karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) (01/28/89)
Please respect the Followup-To:, as this is rapidly leaving behind the basic question of a needed source package. | > uucibg@sw1e.UUCP (3929] Brian Gilstrap) writes: | > + I was recently told that RCS is public domain or copyleft or some | > + such thing. Is this correct? If so, is it available via UUCP | > + somewhere? If so, what does it require as far as OS and such (e.g. | > + BSD versus System V)? | | karl@triceratops.cis.ohio-state.edu (that's me): | > Apparently. Yes, osu-cis. I don't know. | wolfordj@ihlpa.ATT.COM (452is-Wolford) writes: | NO NO NO. It is copyrighted by Walter Tiche at Purdue University. If you | want a email address I will dig it up. It includes some AT&T source, thus | you must have a AT&T source license and he requires you to provide proof | of this. Me-oh-my, I just love having a little excitement added to my day... On first seeing this, I headed off to my archive area, unpacked my rcs.tar.Z, and poked around in the READ*ME* files. Lo and behold, there's an rcs/src/READ_ME which says, in part: "In both cases [of how to get source], we need a signed copy of the non-disclosure form below, executed by the person who handles your Unix licenses." And there follows in the same file this form letter which such a license-responsible person must send to Walter Tichy (note spelling), which says, in part: "2. RCS will not be distributed in any form or by any means without prior written permission by the author, Walter Tichy." With this information in hand, and my attitude deteriorating really rapidly, I wrote something of a flame concerning AT&T-proprietary sources being available for anonymous FTP and the apparent irresponsibility on the part of Purdue for having had such a thing available. In a nearly failed attempt at rationality before flaming, I sent the note to Gene Spafford privately before posting it to get input from anyone around there concerning RCS. He called me on the phone a couple of hours later with some relevant information that I think is in order for all to know: In the past (some years back; I believe Gene said 4), there was in fact some AT&T code in the RCS distribution. Specifically, there was a modified diff(1) which understood the things that RCS needed in order to do its work. Due to the presence of this code, RCS as a whole was in fact restricted to AT&T source license holders. That particular item has been gone from the RCS distribution for some time. Gene wasn't sure for how long, I don't know, and the author is in Germany. As far as anyone at Purdue or anywhere else has been able to determine, nothing in the *current* RCS distribution is restricted with respect to AT&T source licenses. But the package is still copyrighted by Tichy. Therefore (whew), be it resolved that: [1] osu-cis and tut.cis.ohio-state.edu now have the very latest rcs.tar.Z present for uucp and ftp, respectively. [2] There are no evident restrictions on having this source except as per Tichy's requirements. The source is still copyrighted by him, and one is bound to preserve those copyright notices and to give credit where appropriate. However, based on my discussion with Gene and a couple of folks who wrote me mail about this, RCS constitutes shareware at this point, notwithstanding the comment about not distributing source. One must still fill out the form and send it to Tichy (at his Purdue address, I suppose) so that he has a registry of who's using it. [3] The version previously on osu-cis for anonymous UUCP was in an uncertain state. It matches a version currently residing on arthur.cs.purdue.edu dated 18 December 1987 which has been chmod'd to 0, thus inaccessible. I have removed it from my machines, and I do not know with any certainty whatever whether it was in a proprietary state or not. Those still harboring such a copy should check it over carefully for any signs of AT&T notices. Unless you are AT&T-source-licensed, you should destroy such a copy. Pick up the new one via anon ftp to Purdue or anon uucp to osu-cis. Many thanx to Gene for his help in resolving the issue. I'm sure that he'll offer any remaining needed corrections to my understanding above. Thanx also to the other folks who wrote mail (especially David Taylor <taylor@think.com>, who expressed most of what Gene told me). Keeping my cool, --Karl