RMS question

Mirian> On the Xemacs vs. GNU Emacs page, in the "FSF Point of Mirian> View", RMS writes: This is old, and may not reflect RMS's current position. AFAIK he has never repudiated it, however. But in another sense it is not GNU software, because we can't use XEmacs in the GNU system: using it would mean paying a price in terms of our ability to enforce the GPL. Some of the people who have worked on XEmacs have not provided, and have not asked other contributors to provide, the legal papers to help us enforce the GPL. I have managed to get legal papers for some parts myself, but most of the XEmacs developers have not helped me get them. Mirian> I'm genuinely confused as to what RMS is talking about... The legal theory is based on the _fact_ that (at least in US law) only the copyright owner or her/his legal agent (the "legal papers" RMS writes about are an assignment of the copyright to the FSF) can actually enforce a copyright (prohibit copying, require destruction of illicit copies, receive damages, etc). The FSF's legal advisor believes that there is significant danger that unless _all_ the copyright owners join in a legal action, the copyright of those who do take action is diluted, and may not be upheld in court without meeting stringent standards of proof of which parts they wrote.[1] As for your confusion, you are in good company. RMS himself is evidently confused. The statement above clearly applies to the Linux kernel (substitute "Linux kernel" for "XEmacs" throughout), yet we know how RMS feels about "GNU/Linux." The majority of the occurances of the regexp 'copyright.*free software foundation' in a random 2.2.10 Linux source tree I have lying around occur in copies of the GPL! It would be interesting to see what, if any, response the FSF would make to a violation of the Linux copyright. (It is even more interesting now that IBM has assigned its S/390 port to the FSF. I wonder if IBM realizes that according to the public statements of the FSF this amounts to abandoning their copyright to the public domain?) Still, "full sharing" is possible and has always been the norm. RMS has always offered all the code of Emacs for our use, and we have of course returned the favor. That's exactly what the GPL is for. It is a minor point that XEmacs has accepted RMS's gift in full, while RMS has not accepted ours, is it not? In fact, there has always been cooperation between the projects at many levels. Many important Emacs subsystems (Gnus, w3, EFS, dired, cc-mode, ...) have the same maintainer and other developers in both projects. These are nowadays much more characteristic of Emacsen than the substrates (there are dozens of Lisp engines, several display engines, garbage collectors galore, and at least a half-dozen implementations of multicharset buffers). It is true that there is some redundant effort involved in developing the substrates, because Emacs Lisp is an idiosyncratic dialect, and the binding of the Lisp-based editor to the display engine and text-handling routines is not as clean as can be done today. But this is not usually a totally bad thing; sometimes our implementation is better, sometimes theirs, and often some third project's is best---and usually both GNU Emacsen benefit as a result. (Albeit there may be some extra "reengineering cost" when the GNU Project adopts our designs, as they occasionally do, but refuses the code.) Finally, RMS's threat to abandon XEmacs to the wolves is empty, as far as I can tell. The FSF-assigned code in XEmacs is typically nearly identical to that in the mainline version. Refusing to protect XEmacs in court could be interpreted as abandoning copyright in most of GNU Emacs, as any future defendent could then argue that the code they borrowed came from XEmacs, and not GNU Emacs. In that case RMS's many public statements about the relationship of XEmacs to GNU Emacs would return to haunt him. And if the FSF refused to defend its copyrights in XEmacs in court, it would receive a severe blow in public esteem, I am sure. Most of the free software community is well aware of the bad blood between RMS and some XEmacs developers (especially in the early days), and it would be hard for many to interpret the FSF's refusal to act on its own copyrights as anything but spite. Mirian> I'm curious, because although I personally prefer XEmacs Mirian> to GNU Emacs, it bothers me that there might be some point Mirian> of licensing It is not an issue of licensing; the licenses are identical. RMS wants to (believes he needs to, if you prefer) _own_[2] all of Emacs, that's all. So, in sum, the full weight of what RMS is talking about is his own refusal to incorporate XEmacs-derived code in the mainline. There is nothing to be done about that, except to hope that the new maintainer of GNU Emacs, Gerd Moellman, will find ways to be more flexible. In any case, many of the most important developers, ie, the "subsystem maintainers" mentioned above, have already taken the extra effort necessary to ensure that their code is assigned to the FSF with simultaneous contribution to GNU and to XEmacs, avoiding the issue of the mainline "borrowing from XEmacs." The bulk of the code is common; many Lisp programs work on both versions without change. As for eliminating the redundant effort of developing the Lisp, garbage collection, multicharset support, and display engines, it is unlikely the projects will be merged. People who work on XEmacs generally believe its code to be technically superior[3] and more maintainable in those areas where there are important differences. Since a merger on RMS's terms would involve throwing those improvements out, and the GNU development organization is by invitation only (implying you really have to commit to a code base you haven't seen if you want to participate in current development), it is certain that XEmacs development will continue separately for the foreseeable future. Footnotes: [1] There is an alternative theory that in fact if any of the copyright holders sued, the burden of proof would be on the defendent. Furthermore, with many copyright holders, although it is not possible to use "class action", the effect of more and more holders joining over time would have similar effect, making it more risky to violate. However, AFAIK no real lawyer has actually signed off on this theory, although people who should have a pretty good idea have espoused it, and I've never heard a real lawyer deny it, either. We should respect the fact that RMS has actually gotten real legal advice on the matter. [2] Once the copyright is assigned, the assignee has almost all the privileges of ownership, including reassignment, etc., except where specifically limited in the assignment contract. [3] Have you ever met a programmer who didn't believe in their own code? :-) -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."