This is the mail archive of the crossgcc@cygnus.com mailing list for the crossgcc project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
On Thu, 8 Jan 1998, Robert J. Brown wrote: > Why does an embedded system impose special considerations for > licensing? If you are concerned about distribution in binary only > form, this is allowed, provided that you give a reference to where the > source may be found. If you make the compression routine a seperate > task and tie it in with a pipeline, then you need not coply with the > GPL for the rest of your system, so what is the problem? Someone else may be able to give a better explanation but let's take the Every discussion I have ever seen of the LGPL on-line says that if you use an LGPL'ed library in your application, then you must provide the proprietary part of your application in a form such that it can be relinked against the LGPL'ed code. This allows the user to modify the support code your application depends upon. If you think about this a while, you will realize that it means that a user can potentially relink your application, break it, and sue you for damages. They are morally at fault and you may win but will end up bankrupt defending yourself. If this perspective is not right, then why does support code like libgcc.a and the gnat run-time come with disclaimers that linking that code in your application does not impose any restrictions on you. LGPL for example. I have assumed for so long that the GPL is inappropriate for libraries and link-in code, that I have forgotten why it is not appropriate. Someone on this group should be able to comment. I am sure others on the list can offer better explanations. If I am wrong, I would be happy to be corrected. RTEMS is not under the vanilla GPL because the team viewed it as burdensome for fielded applications. That is why versions up to 3.6.0 use a BSD-ish style license. The upcoming release uses a license like the gnat run-time and libgcc. > Joel> I dismissed kermit because I remembered that it could not be > Joel> included in Linux distributions because of the licensing. > Joel> That leads me to believe that its license is too restrictive > Joel> for my purposes. > > I think I commented that I did not remember the details here. Is it > that each user of Kermit must download his own copy, or what? I was a bit lazy for not looking this up. I should have been more diligent. The license terms found on http://www.columbia.edu/kermit/aaxcom.html seem clear to me that you should not user kermit without obtaining written permission. Here is an excerpt from their license which I think makes the Linux situation very clear: "Kermit software may not be distributed, remarketed, bundled, embedded, adapted, given away, or otherwise redistributed by commercial enterprises to their customers, clients, or prospective clients without written permission from the Office of Kermit Development and Distribution at Columbia University, which will be granted under the conditions enumerated in this document. Commercial distribution includes, but is not necessarily limited to, bundling of Kermit software with hardware or software products; furnishing Kermit software to institutions, government agencies, or corporations under contract; including Kermit software on CD-ROM distributions of any kind; inclusion of Kermit software by Internet Access Providers in software kits provided to their customers; embedding of Kermit software in industry-specific applications such as medical claims submission packages; or any other arrangement in which Kermit software is furnished to customers, clients, or prospective clients for any purpose. " --joel Joel Sherrill Director of Research & Development joel@OARcorp.com On-Line Applications Research Ask me about RTEMS: a free RTOS Huntsville AL 35805 Support Available (205) 722-9985