Article:
  The Strange Case of the Disappearing Open Source Vendors
Subject:   Essay sidesteps the demonstrable corrosive effects of some licenses
Date:   2002-06-29 15:12:04
From:   jdepner
Response to: Essay sidesteps the demonstrable corrosive effects of some licenses

OK. I didn't really want to get drawn into this but I just can't help myself. Brett, thanks for pointing out the problems inherent in the GPL. This type of license is usually embraced by young idealists (or old lifelong socialists). You hit the nail on the head when you said that a programmer cannot even look at GPL'ed code without risking a lawsuit. This tactic has been employed many times by companies that produce proprietary code. Microsoft is in the process of trying to do so again by opening some of their code (SAMBA programmers close your eyes).


There are two points that I want to make here. First, even though I agree with most of what you've said, I must point out that no company has a "right" to survive. I have seen, in one instance, a college kid writing code that he released for free put a small company out of business. Why? His product was far superior to the commercial offering. His code wasn't GPL'ed but it easily could have been. My point here is that some people have the attitude that ALL free software is bad because "hey, I'm trying to make money here, you can't just give away that code". No company has an inalienable right to make money. If you can't compete, tough.


Second, Richard Stallman is one of the brightest, most perceptive visionaries on the face of the earth. But, in my humble opinion, he's about three sigma west of strange. I really like the fact that Linux is open source because it is a reliable OS, as opposed to Microsoft Windoze. It's great that some people want to write code that they give away. I'll use whatever they want to give me. But thinking that everyone should give away their code isn't very realistic. I keep hoping that there will be some sort of middle ground. Heck, I've even bought Microsoft Office so I could run it on Crossover Office (work's great by the way).


Bottom line - GPL always seemed to me to be a case of someone writing a piece of code that he is either too lazy to sell, doesn't know how to sell, or is too altruistic to sell, saying to the rest of the world "I wrote it and I didn't make money from it, so you can't either". Sort of reminds me of the kid who takes his ball and goes home. But remember, if you're running a company, and you can't make a better tool than something that is GPL'ed, and you can't write your own code without looking at the GPL'ed code then you probably don't need to be in business.

Full Threads Oldest First

Showing messages 1 through 1 of 1.

  • Essay sidesteps the demonstrable corrosive effects of some licenses
    2002-06-29 17:55:00  brettglass [View]

    We actually agree on a fair number of points here. You write:

    > OK. I didn't really want to get drawn into this but I just can't help myself.

    It's an important issue to discuss, especially since O'Reilly and Associates have lent Richard Stallman credibility by giving him a keynote slot in their upcoming conference.

    > Brett, thanks for pointing out the problems inherent in the GPL. This type of license is usually embraced by young idealists (or old lifelong socialists). You hit the nail on the head when you said that a programmer cannot even look at GPL'ed code without risking a lawsuit. This tactic has been employed many times by companies that produce proprietary code.

    Oops.... The word is "commercial." The FSF has been attempting to promote an incorrect definition of the word "proprietary." (A process, method, or protocol is proprietary if only one company can do it and it cannot be reproduced or emulated by competitors. It also denotes incompatibility.)

    > Microsoft is in the process of trying to do so again by opening some of their code (SAMBA programmers close your eyes).

    Yes, it is true that it is not safe to look at code owned by ANY unfriendly party -- be it a rapacious corporation like Microsoft or a rapacious corporation like the FSF -- and then write an equivalent. This is why Compaq and Phoenix, in the early days of the PC, engaged in a process called "clean room reverse engineering" to ensure that the authors of work-alike BIOS code could not be accused of having copied from the original ROM code (which IBM published in its reference manuals). This is why truly free licenses are so valuable. You explicitly ARE allowed to look at the code, and use it, or write anything you want based on what you've learned. The so-called "Free" Software Foundation's code is not free.

    In fact, some of the code that Microsoft publishes for developers imposes fewer barriers. Visual Basic, for example, comes with dozens of modules -- called OCXes -- that perform useful functions. You can use these in your own programs without fear of being forced to distribute your work for free or pay royalties. And Microsoft also provides sample source code you can use to build your own OCXes. You can derive your own OCXes from this code without paying royalties. So, ironically, Microsoft publishes *some* code that has fewer strings attached than the FSF's!

    > There are two points that I want to make here. First, even though I agree with most of what you've said, I must point out that no company has a "right" to survive.

    I never said that. However, to attempt to put it out of business via anti-competitive tactics -- the purpose of the GPL -- is simply wrong. And since the GPL hurts the little guy much more than the big guy, it hurts the software ecosystem by preserving the Microsoft monoculture.

    > I have seen, in one instance, a college kid writing code that he released for free put a small company out of business. Why? His product was far superior to the commercial offering. His code wasn't GPL'ed but it easily could have been. My point here is that some people have the attitude that ALL free software is bad because "hey, I'm trying to make money here, you can't just give away that code". No company has an inalienable right to make money. If you can't compete, tough.

    I'm not saying that companies shouldn't have to compete to succeed. But it's tough to compete with free! If a no-cost product is merely adequate, many people will use it instead of a commercial product *no matter how much better the commercial product is*. Someone who's competing with such a product must offer everything it has and a *lot* more. In many cases, the only way to make this practical (and keep the technology from stagnating at a mediocre level forever) is to let them start with the no-cost product and build on it. If it's truly free, they can.

    > Second, Richard Stallman is one of the brightest, most perceptive visionaries on the face of the earth.

    Very strongly disagree. Richard Stallman is a demagogue who has, for his entire life, single-mindedly pursued a vendetta upon which he embarked due to a childhood trauma. His goal was, and is, to wipe out the "evil" people who destroyed his Nirvana, and all of their kind. He has done this via deception and unethical practices, perpetrated largely upon the young and gullible. He deserves scorn, not respect.

    I wish there were an opportunity, at the convention, to rebut him and/or contest some of Stallman's statements. But O'Reilly has given Stallman the podium; it hasn't placed him on a panel (or in a debate format, as occurred with Craig Mundie last year) where his pronouncements could be so much as questioned. So, he'll have a chance to spread his demagoguery unchallenged. I don't think it's in O'Reilly's interest to deify Stallman, who has stated that he would love to see O'Reilly fail too. Yet, I'll bet they've given Richard, whose FSF has quite a lot of money to send him places, a free room -- while I, like many non-rich hackers, have to hike a mile each way from a shared room at the Travelodge to make the trip affordable.

    > But, in my humble opinion, he's about three sigma west of strange.

    Four. ;-)

    > I really like the fact that Linux is open source

    Linux is GPLed. It's not open source. (The GPL violates the Open Source Definition -- at least two points of it.)

    > because it is a reliable OS, as opposed to Microsoft Windoze.

    To say that something's more reliable than Windows is not saying much! ;-) It's more accurate to say that Linux is more reliable than Windows, but less so than the BSDs.

    > It's great that some people want to write code that they give away. I'll use whatever they want to give me. But thinking that everyone should give away their code isn't very realistic.

    Strongly agree.

    > I keep hoping that there will be some sort of middle ground. Heck, I've even bought Microsoft Office so I could run it on Crossover Office (work's great by the way).

    The middle ground is truly free software. It is ultimately the only thing that will save us from a duopoly: the FSF/GPL complex on the one hand and Microsoft on the other.

    > Bottom line - GPL always seemed to me to be a case of someone writing a piece of code that he is either too lazy to sell, doesn't know how to sell, or is too altruistic to sell, saying to the rest of the world "I wrote it and I didn't make money from it, so you can't either". Sort of reminds me of the kid who takes his ball and goes home.

    It's actually a lot like the cruel children's game that's sometimes called "Monkey in the Middle." The GPL developers can have the code; the users can use the code; but they'll keep it away from the programmer who seeks to make a living as a craftsman.

    > But remember, if you're running a company, and you can't make a better tool than something that is GPL'ed, and you can't write your own code without looking at the GPL'ed code then you probably don't need to be in business.

    Again, the issue is not being "better." Remember that a *mediocre* GPLed product can wipe out the market for a far superior commercial product. Witness GCC: it has destroyed companies which have made much better, much more highly optimizing compilers. You often can't buy a good compiler for a platform where GCC has taken root; you're stuck with the mediocre GCC.

    It's also important to realize that small companies often don't have the funds and time to "reimplement the wheel" before adding the enhancements that will distinguish their products. If they use all their resources just catching up, they'll have none left to spend on the enhancements that they need to try to compete with the GPLed product.

    This is what Microsoft hopes for. Remember what Jim Allchin said about giving away Internet Explorer (at the time a mediocre browser): "We'll put them on a treadmill. We'll cut off their air supply." That's what the GPL does to small companies. And when it does, consumers will be left with no choice but to go to Richard or Bill. And programmers won't be able to enter markets and innovate. A sad state of affairs indeed.