| |||||||||||
|
| |||||||||||
ODSL awareness campaign misguided?ODSL awareness campaign misguided?Posted Nov 27, 2003 5:09 UTC (Thu) by JoeBuck (subscriber, #2330)In reply to: ODSL awareness campaign misguided? by walterbyrd Parent article: OSDL Launches Linux Kernel Awareness Initiative
What people don't seem to get is that the free software world is actually more careful about this kind of thing than proprietary software companies are. In my career in proprietary software development, I have repeatedly found that some developer grabbed some code out of a book or off the net and incorporated it into a product, naively thinking that anything he or she could find source code for was "public domain". When such things have been discovered, we worked to remove the code, or in at least one case investigated negotiating a license to use the code legitimately, but I fear that some companies are less ethical than those I've worked for, and ask the question "will we be caught?". You'd be surprised how many programmers think that the code in books like "Numerical Recipes in C" is fair game to use. In most cases, it's all proprietary code and almost any use needs to be paid for. OSDL would be better off turning the question around and asking what the process is for proprietary software companies in assuring that they own all the code they claim to own. In many cases the process seems to be "no one can see our source code so we won't get caught". Example: SCO recently demonstrated that they are in violation of the BSD license with respect to the Berkeley packet filter code.
(Log in to post comments)
Non-free code in books... Posted Nov 27, 2003 17:41 UTC (Thu) by dps (subscriber, #5725) [Link] It always wise to read the bits of books about the licence for any code in them. Numerical Recipies in C (and other languages) looks nice but the licence is too drastic---most people in numerical analysis do *not* have a copy so anything they produce could not have been copied from there. At least one person has been sued for simply transfering his code, containing stuff from numerical recipies, to a supercomputer. This made the code avaialable to someone else, contrary to the numerical recipies licence. Other bits of proprietry software are just as bad, albeit in different ways. Prohibition of "unahtorised" benchamrks and using your their program to check the results of yours is not unknown. The people behind one such effort are known to sue people at will. I will not name any names, lest their lawyers read LWN. This sorts of thing makes me appreciate free software a whole lot more... Could be the makings of FUD about using commercial software here?
Numerical Recipes Posted Nov 27, 2003 20:30 UTC (Thu) by andrel (guest, #5166) [Link] Not only the license to <it>Numerical Recipes in C</it> is horrible. The code itself stinks. It is ugly, inefficient, and doesn't handle corner cases well. Existing free libraries (e.g. ATLAS, LAPACK, FFTW, GSL) do a much better job.
|
|||||||||||