Here is some confirmation that Xara isn't expecting copyright assignment.
Bryce
----- Forwarded message from Charles Moir <CharlesM@...1042...> -----
Date: Wed, 12 Oct 2005 21:55:00 +0100 From: Charles Moir <CharlesM@...1042...> To: Mike Hearn <mike@...869...> Cc: bryce@...1... Subject: RE: Linux port
From the discussion on inkscape-devel, a sticking point for some is
the
need for copyright assignment. I'm not sure about the technicalities
of
this, I guess it depends on exactly what Xara want to do and what the Inkscape guys find acceptable. For instance I doubt they'd object to
you
reselling Xara with some proprietary extra bits like Pantone colours, but reselling their code as proprietary code on the components market (as DLLs/OCXs for instance) might be off limits.
What I'd suggest is jumping on the Inkscape list yourselves and trying to hammer out a middle ground - find out what you want to do, what the Inkscape guys find agreeable and see if there's any common ground.
Yep, I've seen this and they are making assumptions that we'll need or want copyright assignment (same feeling on Slashdot from some) - we've never said anything like that. I'll quote from my guy Alex who knows most about license issues;
--------------
Just to clarify, we will *NOT* be requiring copyright assignment. There are two reasons for this:
1. The code will be released GPL. Not a GPL variant, but GPL. This means you can do *WHAT YOU LIKE WITH IT* subject to the normal GPL license conditions (I'm sure you guys are already familiar with them). So saying we are requiring something in order for you to use, modify, or distribute the code is not accurate.
2. We will be maintaining the "official" Xara tree (i.e. the one with the Xara name on it) - anyone else is free to maintain their own tree. That's the beauty of open source. The only sense in which we are planning to "require" anything is that if you want your modification to get into our tree, it has to jump certain hurdles. Some of those are to do with making sure at least one of our trees stays relatively bug-free and the architecture is consistent. Another will be to do with licensing. This is because we want to be able to issue a proprietary version as well. This is for two reasons: firstly, we need to continue to survive as a company. Secondly, there is stuff we simply can't open source - certainly right now - for instance commercially licensed third party components such as the Pantone library. Both of these things, we think, are *good* for the Open Source program. Having an open-source program that is able to produce artwork that is (for instance) easily printed using colour separations and spot colours (even if this involves the use of a commercial version at one stage in the process) is clearly better than being unable to support it at all. Equally, having us able to fund full-time developers means both versions benefit.
3. Lastly, we will NOT be requiring copyright assignment. Assignment means you (the assignor) would lose your rights to us (the assignee). We do not intend to ask for that, and to do so would be totally unfair. What we will be asking for is for some sort of license that will allow us to use your code in our commercial version as well, but will allow the original author to retain *all* their copyright rights. We know the question here is "well why should you guys get any special benefit?". The answer to that is that we're putting a lot of code into the open-source domain. One thing we are thinking about is asking contributors who want their code to get into the official tree to dual license the code GPL/BSD (to the extent they have copyright). That would mean if there is an independent feature someone is contributing, then it's not only us that can use it in a commercial version. We are open to suggestions on how this is achieved.
I'd get involved in the Inkscape list if I had a second. Hopefully when things have quietened down a bit - getting a bit inundated here. But by all means pass these comments on, as from us. I like the idea of dual license GPL/BSD for contributions - allows us to use the code, BSD is hugely popular license, doesn't take anything away from GPL or the rights of the contributor. But as Alex says we're open to suggestions on how to best achieve this. If there's absolutely no way we can incorporate a certain feature in a commercial version, then fine it just wont ever get into our main version. But does that help anyone? We're not taking anything away from the developer, we're not asking for copyright assignment and so I'd hope that wouldn't be a problem.
Of course, this assumes the code bases are similar enough to be mergable. Inkscape uses fairly regular C++ but if Xara is riddled with COM/MFC idioms it might cause issues
Pretty clean OO design and C++ from day one. In fact from day one it was designed to be platform portable. We just never got around to doing any port other than Windows. No COM / MFC in the core at all - that's why and how we got the viewer working so quickly (which is all the main core tree building / rendering pipeline / engine stuff).
For what it's worth I'm also a Wine developer, working for
CodeWeavers.
Ha, I know Jon Parshall, COO of CodeWeavers. Big Xara fan and user (does amazingly detailed illustrations with Xara actually). He's shown me Xara X working on an as yet unreleased Codeweavers Office, so Wine must be getting very near to being able to do this. Obviously not the way we want to go, but would be damn useful to be able to run the Window version side by side with the Linux version just for test and compare purposes.
Anyway thanks for the feedback, Regards, Charles
----- End forwarded message -----
Bryce Harrington <bryce@...1...> writes:
From: Charles Moir <CharlesM@...1042...> Subject: RE: Linux port To: Mike Hearn <mike@...869...>
Of course, this assumes the code bases are similar enough to be mergable. Inkscape uses fairly regular C++ but if Xara is riddled with COM/MFC idioms it might cause issues
Pretty clean OO design and C++ from day one. In fact from day one it was designed to be platform portable. We just never got around to doing any port other than Windows. No COM / MFC in the core at all - that's why and how we got the viewer working so quickly (which is all the main core tree building / rendering pipeline / engine stuff).
FWIW, John Palmieri blogged about Xara's announcement (http://www.martianrock.com/?p=153) and a comment there linked to a nice post about Xara from 1995: http://groups.google.com/group/comp.sys.acorn/msg/8c89e4d2b2fb2f3c?hl=en&...
Cheers, Colin
participants (2)
-
Bryce Harrington
-
Colin Marquardt