that originated from you was factory.cpp, and even that contains code written by Magnus Hagander and I committed long before you left the project. I do not for one minute deny that you also have contributed significant code to those classes, but the fact remains all bar the factory code is derived from the work of others, no matter how rewritten it has become in your work. Given that on your PSE Consulting website at http://www.pse-consulting.de/os.html you speak of how you appreciate the Open Source concept, I find it incredible that you can have such disregard for the hard work of others by ignoring the very basic principles of Open Source. I'm dismayed that someone I considered a friend and held in high regard could act in such a disgraceful manner. Regards, Dave.
Dave, since pgImport was originally written to import from file to MSSQL, with pgsql support added later (and pgConnection/pgSet never supported COPY), you're trying to see stuff that's not there (nor is factory.cpp). Regards, Andreas