On Fri, 15 Oct 2004, Paul Vixie wrote: > > ... notwithstanding, how can a specification be considered a standard > > if over half of the operators on the planet refuse to deploy it > > because of patent/licence issues. > > i can't understand why this matters. if ietf were to change its policies > so that only "open technology" was allowed in the standards, there would > still be patent fights (both from submarines and ships-in-the-night). Well, given that Mr. Vixie told me that he is a supporter of software patents, and that we know he only supports BSD-style copyrights which can be taken closed source, and is someone who has claimed trademarks on other peoples software, it is more obvious why he doesn't understand 'why it matters'. So I'll explain why it matters. There is a pro-patent/closed-source point of view, and there is an anti-patent/open-source point of view. These sides are not difficult to understand. But they take some effort to reconcile into cooperation on standards. Why should they be reconciled? Because there are a lot of constituencies in the IETF: There are many large, closed-source, pro-patent vendors that we want to have participate in the IETF. There are also many groups that implementing anti-patent, open-source technology under a variety of copyrights. The particulars of the copyright vary from group to group, and reflect the goals and ideals of each group. There is no consensus on the "one-great-copyright", even though each group probably thinks theirs is best. In order to obtain benefits of standardization and interoperablilty, the IETF (and other standards organizations) have made some compromises: Patent claims must be disclosed early, and there is no requirement that implementation of protocols be made free or open-source. There is no obstacle to open-source either. Each working group is able to evaluate and select technology based on many factors, including whether the technology is subject to patents. However, patents need not stand in the way of standardization if the working group decides to adopt the technology. Patents were a significant obstacle for the IETF-MX working group. They were not an obstacle for certain PPP RFCs, and perhaps other RFCs. The ITU has similar policies and some ITU standards are subject to patents. Usually, the assurance of fair and universal licensing was an important consideration. It is important not to adopt a knee-jerk anti-patent position. That this is important can be seen: It is not just pro-patent organizations that may obtain patents. Anti-patent organizations may find it necessary to patent technology in order to ensure that the technology remains free. Like many other areas, knee-jerk, absolutist responses are wrong, and a more nuanced approach is necessary. Similarly, Open-Source organizations such as GNU and others depend on the copyright law to keep open-source remaining open. If the copyright law were weakened, then GPL code would be vulnerable to being stolen and made closed. So it is important that copyright remain. When the open-source tide really turns, and the best quality source code and technology is free, then it will be subject to theft of the sort where it is made improperly not-free. Then it will be the open-source community that is trying to enforce the copyright and possibly even patent law. That is why alterations must be careful. However, the tide hasn't yet turned. But even after the tide does turn there will still be many players who will still pursue closed-source, patented solutions. We will still need to accomodate them for sometime even as a minority. So, the IETF will probably be the last organization to be purely open-source. -- Av8 Internet Prepared to pay a premium for better service? www.av8.net faster, more reliable, better service 617 344 9000 _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf