From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-f172.google.com (mail-qk0-f172.google.com [209.85.220.172]) by dpdk.org (Postfix) with ESMTP id 3B54058C8 for ; Thu, 1 Dec 2016 18:46:50 +0100 (CET) Received: by mail-qk0-f172.google.com with SMTP id q130so65251697qke.1 for ; Thu, 01 Dec 2016 09:46:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=jgKdMm0WlCtHy9Q7Bx0Cs5A//56GVVuBhb/R+ZjVx1E=; b=JGGhnQJVPaxx5HSyxQSEqPS3ypEyDjifkeFhw24cwFVsBKag0+/NfMWDbCl+fReV5t 9rRRXz+FdofXyaPu6U4LZ9L8VKYHGDKr0iyH7rZSlpDtQoTbV+s5+P4kubY03WKdMKlW 96gwdgkMxbL7+4EMZM+6yPfFVUeCXxHKDfIROpmSee3PdZeVKxFlVe7VeDo4O/dGT/Z4 kANVgCN/hbg+6eYNRS0F3cDOFY2x0IH1DE70uOti8girXTp7FMWL2jh4237PuUkruQ2Q dIMtY/gXj0cAYdArvCqCnYRVxgFHR4uyAV4jj/iV6SFvdji2dejS7cn8QQeXe1d1nJIc Y9tw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=jgKdMm0WlCtHy9Q7Bx0Cs5A//56GVVuBhb/R+ZjVx1E=; b=UacClQJEcDeiWta2JWHNpGt1gpcVA7OFkBL1Gu9xyl8clYmUZRSUV2CIaSjUHF1G0y 2hTPWbmYD7Jh8c2OF/eviZUpQhUwU7fzuMt0RKjRByAd9Rs6OW9JReO0zxdtMwpiZ/rw GcOpK9/DzSmwM97Qi/Zorzg0dqT3BERQLHNT7LpqiUipLRG3V4wuOiQh3cTlM+6rm2n7 8PKM5b8+x+VF2MwGmC65614Zv9SnBUo1vmQZp2RMmMj53fJ6s84sSGcsvk7VZq2WuUKQ 1eZlGfNzn2d6ywyeAt3RcQpoAs8EmOjPDQXGLBcEJBhHIn3Dwo7yrscodjju/thHxCUw HnZw== X-Gm-Message-State: AKaTC00sFz4r2M7/PEL1mGx0vrr1CrJFKLMthgzJff7KblsyoWdmVTNk6HcxwrqFom39pOoz2Y6dzEOkc6drCg== X-Received: by 10.55.16.134 with SMTP id 6mr39844963qkq.76.1480614409674; Thu, 01 Dec 2016 09:46:49 -0800 (PST) MIME-Version: 1.0 Received: by 10.237.34.202 with HTTP; Thu, 1 Dec 2016 09:46:49 -0800 (PST) In-Reply-To: <26FA93C7ED1EAA44AB77D62FBE1D27BA676277F4@IRSMSX108.ger.corp.intel.com> References: <26FA93C7ED1EAA44AB77D62FBE1D27BA67626AEE@IRSMSX108.ger.corp.intel.com> <26FA93C7ED1EAA44AB77D62FBE1D27BA676277F4@IRSMSX108.ger.corp.intel.com> From: Ed Warnicke Date: Thu, 1 Dec 2016 11:46:49 -0600 Message-ID: To: "O'Driscoll, Tim" Cc: Michael Dolan , "moving@dpdk.org" Content-Type: multipart/alternative; boundary=001a1147bc42f0299d05429c67c8 Subject: Re: [dpdk-moving] Minutes from "Moving DPDK to Linux Foundation" call, November 29th X-BeenThere: moving@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK community structure changes List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 01 Dec 2016 17:46:50 -0000 --001a1147bc42f0299d05429c67c8 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable A major disadvantage to all of the CLA+BSD options in terms of covering patent issues is: most people look to the license for comfort on patent issues, and so the CLA doesn't *really* buy us any comfort for most downstream consumers on those issues. Ed On Thu, Dec 1, 2016 at 11:40 AM, O'Driscoll, Tim wrote: > Thanks Mike. I realise you can=E2=80=99t say too much in public about wha= t is > essentially a legal issue. > > To summarise, these are the options we seem to have: > > 1. Continue with BSD license and DCO: > Advantages: Easy (nothing changes). This combination has worked well for > several years with many companies contributing to the project and deployi= ng > DPDK-based solutions. No CLA required. > Disadvantages: Some Linaro members may not be able to contribute and/or > deploy DPDK-based solutions. > > 2. Use Apache 2 for new contributions: > Advantages: It=E2=80=99s a fairly easy change. Provides patent protection= for new > contributions. No CLA required. > Disadvantages: Doesn=E2=80=99t cover the existing DPDK code so the actual= benefit > of this is very small. > > 3. Use Apache 2 and re-license existing code: > Advantages: Patent protection for everything. No CLA required. > Disadvantage: We need to re-license everything. I suspect that=E2=80=99s = a big > effort and it will be very difficult to get agreement from everybody who'= s > contributed. We would also need to consider DPDK code that=E2=80=99s dual= -licensed. > We have some code that=E2=80=99s dual BSD-GPLv2. IANAL, and I'm far from = an expert > on SW licensing, but I think Apache 2 is not compatible with GPLv2, so th= is > might need to become Apache 2/GPLv3. > > 4. Use BSD and CLA: > Advantages: No license change. Provides patent protection for new > contributions. > Disadvantages: Doesn=E2=80=99t cover the existing DPDK code so the actual= benefit > of this is very small. Need for a CLA is a problem for some contributors > due to the need to get legal approval. Mike expressed concerns below abou= t > the combination of the Apache CLA and BSD license, so we'd need to create > and agree a custom CLA. > > 5. Use BSD and CLA, and have the CLA apply retrospectively to existing > code: > Advantages: No license change. Patent protection for everything. > Disadvantages: Don't know if this is even possible - can a CLA apply > retrospectively to existing code? Need for a CLA is a problem for some > contributors due to the need to get legal approval (presumably an even > bigger problem if the CLA applies retrospectively). Mike expressed concer= ns > below about the combination of the Apache CLA and BSD license, so we'd ne= ed > to create and agree a custom CLA. Same logistical issues as for > re-licensing - we'd need to track down and get agreement from all previou= s > contributors. > > Note that I=E2=80=99m assuming that the combination of Apache 2 and a CLA= isn't an > option because this seems redundant as both include patent protection. > Maybe there are other reasons that would make this a valid combination > though. > > We do need to reach a conclusion on this and move forward. We should aim > to resolve it at next week's meeting, so people should consider their > position in advance of that. My vote would be for option 1. > > > Tim > > > From: Michael Dolan [mailto:mdolan@linuxfoundation.org] > > Sent: Wednesday, November 30, 2016 5:39 PM > > To: O'Driscoll, Tim > > Cc: moving@dpdk.org > > Subject: Re: [dpdk-moving] Minutes from "Moving DPDK to Linux > Foundation" call, November 29th > > > > Hi Tim, sorry I couldn't make it with a LF Board meeting conflict > yesterday. As for 1), most/all of our projects facing this issue decide t= o > go Apache 2. A CLA is less preferably particularly with the BSD license. > Where we do use a CLA on a project it's usually the same as the Apache > CCLA/ICLA and that combined with the BSD license will I'm fairly certain > not achieve what Linaro legal is probably concerned about. > > > > My guess is the members here are 90% or more of the contributors and a > relicensing effort could be done within a reasonable timeframe. The proje= ct > could also start under the LF with all new contributions under the Apache= 2 > license which is compatible with all prior BSD contributions. Or you coul= d > just required Apache 2 on any future contributions and keep the prior BSD > if the relicensing is not agreeable to others. > > > > Just some thoughts on how other projects tackled this question. It woul= d > probably be best if we push any further discussion on this to a small gro= up > of your legal counsel as the various levers have different implications a= nd > I'm uncomfortable continuing this discussion without your counsel being > involved. > > > > Thanks, > > > > Mike > > > --001a1147bc42f0299d05429c67c8 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
A major disadvantage to all of the CLA+BSD options in term= s of covering patent issues is: most people look to the license for comfort= on patent issues, and so the CLA doesn't *really* buy us any comfort f= or most downstream consumers on those issues.

Ed

On Thu, Dec 1= , 2016 at 11:40 AM, O'Driscoll, Tim <tim.odriscoll@intel.com= > wrote:
Thanks Mike. I realise= you can=E2=80=99t say too much in public about what is essentially a legal= issue.

To summarise, these are the options we seem to have:

1. Continue with BSD license and DCO:
Advantages: Easy (nothing changes). This combination has worked well for se= veral years with many companies contributing to the project and deploying D= PDK-based solutions. No CLA required.
Disadvantages: Some Linaro members may not be able to contribute and/or dep= loy DPDK-based solutions.

2. Use Apache 2 for new contributions:
Advantages: It=E2=80=99s a fairly easy change. Provides patent protection f= or new contributions. No CLA required.
Disadvantages: Doesn=E2=80=99t cover the existing DPDK code so the actual b= enefit of this is very small.

3. Use Apache 2 and re-license existing code:
Advantages: Patent protection for everything. No CLA required.
Disadvantage: We need to re-license everything. I suspect that=E2=80=99s a = big effort and it will be very difficult to get agreement from everybody wh= o's contributed. We would also need to consider DPDK code that=E2=80=99= s dual-licensed. We have some code that=E2=80=99s dual BSD-GPLv2. IANAL, an= d I'm far from an expert on SW licensing, but I think Apache 2 is not c= ompatible with GPLv2, so this might need to become Apache 2/GPLv3.

4. Use BSD and CLA:
Advantages: No license change. Provides patent protection for new contribut= ions.
Disadvantages: Doesn=E2=80=99t cover the existing DPDK code so the actual b= enefit of this is very small. Need for a CLA is a problem for some contribu= tors due to the need to get legal approval. Mike expressed concerns below a= bout the combination of the Apache CLA and BSD license, so we'd need to= create and agree a custom CLA.

5. Use BSD and CLA, and have the CLA apply retrospectively to existing code= :
Advantages: No license change. Patent protection for everything.
Disadvantages: Don't know if this is even possible - can a CLA apply re= trospectively to existing code? Need for a CLA is a problem for some contri= butors due to the need to get legal approval (presumably an even bigger pro= blem if the CLA applies retrospectively). Mike expressed concerns below abo= ut the combination of the Apache CLA and BSD license, so we'd need to c= reate and agree a custom CLA. Same logistical issues as for re-licensing - = we'd need to track down and get agreement from all previous contributor= s.

Note that I=E2=80=99m assuming that the combination of Apache 2 and a CLA i= sn't an option because this seems redundant as both include patent prot= ection. Maybe there are other reasons that would make this a valid combinat= ion though.

We do need to reach a conclusion on this and move forward. We should aim to= resolve it at next week's meeting, so people should consider their pos= ition in advance of that. My vote would be for option 1.


Tim

> From: Michael Dolan [mailto:mdolan@linuxfoundation.org]
> Sent: Wednesday, November 30, 2016 5:39 PM
> To: O'Driscoll, Tim <tim.odriscoll@intel.com>
> Cc: moving@dpdk.org
> Subject: Re: [dpdk-moving] Minutes from "Moving DPDK to Linux Fou= ndation" call, November 29th
>
> Hi Tim, sorry I couldn&= #39;t make it with a LF Board meeting conflict yesterday. As for 1), most/a= ll of our projects facing this issue decide to go Apache 2. A CLA is less p= referably particularly with the BSD license. Where we do use a CLA on a pro= ject it's usually the same as the Apache CCLA/ICLA and that combined wi= th the BSD license will I'm fairly certain not achieve what Linaro lega= l is probably concerned about.
>
> My guess is the members here are 90% or more of the contributors and a= relicensing effort could be done within a reasonable timeframe. The projec= t could also start under the LF with all new contributions under the Apache= 2 license which is compatible with all prior BSD contributions. Or you cou= ld just required Apache 2 on any future contributions and keep the prior BS= D if the relicensing is not agreeable to others.
>
> Just some thoughts on how other projects tackled this question. It wou= ld probably be best if we push any further discussion on this to a small gr= oup of your legal counsel as the various levers have different implications= and I'm uncomfortable continuing this discussion without your counsel = being involved.
>
> Thanks,
>
> Mike
>

--001a1147bc42f0299d05429c67c8--