bb@wjh12.UUCP (byer) (09/01/84)
<> > Well, I never thought I'd be here, but: > > A friend just got an Altos running a v7 port. I'd sincerely > appreciate anyone's aid in acquiring the current set of v7 > uucp bug fixes, if any kind soul happens to have them and is > willing to let me have them. > > Thanks, > > Dave Ihnat > ihuxx!ignatz The above request raises a question? Doesn't this endeavor require a source license for an Altos, or a sufficiently-rich cross-development environment (which is source-licensed)? ------- This request also prompts a more general PLEA: Would sites who are fortunate enough to have source licenses be willing to provide updated/improved binaries of popular utilities to sites with binary-only licenses (and the same architecture / compatible U*x)?? Yes, I know that this is the responsibility of the vendor, but most vendors have a very insular view of ``working software'' -- if it doesn't core-dump, it works! FOO!!! If there is sufficient support for this idea, could the net be used as a clearinghouse of `who has what for which'? Note: the binaries would be retrieved via direct uucp; only the pointers would be in net.binary.*, along with a description of the feature level. Brent Byer (decvax!genrad!wjh12!bb) "Hey paro, I think he broke the President!"
bass@dmsd.UUCP (09/03/84)
My understanding of the AT&T stand on binary licenses is that EVERY provider of binaries to a single system must pay the proper license fee... IE: Vendor 1 provides UNIX kernel and basic set of utilities pays first royalty fee. Vendor 2 provides C, F77, assembler, ld, and vi pays second fee. Vendor 3 provides UUCP pays third fee. A source site which provides binaries without a binary sublicense agreement and payment of fees is in violation of their license. Please correct me if I'm wrong ... this is the example I used at a USENIX meeting several years back in questions to a WeCo speaker ... their statement of multiple fees was clear ... of course things do change for the better every once and a while ... John Bass
piet@mcvax.UUCP (Piet Beertema) (09/04/84)
<...> >Yes, I know that this is the responsibility of the vendor, but most >vendors have a very insular view of ``working software'' -- if it >doesn't core-dump, it works! FOO!!! ...which means you have to approach the problem from the other end: have it dump core, so the the vendor *has* to look after it. There's a whole arsenal of tools for doing so. Just read net.bugs.uucp..... -- Piet Beertema, CWI, Amsterdam ...{decvax,philabs}!mcvax!piet
bsa@ncoast.UUCP (The WITNESS) (09/05/84)
[gollum :-)] > From: bass@dmsd.UUCP > My understanding of the AT&T stand on binary licenses is that EVERY provider > of binaries to a single system must pay the proper license fee... IE: > > Vendor 1 provides UNIX kernel and basic set of utilities pays first > royalty fee. > > Vendor 2 provides C, F77, assembler, ld, and vi pays second fee. > > Vendor 3 provides UUCP pays third fee. > > A source site which provides binaries without a binary sublicense > agreement and payment of fees is in violation of their license. Strikes me that some of the stuff regarding SOURCES on here has already broken the equivalent source-code rules (EQUIVALENT, not EXISTING; damned if I know what AT&T's got in its head (?) (:-) re: legalities, I'm just trying to draw a parallel). Okay, we provide 'diff' listings for sources. How about a (non-AT&T) patch program, and a protocol for it which can be posted on the net a` la uuencode? You'd probably have to repost uuencode/uudecode to use it, as we, at least, do not have it (our news link does not like to pass along sources, the hack occurs 'long about at cwruecmp (listening, Case? :-)). If LDOS can do it, SURELY a Unix system can! --bsa