Misters, Let me introduce myself : Guillaume FORTAINE, Engineer in Computer Science. I am currently working on a Virtualization Firmware and I would greatly appreciate to go further into the ideas mentioned in the coreboot GSOC project AVATT [1]. I am definitely aware about the UEFI Forum and especially the Hypervista's effort to implement an UEFI Hypervisor [2]. From my analysis, Hypervista is of particular interest as business case for an SMB trying to innovate around UEFI. Their effort has been a failure and I am not surprised, because three years ago (in 2006), I was the first to post on the edk2 website to kindly ask some hardware support from the UEFI Forum [3]. To quote the reply from an Intel engineer [4]:
"The Edk2 project on www.TianoCore.org does not have silicon enabling code but it does implement UEFI conformant interfaces. It currently builds under Windows, Linux, and Mac OS X. So this code may be helpful to you." The first point is to show you that I followed the UEFI's effort since the beginning and that as a small organization, we tried to contribute to something new. And I didn't receive any support. What is the point of a Platform Initialization software that doesn't provide any hardware ? At least an Intel Customer Reference Board with an initial UEFI port would have been an invaluable help to provide further innovation. Two years later, the situation was the same for Hypervista. To quote [2] : "During our recent search for a UEFI compliant PC motherboard, even Intel®, AMI®, and Phoenix® had difficulty identifying PC motherboards that support UEFI 2.x in June 2008." That's why I have chosen coreboot as my hardware initialization firmware. However, to date, there was no case of large scale adoption of coreboot and Intel gave the main reason. To quote [5]: "And with the GPL issues around intellectual property, hardware vendors and OEM’s will be unlikely to adopt LinuxBIOS going forward." We can see that the main concern for Hardware Vendors to deploy coreboot is not because it is open source, but, because it is under a GPL License. I believe that it could be valuable to find a compromise for letting the big players use coreboot technology. And my conclusion is that a switch to a BSD-style Open Source licence could be a smart move, thus letting Hardware Vendors (CPU, Chipset, OEMs, IHVs) protect innovation. To prove the business case for a BSD-Licensed coreboot, I contacted two prominent hardware vendors from the UEFI Forum. To quote : a) "If you are able to make a clean break with the GPL-based Coreboot for your project to be BSD licensed, there may be elements that we may be interested, or maybe it could be of interest to a special platform. We don't need to open source firmware in that scenario and we can use any of the code for whatever reasons we want to. " b) > "And with the GPL issues around intellectual property, hardware vendors and > OEM’s will be unlikely to adopt LinuxBIOS > going forward." [5] I think that's nonsense. VIA, AMD et al are quite happily contributing to coreboot. Not as much as we might like, perhaps, but they are doing more than X. That's why despite all the work that X engineers are putting into fast-boot work for Moblin, we still see people out there saying "If you want it to boot fast, use AMD or VIA hardware". There is absolutely no reason why a platform firmware couldn't be GPL licensed. There's no real need for people to hide how the memory controller is initialised. It's just baseless paranoia. If there is _anything_ that actually _needs_ to be kept secret, then it'll be crypto (and DRM) stuff -- and if that stuff actually _does_ need to be kept secret then it's a clear indication that the hardware is fundamentally broken anyway. > We can see that the main concern for Hardware Vendors to deploy coreboot is > not because it is open source, but, because it> is under a GPL License. > > That's why we believe that to switch to a BSD-style Open Source licence could > be a smart move, thus letting Hardware > Vendors (CPU, Chipset, OEMs, IHVs) > protect innovation. This is still a good move anyway, I think. People are too entrenched in the idea that the world will end if they actually tell people how to initialise the hardware, which is insane. We'll never cure that brain damage all in one go and get to a fully open implementation; I think it _has_ to be done piecemeal, which a BSD solution allows. We can start off with some binary-only parts, and then pin the nay-sayers down on one at a time, and make them tell us _precisely_ what form the apocalypse will take if we publish the source code to each one. > Our question is fairly simple. Would X consider any interest to coreboot > AVATT under a BSD License ? I cannot speak for X on such matters. But there are certainly individuals who would _like_ the company X to be interested, because we believe it's the right approach. My question is fairly simple : Would people from the coreboot project agree to release the source code under a BSD-style Open Source licence instead of the GPL ? I look forward to your answer, Best Regards, Guillaume FORTAINE [1] http://www.coreboot.org/AVATT[2] http://x86asm.net/articles/uefi-hypervisors-winning-the-race-to-bare-metal/[3] https://edk2.tianocore.org/ds/viewMessage.do?dsForumId=139&dsMessageId=13751[4] https://edk2.tianocore.org/ds/viewMessage.do?dsForumId=139&dsMessageId=13804[5] http://www.uefi.org/learning_center/A_Tale_of_Two_Standards.pdf _________________________________________________________________ Windows Live Hotmail: Your friends can get your Facebook updates, right from Hotmail®. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_4:092009 -- coreboot mailing list: [email protected] http://www.coreboot.org/mailman/listinfo/coreboot
