[MPEG-OTSPEC] MATH Encumbrance

Sat Aug 22 21:37:41 CEST 2020

You can find lots of info regarding ISO policies here:

https://www.iso.org/resources.html

(Btw, as this list was created by an ISO technical body for purposes of its technical work, it’s my understanding that all applicable policies would apply to this list.)

For instance, wrt copyright, among other details:

“If you offer such content, you undertake to declare this to ISO or ISO/IEC, identify the name of the copyright holder and assist ISO or ISO/IEC in obtaining appropriate permission to i) share the content in the standards development process, ii) to publish the content, or parts thereof, in original or modified form in ISO or ISO/IEC standards and iii) to exploit the content as part of an ISO or ISO/IEC standard according to ISO and IEC practice. If you or the organization you are representing own/s the copyright in the content, such permission is implicitly granted to ISO or ISO/IEC, upon submission of the content to the standards development process.”

Or regarding communication of committee work (which, in my understanding, would include this mail list*), among other details:

“[Externally shared] information shall not include details about the identity of the individuals or companies (including the names of the person, the company they work for, etc.)… shall not include information about how any NSB or committee member voted… Committee and working group documents such as working documents, minutes, or working group recommendations shall not be shared externally… personal opinions… are clearly identified as personal opinions… do not convey the views or positions of others… do not criticize the views or positions of others…”

*Based on stated ISO policies that I’m aware of, it’s a bit surprising that activity can happen on a publicly-archived email list unless
i) the (sub-)committee had sanctioned creation of a public list for committee work
ii) the subscription page explicitly calls out the general policy, calls out that the list is public, and implicit or explicit acknowledgement that one’s opinions will be made public (not true in this case)
iii) there is also a non-public forum in which experts that wish to can participate in discussion
In this case, I assume (i) is true; (ii) is not true, though probably everyone here is aware that they are participating in a publicly archived list. And (iii) hasn’t really been necessary up to now because, up to now, nothing has been brought _into ISO processes_ that has been of such a potentially-contentious nature as to require a non-public forum for open discussion among experts without needing to worry about unintended repercussions.

And just to clarify: ISO principles are concerned about openness and transparency of _outcomes_, but chooses to protect views expressed by individuals during the process of arriving at those outcomes because that leads to more openness among experts during the process. ISO is not the only standards-development organization that has such policies.

Note that the communication policy is relevant to patent discussions. I’ll get back to this below.

As for ISO patent policies, there’s a link on that resources page, repeated here for convenience:

https://www.iso.org/iso-standards-and-patents.html

Now, it appears that the details might only be available if have access to internal documents (e.g., if you were added because your NB nominated you as an expert to participate in a WG). Some of the high-level points are

  *   Participants should, as early as possible, “draw attention to” known patents that would be essential to implementation of a specification
  *   Patent holders must submit a “Patent Statement and Licensing Declaration” form and agree to RAND licensing terms for any implementer in relation to essential patents; however, they are not required to enumerate specific details regarding the patents.
  *   In committee or WG discussions, “the Technical Bodies may not take position regarding the essentiality, scope, validity or specific licensing terms of any claimed Patents.” IOW (my paraphrase), the tech nerds should not engage in legal discussions.
  *   ISO maintains a database of declared patents but is not accountable for that being complete.

For US participants in work of SC29, INCITS provides more public visibility into its patent policy, and other legal info:

https://www.incits.org/standards-information/legal-info

It calls out things that are not to be discussed in meetings. Now, that doesn’t include discussion of specifics regarding patents. But here, I think there is an unwritten rule because anybody participating in an ISO technical body and who works for a company with lawyers that are concerned with IP-related risks has (I would think) at some point been told by their employer, “Do not ever discuss patent claims—either ours or theirs—with anyone from another company.”

Certainly early in my time at Microsoft (probably in the required “PM Bootcamp” training), I was told this, along with the elaboration that, if people from another company start discussing patents, ask them not to; and if they continue, then _immediately leave_. The general issue here is, “We do not want you learning about any other company’s patents.”

And the reason for that has to do with penalties. In at least some jurisdictions—and certainly in the US—there are additional penalties if a company is found to be in violation of patent protections (i.e., implementing protected claims without a license) _and it is found that they were aware of the patents_. In the US, prior knowledge of the patents results in automatic _tripling_ of financial penalties.

For that reason, in an ISO (or INCITS or Unicode…) meeting, I would never say more than “I am aware of patents that may be essential for implementation of this specification,” since given any specifics may shut down discussion.

On this list, if someone mentioned patent details in the course of some subject, I would anticipate that some would feel it necessary at a minimum to discontinue participating in or reading any messages in that thread (perhaps even immediately deleting them from their mail). And if that happened repeatedly, I would anticipate people might unsubscribe from the list and, perhaps, insist on a more formal WG process in which details of discussions are not made public.

Consider the presence of a public archive in relation to a company’s concern about protection from patent suits: Suppose person A, even in a single message, mentions specifics about some patent. And suppose the public archive shows that person B working for company X was a member of the list at that time. Even if person B had not read any message mentioning the patent details, the fact that they were receiving those messages from the list would likely be taken as material evidence that they had prior knowledge of the patent, hence making company X liable for triple indemnity if found guilty in a legal suit.

Now, IANAL. But my understanding is that any discussion of any specific details regarding essential patents becomes a significant obstacle to technical discussion. It’s best to leave the patents to the lawyers.

Sorry this got long, but I hope it’s at least somewhat helpful.

Peter

From: Dave Crossland <dcrossland at google.com>
Sent: Friday, August 21, 2020 9:21 PM
To: Peter Constable <pgcon6 at msn.com>
Cc: mpeg-otspec <mpeg-otspec at lists.aau.at>
Subject: Re: [MPEG-OTSPEC] MATH Encumbrance

On Sat, Aug 22, 2020, 12:02 AM Peter Constable <pgcon6 at msn.com<mailto:pgcon6 at msn.com>> wrote:
It’s best not to bring up specific patents in cross-company discussions since some companies have policies requiring employees to exit discussions if other parties mention any specifics regarding patents. (Triple indemnity being a bit scary and all.)
There are a lot of new people here, like myself, who have no idea about this kind of "unwritten rule," so I'm very grateful for your shedding light on this.

Does "exit discussions" mean not posting on this thread, or this mailing list?

I will say that I was a party in some patents in my time at MS that were filed solely for defensive purposes, never with any intent to charge licensing fees.
Right, and my name is on some Google patents for font UI stuff, which I am fine with for the same reasons.

But isn't the point of a formal standards body to get that intent turned into something in writing? And is that what OFF means? Or isn't it?

Since the MATH table is in OFF, this seemed like a good case study for informing the future of 32 bit GIDs and so on that Li Renzhi just mentioned.

But I'm happy to talk, without a specific existing case study, about hypotheticals.

 In the font space, I’d suggest that licensing fees could result in diverging formats and renewed font wars, which nobody wants. It’s not like video formats or something that takes rocket science to eke out the next leap in performance. Nobody significantly benefits if fonts don’t just work everywhere.
I certainly don't want renewed Font Wars, and that's why I'm raising this topic.

In order to prevent it, I want to see a clear path forwards for fixing long-standing problems with OFF within the OFF development process...

And I want to confirm that the OFF process does actually guarantee "just works everywhere," because it ensures patents don't require licensing as written policy.

Peter

From: Dave Crossland <dcrossland at google.com<mailto:dcrossland at google.com>>
Sent: Friday, August 21, 2020 8:09 PM
To: Peter Constable <pgcon6 at msn.com<mailto:pgcon6 at msn.com>>
Cc: mpeg-otspec <mpeg-otspec at lists.aau.at<mailto:mpeg-otspec at lists.aau.at>>
Subject: Re: [MPEG-OTSPEC] MATH Encumbrance

Is that because of a MPEG OFF patent policy applying to https://patents.google.com/patent/US7492366<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatents.google.com%2Fpatent%2FUS7492366&data=02%7C01%7C%7C7c6b72474e5d4370297908d84652bf08%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637336668484634086&sdata=1HnKCddnRJw89dNzxZ2%2FehO%2FiqN6hDrvzpyx4poxVP0%3D&reserved=0> that means a license isn't required?

On Fri, Aug 21, 2020, 9:29 PM Peter Constable <pgcon6 at msn.com<mailto:pgcon6 at msn.com>> wrote:
I’m not aware of any such requirement.

Peter

From: mpeg-otspec <mpeg-otspec-bounces at lists.aau.at<mailto:mpeg-otspec-bounces at lists.aau.at>> On Behalf Of Dave Crossland
Sent: Friday, August 21, 2020 5:51 PM
To: mpeg-otspec <mpeg-otspec at lists.aau.at<mailto:mpeg-otspec at lists.aau.at>>
Subject: [MPEG-OTSPEC] MATH Encumbrance

Hi

Does MATH table implementation require a license?

Cheers
Dave
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.aau.at/pipermail/mpeg-otspec/attachments/20200822/fea949ef/attachment-0001.html>