Slashdot Mirror
B. Gates Rants About Software Copyrights - in 1980
The following is a transcript of the original recording of Dennis Báthory-Kitsz interviewing Bill Gates. This is a behind-the-scenes look at a programmer who's also a writer interviewing a programmer who's also a businessman. The material is unedited. The author had expected a discussion of philosophy and alternative ways of thinking; but the author was nonplused by Gates's emphasis on economic considerations rather than philosophical or legislative ones, or even the excitement that was a strong motivating force throughout the microcomputer community at the time.
This interview, conducted in March 1980, was one of several dozen written and taped exchanges between Báthory-Kitsz and Bill Gates, John Hersey, Bryan Mumford, Hank Watson, P. T. Wolf, and other software authors, computer club members, publishers, program traders, and general users, as well as Sarah Basbas of the Copyright Office.
The final article appeared in 80 Microcomputing, a magazine dedicated to the Radio Shack's TRS-80 Micro Computer System (later known as the Model I). It was the magazine's cover story, Have the Courts Smashed Software Copyright?, in the fall of 1980.
In the interview, the reference to Datacash vs. JS&A is a decision (79 C 591, September 26, 1979) in Illinois District Court that unequivocally held that "the object phase of a computer program was not a 'copy' within meaning of the Copyright Act of 1909 or common law" and "The Copyright Act of 1976 applies to computer programs in their flow chart, source and assembly phases, but not in their object phase." The decision terrified the software community, and was the reason for this article being prepared. CONTU was the National Commission on New Technological Uses of Copyrighted Works, which held hearings on the validity of copyright as applied to computer software, and issued a report on July 31, 1978 (SuDocs No. 030-002-0143-8). The heart of the problem was human readability, which CONTU Commissioner John Hersey (author of Hiroshima and president of the Authors League of America) found absent from the "machine part" character of object code. However, in the intervening 20 years, copyright has been extended to computer object code and other material in non-human-readable form, and displayed copyright notices are no longer required.
Dennis Báthory-Kitsz, composer and technologist, was - in the halcyon days of small computing - the author of The Custom TRS-80, a best-selling book of hardware and software improvements, and Learning the 6809, a programmed learning text that focused on the Tandy Color Computer. He was president of Green Mountain Micro until its demise in 1986. He presently co-hosts the radio/cyber show Kalvos & Damian's New Music Bazaar, writes and edits for The Transitive Empire, and consults in Web site accessibility for OrbitAccess.
Bill Gates was until recently CEO of Microsoft Corporation. Gates and Paul Allen wrote a BASIC interpreter in ROM for the TRS-80 (Level II, replacing the rudimentary, non-Microsoft Level I BASIC that was offered in 1977) that was both good enough and flawed enough to make the TRS-80 a hobbyist's dream. A generation of programmers and hardware jockeys were raised on its workings, and the venerable Model I-once called the Trash-80-has recently undergone a resurgence of interest along with other old "eight-bitters."
There is a prefatory exchange before recording begins.
TRANSCRIPT
Dennis Báthory-Kitsz: So you do put a copyright notice in your programs that is displayed on the screen?
Bill Gates: Well, we certainly, um...
B-K: The approach is that, because programs are not in fact protected by the copyright law in magnetic medium, I was wondering about your particular approach to it-your recommendations. The National Commission on the, um, what is the title...
G: CONTU.
B-K: Right, you know about it, okay, good. ...has recommended that, and legislation hasn't been introduced yet on the question.
G: There's no, there's nothing that says it either is or isn't. Unless you take Datacash vs. JS&A as a final determination of the question. There's no answer to whether magnetic media is presently protected by the copyright laws.
B-K: Well, on the Section 117, it specifically excludes from...
G: No it does not. It says that this law, the new copyright law, does not change in any way the degree to which the copyright laws cover such items. It leaves it totally open. There's nothing that explicitly, unless I'm mistaken, that closes that question.
B-K: One of the points of CONTU was because it did in fact exclude that. That was what the Commission was trying to approach and get at-revisions to the law that would definitely include such things, including ROMs, and other matters which are not currently covered by the law.
G: Okay, I was involved in the CONTU thing, and let me explain what I understand about that. Specifically in the new copyright law, there is a section that says this law is not intended to change in any way the degree to which copyright laws cover ROMs, magnetic media, diskettes, other forms of software programs, and that Congress will appoint a commission to look into this issue. We're not changing it in any way, shape, or form. So CONTU comes directly from that provision, okay, and they're supposed to recommend to Congress how to treat those things explicitly. There's never been a statement of any kind that says that the copyright laws don't cover magnetic media. There's now Datacash vs. JS&A which says that it's in machine-readable form only, and therefore wouldn't be subject to a copyright, if that's the way you read Datacash vs. JS&A. It's now an Illinois District Court decision. It has a lot of people very upset. But there's cert... As far as I'm aware, there's nothing that eliminates that. I mean, every manufacturer around, Digital Equipment puts copyright notices on, we put copyright notices on. Also, on most of our software packages, we require a non-disclosure agreement from the user before they can use it. B-K: Right, that is an agreement of another nature. However, a person obtaining a copy of yours outside of the agreement, through any source... Now again, you're interpreting it, apparently, from a point of view different from that taken by the Commission, because the Commission, specifically Hersey's dissent...
G: What about Hershey's [sic] dissent? That's something I spent a lot of time on.
B-K: Oh, you did? I had written to him recently, and got a letter in which he specifically analogizes it to the actions of cams and other devices.
G: That's what he thinks!
B-K: Yes, indeed.
G: Hershey was just way off in left field as far as the Commission?s concerned! He takes this thing that we're dehumanizing society by protecting our work, you know, that everybody should be... that somehow if people can feel mechanistic things, then society is a little more human.
B-K: It seemed to me that Hersey's point was - I didn't agree with him - mostly from the viewpoint that he seemed to be trying to protect the works of the traditional author as he viewed them. And I myself write extensive machine language programming, do a lot of that, and sort of resented that point of view personally. But, it seemed to me, again, I'm going from the point of view that the Commission would not have been necessary if the law was clear enough to cover those sorts of things, the format...
G: The law is not clear at all, and the Commission was set up specifically by a provision of the law. I'm talking about... I can go get my file on this stuff, which is about eight inches thick, and read to you that portion of the law.
B-K: Yeah, I have it in front of me at the moment.
G: It does not say that the copyright laws do not cover magnetic media.
B-K: The copyright laws do cover magnetic media, but not computer programming magnetic media.
G: No, it doesn't say that either.
B-K: Okay, what I'm looking for is not a discussion of that so much as your opinion of the current situation in regard to this. How do you feel about it? Or do you...
G: Well, we spend millions of dollars a year creating software programs, and we are protecting those in several ways. There's the trade secret laws where we get non-disclosure, and that's how we handle our source codes and our so-called commercial packages that are high priced. But for our low-cost software, we simply can't do that. Like the things... We have that consumer products division that sells things selling for less than $100. We can't afford to go out there and get non-disclosures on those things. So we rely on the copyright laws.
B-K: You haven't had an occasion to test the law? Your organization, specifically?
G: There is no case testing the law, unless you consider the Datacash vs. JS&A such a case. Everybody relies on unfair competition laws, which are just fine. If you have somebody going out using your work, and selling it, then the unfair competition laws come in, or there's laws for instance in the State of California, that were enacted to protect cassette tape rip-off, and those have been used, and unfair trade practices have been used, but as far as the copyright law, no, nobody's ever...
B-K: What about modifications? Minor modifications to programs and reorganization, and then re-issuance? How do you feel about that? There are rumors, which you probably have more reliable information on, about the copies of machines like the TRS-80 to appear using essentially your writing of Level II. How do you feel about a minor rewrite of all of your work? And how do you think you would approach that if such a machine were in fact to appear?
G: Well, you must be talking about the EACA machine, which is a Hong Kong company that's coming in and licensed our BASIC.
B-K: They have licensed it from you?
G: Certainly.
B-K: Okay, I was not aware...
G: If they didn't license the BASIC... I mean, the unfair trade laws protect us there 100%, you know. That has nothing to do with the copyright thing. There you have somebody who is using another person's work.
B-K: How can it be proven that it is in fact your work?
G: How can it be proven that it's my work? I'll get, you know, any number of experts to testify that it's my work. It's quite simple. It's proof in a court of law.
B-K: What evidence is there within the context of the work that you would consider enough evidence? In other words, we are dealing with a group of machine instructions, which could be, like a bibliography, could be also arranged in an order that would produce a certain series of effects. What is distinctive...
G: You?re talking about a thing that is 16,000 instructions with 256 possibilities for each one. I can certainly prove if it's derived from my work that it's derived from my work. It's just like any form of plagarism [sic] where, you know, you can modify musical tunes. There's nothing novel about this. If you change the page numbers on a book, or if you translate a book, or if you rip off somebody's musical tune, the court has to determine, was it in fact borrowing from the previous work, and in the case of software, they'd rely on expert testimony.
B-K: In other words, from your point of view, there is nothing difficult at all about the question?
G: Well, you have to have exp... You know, if somebody's camouflaged the thing pretty well... That's a question of fact, not of law. The question is, did they borrow from my work, okay? Assume that I can prove that, that I could convince the court that they borrowed from my work, then you have the question of law, what's the law going to do about that. Okay, when somebody's commercializing a machine based on my product and earning money from it, that's unfair competition, when they haven't licensed it from me. That's ignoring the copyright law. The thing the copyright laws are really necessary for is where you have people exchange the software essentially on a free basis with each other, say inside a computer club or, you know, people there are saying, here, take a video tape or things. That's where copyright laws have to come in and say, "Is this a legitimate use?"
B-K: Or if someone were merely to take your material, and from their point of view, a gesture of humanitarianism to users who couldn't afford it, to give it away to them...
G: Just like you go to a bank, and as a gesture of humanitarianism, you take their money and you give it away! That's a gesture of humanitarianism! In society, we don't need to pay... If something's expensive to develop, and somebody's not going to get paid, it won't get developed. So you decide: Do you want software to be written, or not?
B-K: Okay, in such a situation, how would you approach, if, say, major computer club like a Cleveland group or a Chicago group decided they had had it with paying what they considered a high price and started giving away your material publicly? How would you approach that?
G: Well, they put, they put software companies out of business, okay? If the law wasn't going to protect it, there wouldn't be any software written. I mean, it's just like, if somebody... What would an author do if everybody starts giving away their book, or anything?
B-K: There's a clear protection...
G: Take them to court and see if the courts are willing to protect the work that we've done.
B-K: Okay, do you think you?d be successful?
G: Oh, absolutely! I mean, you've got to realize that the courts are basically looking for equity solutions, and when you have somebody creating a work, and putting a lot of effort into that, and you have somebody else who is giving it away, the court is looking for the equity of the situation.
B-K: What piece of law would they base it on?
G: Oh, they'd generally use unfair trade practices, unfair competition.
B-K: But neither of those cases are legitimate here, because there is no profit being made.
G: Well, it's things like the AP wire, where there were people giving away AP wire stuff, and AP wire can't be copyrighted because it's simply news material, and I forget how they handled it in that case, but... Well, they'd probably look into state law to see what they could get. I mean, you know, if you have somebody running advertisements and distributing media, they're probably going to have to charge some money, in which case you can definitely get all those laws into play. Also, if it's something requires non-disclosure, which is true for most of our packages, everything but super low-end stuff and [inaudible] stuff, like Typing Tutor, then we've got a trade secret case against them.
B-K: Once you've gotten past the first step of non-disclosure, you may have a case against the initial person, however...
G: But if a trade secret's released, the people are taking advantage of a trade secret are subject to a penalty. It's not like you can just go in and rip someone off, and they can't figure out who it is, you can use the thing freely. That's not the way trade secrets work.
B-K: If that's the case, then why do you license it?
G: Why do I license it? So I can get coverage under the trade secret law.
B-K: I'm looking for a conflict there somehow.
G: You don't get trade secret protection unless you have a specific agreement with the party not to disclose. But you can have, if you prove it's a trade secret, for anybody who's using it.
B-K: Even those with whom you do not have a specific agreement?
G: That's right. It's a trade secret. If you prove that the thing has never been published, and you've covered it under those agreements, then anybody who is benefiting by it suffers. Otherwise the thing is so ludicrous. What do you think the trade secret... What good would the trade secret laws be?
B-K: In the trade secret laws as I had understood them, that was generally material that was not accessible. The process...
G: No, ours is not accessible except by non-disclosure. That's what trade secrets are for.
B-K: No, that's not what I was saying. I was saying that the material, the process of operation, or one aspect of it was not discernible...
B-K: ...and that it requires some covert act to obtain the balance of information. In the case of your code, the code is all. There is no secret beyond that.
G: I really don't know what you?re saying at all. Our code is the trade secret, and people who are...
B-K: But you're giving it way.
G: No, we're not giving it away, we're selling it. Just like Coca-Cola. The bottling companies make the Coke, okay? The overseas affiliates make the syrup, okay? They license it to those people They're just like people signing our non-disclosure. Or, you know, Dow Chemical has a process for making ethylene. They license it to people. It's a trade secret. Those people who have that process pay.
B-K: That's right, yeah.
G: They sign the non-disclosure, just like people who receive our software sign our non-disclosure. Okay?
B-K: And if the material is taken, modified so that it's still structurally similar, and then re..., and then given away at that point? Do you think that has been clarified?
G: Absolutely! I mean, when you're talking about a chemical process, what is an exact duplicate of the chemical process? There have been plenty of cases where you come in and you have a question of fact: Is this... Did somebody else use somebody's... Did they use proprietary information of this company to come up with this process? And that's a question of fact. You have experts come in. Once again, you've got to separate these things out; don't confuse questions of fact with questions of law. First, you decide if you've used somebody else's work. Then you decided if the law protects it. If it's a trade secret, they've received non-disclosure, fine, the law protects it.
B-K: Why, then, was the Commission necessary?
G: Because there are certain things that are not suitable for covering under, covering with trade secret protection. You don't want to go out for every $14 piece of software you sell, get a non-disclosure agreement from the guy. That's bad news. Okay, you don't want, when you sell a TRS-80, to have to have the guy sign a non-disclosure agreement that he won't screw with the ROM. Okay? For things that are going to have widespread distribution, the copyright laws are certainly more appropriate, and yet, the copyright laws, there's nothing clear in them about how they would handle something that's not human-readable. It's not clear that they don't, I mean, certainly when you have tape... everything in a form is human-readable, just like you can disassemble binary bits, you can take a magnetic tape with voice on it and play it back on a recorder, and certainly cassette tapes are protected.
B-K: Um, okay.
G: It's a matter of degree. If they want to set up some procedures, some clear boundaries, of what was and wasn't protected, because nobody's willing to pay the cost. I couldn't afford that kind of pioneering effort, taking something like that to court. I'd be out of business.
B-K: What... Would you favor, then, the specific extension of the copyright law, in specific language, to cover...
G: Making it clear that the copyright laws covered, yes.
B-K: Yeah, you would.
G: Not an extension! It is not an extension!
B-K: Well, some people do feel that it is...
G: Who?
B-K: ...and Hersey's pretty clear...
G: Who?
B-K: ...that he does feel that way...
G: Who?
B-K: ...so it's not a manner of unanimous...
G: Does he have a court decision or something interpreting the law? Or does he see, or is there some provision in the law? I mean, look at the thing which commissioned CONTU to start with. Okay, I can go pull it out. It doesn't say that the laws don't cover these things. It says that they've been asked to come up with a clear position clarifying the exact procedure. And in particular in the case of databases that's tough. CONTU had a very broad scope.
B-K: Then why, as of four weeks ago, had the Copyright Office not accepted any software in magnetic media for copyright?
G: It doesn't really matter. I mean, when cassette tapes come in, they actually deliver them. They don't deliver a cassette tape on the thing, they deliver like a musical sheet rendition or something like that.
B-K: I'm not talking about the printed version, which is acceptable, but the magnetic version, which is not.
G: Are you telling me music cassette tapes are not protected?
B-K: No, I didn't say music. I was talking about computer programs on cassette tape. Music is specifically protected by the law.
G: Why? It's not in human readable form. It's on a cassette.
B-K: Well, I'm not asking you for an interpretation. I'm stating that as of four weeks ago, they had not yet accepted them. If the law to your mind makes them acceptable, why does the Copyright Office, the Registrar of Copyright, still refuse to accept magnetic media and ROM?
G: For the same reason they don't take music cassettes. They want to see the thing in the most human-readable...
B-K: It's the phonogram copyright and it does in fact copyright the musical version.
G: Well, they want to see the thing in the most human-readable form, so people have submitted tons and tons of source code to the thing to the Copyright Office and received copyright registrations. Just like in the Datacash vs. JS&A.
B-K: The office is specifically excluding magnetic media. They still continue to. I want you to address that question if you would - that of magnetic media. You have said the law does protect the programs, and the specific question of the law was whether it protected them in magnetic media and in ROMs, and whenever it was in the machine.
G: The medium is not really the issue...
B-K: The medium is in fact an issue.
G: The Copyright Office wants to see something they can understand. If you put a computer program on magnetic media, they have a hard time knowing what they are copyrighting. They have no way of comparing things, or doing a search, or anything of that nature. They're really unequipped for it, okay, and they want a clear mandate that they should be equipped for that, and then they'll go out and spend a bunch of money. So they want it, right now, they want to see the thing in listing form.
B-K: And, as contrasted with recorded discs and tapes, the magnetic media and the ROMs they are refusing to accept. So you would favor this extension? This is what I am trying to get at. It is in fact from the Copyright Office's point of view an extension of the law.
G: Well, what the Copyright Office does... You know, they should be equipped to accept it in that form, well we can certainly deliver it to them in other forms. Like initially, they weren't equipped to receive video material in VTR form, or they're not equipped to receive it in video disc format, and they definitely should get a setup to do that.
B-K: But it's definitely copyrightable under the phonogram copyright and magnetic media as long as it is not a digitally rendered piece of information, which they are not accepting...
G: There is tons of... I'm sorry?
B-K: Which, as I say, the digital program they are not accepting, yet they are accepting by contrast a digital recording of audio material. That is an interesting contrast. I had asked them of that, and they said yes, they are accepting digital masters, but they are not accepting computer programs.
G: What if I put my source code on that?
B-K: Yes.
G: I put the characters in perfectly readable human form? They'll accept that.
B-K: They will not.
G: No, the problem is object code vs. source code.
B-K: The problem is a machine-readable code.
G: Well, wait a minute. The VTR is a machine; a cassette tape recorder is a machine.
B-K: They made the differentiation. I haven't made it. They have apparently made it.
G: Okay, in a sense that only a machine can execute it, yeah. They have not accepted object code for copyright as far as I'm aware.
B-K: They have not. That is correct.
G: I know some people who have sent in paper tape object codes and they weren't rejected, but that doesn't really prove anything that they have them sitting in their vaults.
B-K: If they issue the copyright certificate for that piece of material in that form, which they have not done. They have issued it for listings, they have issued it for other forms, but not that which sits in either of those two, and I was trying to get at that. Would you like to make any generalized comments about your experiences with your smaller software, and the violation of any rights that you see that you have there? The smaller pieces, those less than $100 that you were talking about earlier?
G: Well, you know, people... When it's more convenient to get it from your friend than it is to go out and buy the thing, both from a cost and a speed of delivery point of view, there's a great temptation to copy it, and that does take place. Fortunately, there's a great number of people who are honest enough, who realize the effect of that type of stuff, and come and purchase the package. And, you know, as... That's really what determines whether people will come up with software packages.
B-K: Would you like to address some comments to a generalized public, as, I would say, probably the acknowledged leader of the microcomputer software industry?
G: Well, as far as the whole rip-off issue, unfortunately, it gets... The trade is something totally different than what it is. It's simply a matter of... It's not manufacturers trying to rip anybody off or anything like that. There's nobody getting rich writing software that I know of. There are people who would like to stay in business and earn a salary writing packages for these low-cost computers, but... And every time somebody comes up with a scheme for making it difficult to protect, there's a great deal of unhappiness expressed on the part of the user, because they want to be able to make the duplicate copies.
B-K: Yeah, absolutely. I sympathize with that, from my point of view, because I have ruined enough of my original copies of things because I'm terribly klutzy about such things.
G: There's too many people out there who are willing to just exchange the stuff, are willing to go to a club meeting and see that stuff being done without speaking out against it.
B-K: Absolutely. I've received in a survey that I was doing of users from my own personal mailing list, a list, a photocopied list of over 120 pieces of commercial software that one individual was trading, offering for trade, most of which he had gotten in trade. And I was really unhappy about that. I'm trying to be objective about my point of view on the article, but I really hit the ceiling when I saw it. I said to myself, what do I write to this guy, what do I say to this guy to say, you know, "Would you stop? Maybe it doesn't seem like a big deal, but you've got 120 programs, and some of them are phenomenally big programs that the authors, who sweated their asses off for this thing, have not seen any compensation for." But it's pretty extensive. I don't know how much of it you see. You probably see a lot of it, but maybe people are wary of telling you, but it's tremendously extensive.
G: There's a great deal of it. But I, looking at the thing, and with the amount of software we offer, we are the most ripped-off company around in this [inaudible]. It's because we offer a broad range, and we try to offer it for these low-cost computers.
B-K: Absolutely. And there's probably where you're getting the greatest portion of your loss.
G: And we view this thing totally as an experiment. If there isn't enough, if there aren't enough honest people out there to buy the stuff, we'll end it. And we won't... At least, most of our packages we won't put down at the low end.
B-K: Okay. Your experiment has been successful so far, I take it from something you said earlier. You felt that there are enough honest people out there.
G: Yeah, I really think there are. And it's not, it's not overwhelmingly acceptable, but it's at an acceptable level where, as the base of personal computer users grows, and we come up with... I think people will have a growing awareness that it's just taking someone else's work is not the appropriate way to handle it.
B-K: As a related question - you can tell me to stop at any time as it gets toward, or maybe it's past your lunch hour - but as just one related question. I'll probably make it my final one. We've heard a lot of feedback from readers about wishing that Microsoft and Radio Shack in this particular case could possibly make available some more, a lot of detail about the ROM. People feel helpless trying to take it apart bit by bit, spending months doing that so they can essentially write a simple routine to patch into it for their own use. How do you feel about the obligation one reader particularly expressed that the, more details should be made available about, particularly, large-scale works like Level II ROM?
G: Well, if there's a problem with Level II ROM, people can work with Radio Shack. They have their own support team.
B-K: Well, Radio Shack apparently doesn't know very much about it, or is not willing to talk about it. They have not answered...
G: The Level II ROM operates as documented in the manual, and they're perfectly willing to talk about that. As far as whether PMOS was used, and whether the third layer of the Z-80 chip has...
B-K: I'm talking about the software specifically.
G: Why is that an area of such interest? Why don't they demand that the circuit diagrams for the Z-80 chip be included then? All that stuff be documented? I mean, they bought the box for one function, and the price was set to support them.
B-K: Maybe they didn't buy the box for one function.
G: Well, if they got it to get the source code of the ROM, then somebody misled them.
B-K: Don't misapprehend my question. I'm saying that there are a lot of places, and I myself have had the difficulty of wanting to perform a certain transparent function that I could use all the time for my particular orientation, which is as a composer, and I need some special activities which I would like to patch into place. It's taken me an extraordinary amount of time to find how those patch points operated, the timings of them, etc., for my own use. I do not have personally the cash to go out and buy a $10,000 controller. I needed an inexpensive machine onto which I could expand with my main interest, which is composition - which is not computer hardware.
G: Well, a great deal of information about that ROM has been made available.
B-K: Through what sources?
G: Through Radio Shack. They tell you where, they tell you certain things about the entry points and what the device parameters are. Beyond that, do you know what kind of support burden you create by trying to explain to everybody what's in that ROM? Not to mention the fact that that ROM, the source code that's in there is what keeps us in business.
B-K: That's true, but do you see as likely that someone is going to copy that when it is so inexpensive?
G: To copy the ROM?
B-K: Yeah.
G: It isn't inexpensive to copy the ROM.
B-K: What I'm saying is, it's so inexpensive for the device as a whole. Nobody's going to build a TRS-80 from the ground up.
G: That's true. Well, if they want. If they want to know every last bit and byte in the thing and they for some reason which I guess I...
B-K: The reason I'm trying to express is, I have found three or four routines which I've found essential to my use, and it took me a long time to find out how the various interpreter sections worked so that I could patch it in completely transparently, do other things, and go into it without having to completely bypass it, to not use it at all. A lot of the convenience functions I wanted, but I didn't, there were a lot of things it didn't have that I wanted to add to it. A lot of people have expressed to me, why aren't the patch points, that whole row, that whole, virtually, page of patch points explained?...
G: Well, the reason...
B-K: ...That kind of support?
G: We put flexibility into the ROM which isn't documented at all, and which is not sold as a feature on the thing. Radio Shack doesn't have the expertise or the money to go out and explain to their users how to work with that stuff at the machine-language level. The support burden would just be unbelievable. I mean, they would virtually have to hire experts to answer the phone and spend hours educating people about machine language, about the stack, about interrupts, about how it works together, about the problems with hardware interface, and explain it to me and my eight-year-old son. [tape side 1 ends; discussion about clones missed] You know, I assume they're not offering the same type of support that Radio Shack would on the thing. Fixing mistakes, and coming up with new versions, fixing the thing, and also the fact that you have to go and buy the thing from a club or something like that, it's very clear what the rules of use are, and I don't see why Radio Shack should do that. Do you think a lot more people would buy the computer, or that they've misrepresented the computer?
B-K: No, I don't think anyone's said that. I think people are anxious, are very anxious for additional information about what they have. You can order, for example, a manual, a repair and service manual...
G: Yeah, on a hardware level, they've done a good job.
B-K: ...no I mean, for an automobile, for example, you can order a detailed manual about all the aspects of it, how to tune it up, and keep it in good operating condition, buy you cannot...
G: Yeah, that's true. There is a thing called the TRS-80 Technical Manual. From a hardware point of view, they've done a good job.
B-K: That's from a hardware point of view.
G: Well, the software's just... B-K: But there's no software technical manual?
G: Well, there's two issues there. One is the complexity of it, and two is the issue of protecting what we own, and what we do business with. I think, you know, there's a gap there. More information should be made available by Radio Shack, and they would have to come to me and say, "Is this too much?" Because of the way our contracts work, they can't really release anything without my permission. There's a certain amount that could be, that would aid people in their use of the ROM without threatening my proprietary rights. If Radio Shack's emphasis was to do a good job supporting hobbyist-type users, and people who were interested in that sort of thing, they would do that. They'd go out and write a book. As it is, the people who are doing this thing have to be careful, because they tread the line of violating my copyright. Like, oh, [Harvard C.] Pennington coming out with his thing [TRS-80 Disk & Other Mysteries]. He's putting comments in.
B-K: Right, right. There's a TRS... I don't know if you've seen it - the TRS-80 Disassembled Handbook, which is a piece of work in which a person deciphered a large number of your subroutine calls, and provides virtually all of your code in hex, with the exception of a few bytes, which he says that he's taken out to make sure that the person bought the product. How do you react to something like that?
G: He's got our code in hex?
B-K: Your entire code in hex less maybe about two dozen bytes that he took out specifically to say that he didn't want to violate your rights by providing the whole code.
G: Well, he certainly violated our rights!
B-K: Okay. If you're interested, there's a review in the upcoming issue of 80 Microcomputing, and it has the address, and you might want to talk to him, because he feels he has not...
G: Certainly he has. I mean, that's my material. Whose does he think it is? Does he think that he has the right to go out and commercially profit by republishing something that we created? I mean, that's ludicrous! You know, why should he be making money from that? All he did was take our stuff!
B-K: Okay, in the level of your support, then, do you feel that giving detailed information about what you've doing violates your rights?
G: No. No, like the thing with Pennington where he gives comments, that doesn't violate my rights, because he did not include any portion of the ROM itself. You know, it kind of encourages people to run the disassembler, and as long as they use it for their personal use, okay, that's fair use. It might make it easier for someone who had something bad in mind... But that's fine; what Pennington has done is okay. [Later that year Pennington published Microsoft BASIC Decoded with a complete source listing.] But if this guy actually included the contents of the ROM in his book, he's crossed the line. That's our material!
B-K: Okay, I would be interested, if you had any contact with him, I would very much be interested in hearing something further on that, because I thought that it was a little out of line personally.
G: Okay, I have to go.
B-K: Thank you very much.
G: Nice talking to you.
-----------------------
For a truly strange audio experience, check out Dennis's MP3.com page. The piece titled No Money (Lullaby for Bill) is made entirely of audio clips from the above interview.
Or you can download the entire original interview in either RealAudio or MP3 format from this special Geeks in Space page. Be forewarned: it's almost 40 minutes long, and it's a 19-year-old, low-quality recording.
How can you people get trolled by a FUNNY POST!?!?!? In my opinion, it's even a bit overdone, but no, you have to go and REBUKE the post!
I'll summarize my claim that it was a sarcastic post: You can't say this, believe it, and be able to spell, all at the same time. This is a collective flame. Those reacting violently to the aforementioned post are hereby sentenced to get a clue or join the Katz Club for People Who Don't Belong But Want to Participate All The Same And Also Change The Place A Bit Because It's Not What They Expected After All.
BillG sez...
G: Just like you go to a bank, and as a gesture of humanitarianism, you take their money and you give it away! That's a gesture of humanitarianism! In society, we don't need to pay... If something's expensive to develop, and somebody's not going to get paid, it won't get developed. So you decide: Do you want software to be written, or not?
Of course he's wrong about the banking metaphor. Money is unitary and only can increase through interest accruing loans (AFAIK). Software is like fire - it can be freely distributed without lessening the original flame.
He is right about the second comment, in a way, but in the context of freely distributed software, he doesn't understand what "expensive" really means. Not that I blame him for being unable to predict the future or understand a culture he had no hand in.
"Expensive", in terms of open source/bazaar software, is measured in units of *glamour* rather than units of *money*. Projects that are highly glamorous, like an OS, a compiler, or a web server, are built quickly and voluntarily. This is because the reward to a developer on an OSS project is the personal excitement and the renown that accrue to them through their work. But on the other hand, projects that almost anyone would consider tedious - like documentation of the third UI widget to the left three dialogs deep - is "expensive". It won't get done unless there is some other form of reward.
In OSS, that reward seems to be the incentive of converting newbies and defeating MS, but it's clearly not the overpowering drive that's enjoyed by, say, the KDE project.
I would be very interested to see how the future of the free software movement pans out with relation to these unglamorous undertakings. Will they eventually be assimilated into OSS, or will they remain in the purview of money-driven operations like MS?
-konstant
Yes! We are all individuals! I'm not!
-konstant
Yes! We are all individuals! I'm not!
all the proof you need that software innovation and quality was NEVER Gates's watchword, being wholly dumped for the holy $. Not only does he come across as arrogant, but he seems to regard with distaste anything he's not making a profit out of. Jeez, Bill! In those days you NEEDED to share ideas just to build a computing community. His tone when he finds that a piece of code has been shared in hex is quite a nasty thing to hear. It would be deliciously ironic if one of his programmers today started with that little bunch of hex back at the turn of the '80s.
Sounds to me like he had his heart set on a monopoly from the very beginning.
- "How do we do it? Volume!" - The Bursar of Unseen University.
The fact is that code is like art or literature, you spend your time and energy to create something, and you would like it to be protected from some loozer coming along and claiming it as his.
Get off the "I hate Bill" and concentrate on what he is saying. He's saying that he's entitled to protect his work. If someone publishes his work without his permission, he should have legal rights to sue or otherwise get reparations (SP?).
This isn't something that was invented by Bill and Microsoft, it's an idea thats been around for years and years before.
Most interesting to me are Gates' opinions about supporting software and the costs associated with such. I'm kind of brought back to the original story post, which says:
[Gates] talks at length about how no one will ever write software if they can't make money at it.
First off, Gates is talking about commercial software, not the garage software or college project software that frequented this time. Secondly, in the context of the time of the interview, he's absolutely right. I mean, we can look at the Open Source world today and say, "Look, Redhat's making money and giving away their software! So is Debian! Blah blah blah!" but really, they're not making money off the software. They make their money off of support, selling the manuals and consulting with large businesses. Gates mentions the incredible burden that supporting software could be, especially more technical issues. He's absolutely right. The reason it works so well today is because the base-entry level for technical support is so far above what it was back then. Kids routinely know more about the systems they use when they join the workforce than even their bosses do, and entry level tech support is around minimum wage. These types of people were 'experts' before, and commanded reasonable sums of money, not to mention they were rare. Back then, a company who said, "Hey, we'll give away our software and code and charge for support," would've lost money because of a) the more expensive costs of developing software and b) the more expensive costs of support.
You may or may not agree with Bill Gates, but he has an excellent business sense and does a good job of recognizing what is happening in the computer world around him. It's this kind of business attitude that has made him very successful. In this interview, you see a little of that, and especially with issues like support and free code, he clearly understood where money could be made in the industry and where it couldn't. That's the mark of a very successful (future) businessman.
However, there are people in open source who point out all of the things they need or want that MS doesn't provide:
The usual answer from MS or its defenders is that Windows is as good or good enough or that the item isn't necessary. The point that usually goes unstated and shouldn't is really very simple: It's my machine. These are my computing tasks. I say these things are necessary. Attacking my opinions and my best judgement concerning how to fulfill my needs is a quick way to alienate me.
Alienating potential allies is a good way to turn them into opponents. That's no problem when they are your competitors. But when they were potential customers, it isn't good business. Open source OSs for the i386 architecture gave a lot of people like me the option of saying, "Fine, I'll be over here getting some work done." They gave the zealots a rallying point to try to take market share from MS. I'm not in the zealot camp, but every bit of FUD I read pushes me that way.
I am a cynic in some ways. I have eyes and a brain. When someone tells me something that contradicts my experience, I take a look. If I can't find a hole that makes it at least plausible that I am wrong and I find that their version would benefit them, I assume that I am being lied to. Note my choice of wording. When I am told something that in my experience can't possibly be true but would benefit them, I don't assume that the speaker is merely mistaken.
My reason is simple. I hold people to a standard of competence when they demand a change from me. People have a right to be mistaken without a presumption of malice. When they apply their own judgement to me, for their own benefit, I expect them to be right. I don't apply that standard to the neighbor who offers me a cookie with the words, "You'll love this new recipe." I do apply it to the painter who says he can paint my house less expensively than his competitor.
If somebody wants me to buy, use or recommend software, they have to convince me that it can do the job. There are some jobs that Windows can do. All but the most hardcore of zealots will agree to that. I'm not ready to recommend to my mother that she replace Windows with Linux on her machine. There are things that Windows can't do. There is no Windows equivalent to Beowolf that I known of. Windows won't run on a lot of hardware platforms that Linux has been ported to. The middle ground is heavily disputed.
Wanting to be better than the competition isn't enough. Wanting people to believe that you are better than the competition isn't enough. Being better and convincing customers of it is the bottom line. My mind is still open. Windows didn't freeze at 3.1. Linux has evolved since the 1.0 kernel. If MS wants me, I'm out here, but they have to have something to say.
The net will not be what we demand, but what we make it. Build it well.
February 3, 1976
An open Letter to Hobbysts
--------------------------
To me, the most critical thing in the hobby market right now
is the lack of good software courses, books and software itself.
Without good software and an owner who understands programming, a
hobby computer is wasted. Will quality software be written for the
hobby market?
Almost a year ago, Paul Allen and myself, expecting the hobby
market to expand, hired Monte Davidoff and developed Altair BASIC.
Though the initial work took only two months, the three of us have
spent most of the last year documenting, improving and adding fea-
tures to BASIC. Now we have 4K, 8K, EXTENDED, ROM and DISK BASIC.
The values of the computer time we have used exceeds $40,000.
The feedback we have gotten from the hundreds of people who
say they are using BASIC has all been positive. Two surprising
things are apparent, however. 1) Most of these "users" never bought
BASIC (less than 10% of all Altair owners have bought BASIC), and
2) the amount of royalties we have received from sales to hobbysts
make the time spent of Altair BASIC worth less than $2 an hour.
Why is this? As the majority of hobbysts must be aware, most
of you steal your software. Hardware must be paid for, but soft-
ware is something to share. Who cares if the people who worked
on it get paid?
Is this fair? One thing you don't do by stealing software is
get back at MITS for some problem you may have had. MITS doesn't
make money selling software. The royalty paid to us, the manual,
the tape and the overhead make it a break-even operation. One thing
you do do is prevent good software from being written. Who can af-
ford to do proffesional work for nothing? What hobbist can put
3-man years into programming, finding all bugs, documenting his pro-
duct and distribute for free? The fact is, no one besides us has
invested a lot of money in hobby software. We have written 6800
BASIC, and are writing 8080 APL and 6800 APL, but there is very lit-
tle incentive to make this software available to hobbists. Most
directly, the thing you do is theft.
What about the guys who re-sell Altair BASIC, aren't they mak-
ing money on hobby software? Yes, but those who have been reported
to us may lose in the end. They are the ones who give hobbysts a
bad name, and should be kicked out of any club meeting they show up
at.
I would appreciate letters from any one who wants to pay up, or
has a suggestion ot comment. Just write me at 1180 Alvarado SE, #114,
Alburquerque, New Mexico, 87108. Nothing would please me more than
being able to hire ten programmers and deluge the hobby market with
good software.
Bill Gates
General Partner, Micro-Soft