Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!zaphod.mps.ohio-state.edu!rpi!uwm.edu!ux1.cso.uiuc.edu!tank!gargoyle!ddsw1!karl From: karl@ddsw1.MCS.COM (Karl Denninger) Newsgroups: comp.sources.d Subject: Re: What does "free" mean, eh? (Re: Scareware) Summary: You have a short memory. Message-ID: <1990Mar12.155911.8245@ddsw1.MCS.COM> Date: 12 Mar 90 15:59:11 GMT References: <14010@s.ms.uky.edu> <125816@midas.UUCP> <635@magnus.Hotline.Com> <34812@watmath.waterloo.edu> <10612@hoptoad.uucp> <796.25f66867@ccvax.ucd.ie> <1142@mtxinu.UUCP> <1990Mar11.192604.7216@ddsw1.MCS.COM> < Reply-To: karl@mcs.MCS.COM (Karl Denninger) Followup-To: alt.flame Organization: Macro Computer Solutions, Inc. - Mundelein, IL Lines: 177 Note: followups redirected to alt.flame (yes, I warn you now) In article <1150@mtxinu.UUCP> frk@mtxinu.UUCP (Frank Korzeniewski) writes: >In article <1990Mar11.192604.7216@ddsw1.MCS.COM> karl@mcs.MCS.COM (Karl Denninger) writes: > >Ah, Mr. Denninger, we cross swords again. Well, lets have at it. How about discussing the issues rather than dueling? >#Let's stick to the issues -- the Gnu license in this case. I have heard >#several reasoned voices saying that it is in many ways a "virus", in that it >#tends to infest code that comes in contact with it. ># > >I find the examples that you and peter have chosen (of getopt and bison) >to be somewhat contrived. Setting aside the issue of including GNU >source into your programs for the moment, we are left with the following: > >The GNU C compiler has only the restriction that the inclusion of the runtime >library in your program then subjects your program to the GNU license. >This library consists of a handfull of machine dependant math routines, >and can be rewritten in a day. So, in effect, the GNU C compiler has >no real restrictions on its use. For now. What happens when there is a real libc.a? Not that it matters; the solution is simply not to develop for GNU C if you want to keep your code free of external encumberances. >So getopt and bison are really special cases in that the generate code >to be included or are themselves included in your programs. If you >look around you, the streets are almost crawling with parser generators, >that you can purchase for little money. It is not a loss to not have >access to bison. The exists a public domain getopt in the net source >archives. Getopt is likewise expendable. So why does GNU have the restrictions at all? >So the "nasty" virus-like qualities you and peter attribute to GNU >just do not really exist for the use of the GNU programs. > >Including GNU source on the other hand, does subject your code to the >provisions os the GNU license. I see nothing wrong with this. In our >legal system of property rights, you have to put restrictions on in order >to protect your work. That is just how life is in our case. Yep. My complaint is that people are >not aware< of the ramifications of all of this until after the fact, and that the GNU people have promoted their software as "Free" when in reality it is a way to make other people's software "Free" -- by their definition. As in the "Free Software Foundation". It sounds awfully contrived. As I said before, but you carefully omitted from your followup, I do not object to the FSF itself, or the code, or the restrictions themselves. I OBJECT TO MISLEADING PEOPLE WITH TERMINOLOGY. It is not GNU that I object to, or their code. I am not Peter. I object to misleading statements or innuendo. Is that such a terrible thing? Now we have that straight, yes? >#My response is not to improve GNU packages. Those we find useful, we >#compile up and use. Support and enhance them? No bloody way, since we >#then HAVE TO GIVE AWAY OUR ENHANCEMENTS TO ANYONE WHO ASKS. Support is >#also in this picture, since support often entails bug fixes, modifications >#for specific user needs, and enhancements -- all of which GNU expects us >#to give away. The only exceptions are those few packages which we make >#trivial changes to, and don't mind giving away the enhancements. Once we >#put more than an hour or so into the changes, we stop. Simple enough. > >You object to giving away your enhancements to something you acquired >for nothing. I find that bizarre. If it were not available you would >have nothing to enhance. I have seen plenty of your postings on the net >criticizing SCO and Interactive for bugs in their software. They do not >provide you the source, so you cannot fix or enhance their products. >You have made all of us aware of your displeasure in this case. So what? ISC at least has been very good about fixing those bugs, and SCO has been too with their Xenix product. Let me ask you -- what is the response when there is a bug in a GNU product? Is "fix it yourself" an option, even for those with source, in a commercial production environment? Is the response "hire a programmer"? There went the "free" in the Free Software Foundation..... Do you see me flaming about GNU product bugs? Nope. Why not? We didn't pay anything for them or the code that they contain. Why would I complain about the bugs? You see, we can fix them, and in addition I had no expectation of quality, since the code was >free<. Do you find that bizarre? >#Tell me -- let's say you could get GNU C freely, OR you could buy a GNU >#C-based compiler for $500 from some company, but the enhanced version could >#not be distributed. Why object to that? If you don't like the fee, get the >#free version! Of course, the free version doesn't include the efforts of >#the people who enhanced it later on. Such is the price of the >#enhancements; you're not being forced to buy them! > >But this is what we have now. You can buy compilers like GNU (without source), >and you cannot distribute them. GNU is just providing the other option. >What is YOUR objection? GNU now gives you access to the numerous enhancements >that have been added by others, and people continue to enhance it. The >quality of the current GNU programs exceeds what you can get commercialy, and >I can only see that situation continuing. In the future I see GNU will >outpace commercial products in quality and functionality. I can't say that in many cases the quality of the GNU products exceed that of those which are commercial, especially if you start counting security holes (ie: the problem with GNU-EMACS which allows anyone to become root in a few minutes, no I will not elaborate further). I suppose that if you don't count security as important you might have a point.... You feel that the quality of GNU software exceeds that which is available for a price. I understand your opinion.... but you are omitting the fact that commercial software houses, which by definition are in business to make a profit, have slightly important things like a operational kernel, all the standard Unix utilities, etc..... and GNU does not at this time. They also have "load and go" distributions for different machines; an important thing if you want to have other than "hackers" use the stuff. I'm sure funding has something to do with that; their functionality exceeds GNUs by a rather large margin when taken as the sum of the pieces rather than each individual piece. GNU has been around for quite a while. Where's the kernel that has been talked about? Where's the packaged implementation? How are you going to convince >business< to use it? The cost of a Unix binary license from ISC, DEC, or anyone else (or even a source license from AT&T!) pales beside the expense of keeping a high-caliber kernel and utility programmer on-staff to port, maintain and enhance the GNU packages! As I said before, my objection is not with GNU packages, or even with the FSF itself. Nor is it with Richard Stallman. It's with misleading people into thinking that Copylefted code is truly free. It most certainly is not free -- in the truest sense of the word. It doesn't come with a pricetag, true, but it does come with all kinds of encumberances and costs which lurk waiting for the unsuspecting. The code which I want to be freely available, without encumberances, I do not Copyleft. I Copyright it with specific conditions -- usually with the conditions that the code not be sold, my notices and name be left in, and disclaim warranties. The code which I do this to is more "free" than that covered under the GPL. Using or distributing it doesn't obligate anyone to do anything. >The loss of any >enhancements by people turned off to the GNU license is just a myth. You >would not be able to get them free in any case, so who cares about them. Really? Is it a myth? How do you back up that statement? Why don't you state it as your opinion instead! I'd like you see your evidence, since you state this as an axiom. >#(Donning asbestos flame-proof suit!) ># >#-- >#Karl Denninger (karl@ddsw1.MCS.COM, !ddsw1!karl) >#Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910] >#Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price" > >Nice suit. I remember a hue and cry raised about you signature a year ago. >After complaints about you advertising there, you removed the ad. I >notice that the ad is back. Perhaps you have a short memory. We don't. Frank, how soon you forget. I took a survey on the signature line. It was overwhelmingly in support of leaving it alone (over 80% in favor; I still have the responses here). Thus I did exactly that -- the line was never removed in the first place. Sheesh. Talk about off the subject! Perhaps it is time to redirect followups to alt.flame.... -- Karl Denninger (karl@ddsw1.MCS.COM, !ddsw1!karl) Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"