- From: Daniel E. Maddux <dpaladin@hal-pc.org>
- Date: Tue, 30 Oct 2001 23:50:46 -0500 (EST)
- To: www-voice@w3.org
World Wide Web Consortium Voice Browser Working Group www-voice@w3.org Dear W3C Voice Browser Working Group: I am submitting these comments in response to Janet Daly's request for comments posted to SLASHDOT on October 23rd (http://slashdot.org/articles/01/10/23/2041217.shtml). Specifically, I am responding to Ms. Daly's request for comments on the *Patent Licensing* issues arising out of the disclosure and licensing statements made regarding the Voice Browser Working Group ("VBWG") and the VoiceXML 2.0 Working Draft. For the following reasons, I support the W3C's current Patent Policy of Royalty-Free ("RF") Licensing and oppose Reasonable And Non-Discriminatory ("RAND") Licensing of Patents that read on the VoiceXML 2.0 Working Draft. 1. The VBWG should maintain the Status Quo - RF Licensing RF Licensing is the W3C's current Patent Policy (see my Comments to the Patent Policy at http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html). Since RF Licensing is the W3C's current Patent Policy, the VBWG should continue to follow the W3C's current Patent Policy. 2. The issue of RF versus RAND Patent Licensing is premature The issue of RF versus RAND Patent Licensing is premature. The Patent Policy Working Group ("PPWG") has not yet established a new Policy for patent licensing (See http://www.w3.org/2001/ppwg). Since the PPWG has not yet established a new Policy for patent licensing, the VBWG may have to re-charter and re-debate this issue if the PPWG establishes a new Policy on patent licensing. Since the VBWG may have to re-charter and re-debate this issue if the PPWG establishes a new Policy on patent licensing, the issue of RF versus RAND Patent Licensing is premature. Since the issue of RF versus RAND Patent Licensing is premature, the VBWG should wait to see if the PPWG establishes its Policy before changing its patent licensing policy from RF to another patent licensing scheme. 3. RAND Licensing will destroy Open-Source Software on the Web RAND Licensing of patents covering the VoiceXML 2.0 Standard will prevent Open-Source developers from writing software for the Standard. The cost of a license just for IBM's patents is around US$40,000 for the VoiceServer software (See the response to Janet Daly's post, entitled "Who needs royalties?" by fonebone at http://slashdot.org/articles/01/10/23/2041217.shtml) [NOTE that another Poster, pdqlamb "With patents, W3C doesn't need to approve", cites a license fee of about US$10,000 for each corporation. However, pdqlamb does not provide supporting evidence for how he/she arrived at this amount. Therefore, I discount pdqlamb's royalty figure for purposes of this discussion.]. And this price is just for 1 company/W3C Member. According to the VBWG's Voice Browser Patent Statements, 8 members of the W3C (Avaya Communications, Cisco Systems, IBM, Motorola, Nokia, Phillips, Rutgers University, and Telera) have disclosed that they own patents covering the VoiceXML Standard and will require RAND Licensing (See http://www.w3.org/2001/09/voice-disclosures.html). Assuming that the other 8 W3C Members charge a similar amount to license the patents covering their software/technology, and further assuming that none of the other 52 Members (note that France Telecom acquired Orange) choose RAND Licensing, prospective licensees can expect to pay licensing fees of 8 x US$40,000 = US$320,000. Furthermore, this assumption is a best-case scenario. If any of the other Members (like BT, France Telecom, Hitachi, Intel, Lernout & Hauspie, Lucent, Microsoft, Nortel Networks, Sun Microsystems, Unisys, Verizon, or Voxeo) decide to jump on the RAND Licensing bandwagon, then the licensing fees will rise above US$320,000. Very few Open-Source Developers can afford to pay US$320,000, or more, in licensing fees. Since few Open-Source Developers can afford to pay US$320,000, or more, in licensing fees, few Open-Source developers will be able to write software for the VoiceXML 2.0 Standard. Since few Open-Source developers will be able to write software for the VoiceXML 2.0 Standard, RAND Licensing will effectively prevent Open-Source developers from developing software for the VoiceXML 2.0 Standard. RAND Licensing of Patents that cover W3C standards, including the VBWG's VoiceXML 2.0 Standard, will destroy Open-Source Software for the Web. The Patent Policy Working Group has stated that the VBWG is a beta tester for its new draft Patent Policy (See the second paragraph of Section 11. Acknowledgements of the Patent Policy Framework at http://www.w3.org/TR/2001/WD-patent-policy-20010816). Since the VBWG is a beta tester for the PPWG's new draft Patent Policy, other Working Groups will probably follow the VBWG's example in setting their Patent Policies. If the VBWG breaks with the W3C's current RF Licensing Policy and embraces RAND Licensing, then the other Working Groups will probably embrace RAND Licensing as well. Since RAND licensing of patents covering the VoiceXML 2.0 Standard is more expensive than most Open-Source developers can afford to pay, RAND Licensing of other Working Group Standards will probably also be more expensive than most Open-Source developers can afford to pay. Since RAND Licensing of other Working Group Standards will probably also be more expensive than most Open-Source developers can afford to pay, RAND Licensing will effectively prevent Open-Source developers from writing software for any W3C Working Group Standards. Since RAND Licensing will effectively prevent Open-Source developers from writing software for any W3C Working Group Standards, RAND Licensing will destroy Open-Source Software for the Web. Alternatively, RAND Licensing may remove Open-Source Software from the Web through optional discount provisions. For example, Microsoft recently prohibited the use of "publicly available software" as a condition of licensing its Mobile Internet Toolkit (See http://slashdot.org/articles/01/06/21/18102588.shtml). In the case of the VoiceXML 2.0 Standard, Microsoft may offer a discount on the licensing fees for its patents covering the VoiceXML 2.0 Standard provided that the licensee agrees not to use Open-Source Software/"publicly available software" (Microsoft, being Microsoft, may even go further and require licensees who cannot pay the full license fee to use only Microsoft's proprietary software to develop applications for the VoiceXML 2.0 Standard). Since most Open-Source developers cannot afford to pay the above-mentioned license fees, they would have to choose between using closed-source/proprietary software or being denied access to the VoiceXML 2.0 Standard. Since most persons, including Open-Source developers, wish to maintain access to W3C Standards, Microsoft and other patent owners would be coercing these Open-Source developers into using closed-source/proprietary software to access the VoiceXML 2.0 Standard. Since Microsoft and other patent owners would be coercing these Open-Source developers into using closed-source/proprietary software to access the VoiceXML 2.0 Standard, RAND Licensing would remove Open-Source Software from the Web. 4. RAND Licensing will delay or halt the development of W3C Standards RAND Licensing of patents covering W3C Standards will delay or halt the development of W3C Standards. The W3C comprises many Working Groups, each Working Group comprising many Member Corporations and other patent owners (See W3C Home Page at http://www.w3.org). Many Member corporations, like Microsoft, belong to more than 1 Working Group (See http://www.w3.org/2001/09/voice-disclosures.html and http://www.w3.org/XML/Activity). For example, Microsoft belongs to the Voice Browser Working Group and the XML Query Working Group ("XQWG"). As a Member of both WGs, Microsoft could refuse to license essential Patent Claims covering the VoiceXML 2.0 Standard in the VBWG unless the XQWG agrees to include Microsoft's patents in the XQWG's Standard. By refusing to license essential Patent Claims covering the VoiceXML 2.0 Standard in the VBWG unless the XQWG agrees to include Microsoft's patents in the XQWG's Standard, Microsoft could block development of the VoiceXML 2.0 Standard in the VBWG and a Standard in the XQWG. Furthermore, other Member patent owners could make the same threat in other WGs, thereby blocking other (and perhaps all) WGs from developing W3C Standards. Since RAND Licensing allows Members of 1 WG to block development of W3C Standards in every WG to which they belong, RAND Licensing of patents covering W3C Standards will delay or halt the development of W3C Standards. 4. My previous Comments I also reiterate the reasons that I stated in my comments to the Patent Policy Working Group (See my Comments to the Patent Policy Working Group at http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html). 5. Conclusion For the above reasons, I think the Voice Browser Working Group should retain the W3C's current Policy of Royalty-Free Licensing for Patents covering the VoiceXML 2.0 Standard. Furthermore, I think the VBWG should prohibit RAND Licensing in any form for the VoiceXML 2.0 Standard or any other Standard. Sincerely yours, Daniel E. Maddux 4100 Greenbriar Street Number 342 Houston, Texas 77098
Received on Wednesday, 31 October 2001 09:19:21 UTC