I agree. To illuminate that point, a friend of mine applied for a NDA with Xerox to develop a Ghostscript filter for some of their inkjet printers. Upon receiving the NDA document, he realized that it specifically barred him from releasing any source code that he would develop based on their documentation. Xerox would not negotiate, so he gave up on it and bought a new printer from a different company.
It is possible that NDAs that do not explicitly outline terms under which you can release source code could be even more dangerous than NDAs that come right out and say "you can't publish source". The latter stops you from going any further right at the beginning, but the former could waste a lot of time and money and ruin your day if the company got the wrong attitude down the road.
When S3 contracted with me to do the Savage driver way back when (3.3.4!), I put explicit language in the proposal stating that the resulting driver would be open source, and would be released to the XFree86 team on a periodic basis. I mentioned it several times during the negotiation process, just to make sure everyone understood what I was saying.
It raises an interesting question, since you can actually glean more information from the source than you can from the confidential chip specs (which are extremely terse), but they had no problem with it.
-- - Tim Roberts, [EMAIL PROTECTED] Providenza & Boekelheide, Inc.
_______________________________________________ Devel mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/devel
