Recording of meetings #334
Comments
|
(Also IANAL but many jurisdictions have laws around recording meetings, and I doubt anyone in the room at the time had adequate knowledge of the relevant legal situation pertaining at the time of the meeting.) |
|
In other orgs with less developed scribing practices, I've seen meetings being recorded by the person responsible for making a transcript, but this comes with announcements of "is everybody OK with this?" and "I will delete the recording once the transcript is made." every time someone who's not already familiar with the practice attends the meeting. I would not necessarily go as far as strictly banning recordings, but given that they are known to make a number of people uncomfortable for a number of reasons, I would say that there should not be an assumption that it's OK just to record, and that if you want to, you need to get everybody's permission. |
|
To be clear, in this case that question (“is everyone okay with this?”) was
asked. (Not speaking for or against recording, but permission was asked.)
…On Thursday, October 3, 2019, Florian Rivoal ***@***.***> wrote:
In other orgs with less developed scribing practices, I've seen meetings
being recorded by the person responsible for making a transcript, but this
comes with announcements of "is everybody OK with this?" and "I will delete
the recording once the transcript is made." every time someone who's not
already familiar with the practice attends the meeting.
I would not necessarily go as far as strictly banning recordings, but
given that they are known to make a number of people uncomfortable for a
number of reasons, I would say that there should not be an assumption that
it's OK just to record, and that if you want to, you need to get
everybody's permission.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#334>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAD3Y6KSNIWELKA25VC5IUTQM2FVNANCNFSM4I5GH6QQ>
.
|
|
The reason for the recording or its intended purpose is a consideration.
Is it an official record of the meeting, or just a format some people
might find easier to consume after the fact than reading minutes. I
believe it's often easier to listen than read in a second language for
example, it's also easier to listen than read if you're Dyslexic or similar.
We may already have complete historical records of what is said at
meetings. If a remote call is transcribed automatically into captions
for deaf people, that will include someone saying "not for the minutes".
To a lesser extent, the IRC logs can be edited before the minutes are
published, but the IRC logs themselves (especially those on people's
local machines) cannot.
Do we know why the request was made in this instance? A use case might help.
|
Yes...but it seems that the answer "no" wasn't readily accepted. So perhaps we need to ask what the process is for resolution; I would have thought that anyone in the room could say "no" and that would be decisive, but there are obviously other possible outcomes. |
|
Actually, no, the answer "no" was readily accepted in this case; there was, as @hober said, a bit of surprise on both parts, but no significant issue afaict. (I support Tess' objection to being recorded and her query of whether this should be clarified in the Process, fwiw.) I'll ask why the request was made. |
|
I was the one who requested to record the meeting, for two reasons:
As noted, I was slightly surprised by the objection, but fully accepted it, and avoided recording the meeting in its entirety. |
|
I think we've gotten distracted from the issue I was trying to raise by focusing on the particulars of a specific event. |
|
Personally, I think the process should discourage (SHOULD NOT) official recordings of meetings by chairs, and forbid (MUST NOT) unofficial recordings of meetings by anyone. One of the primary purposes of having a process in place is to encourage lively participation in meetings; recording has the opposite effect. Also, I don't think we should put chairs in the position of having to be aware of the laws in the relevant jurisdictions. |
|
I agree we should not put chairs in the position of needing to adjudicate local legal restrictions, and supportive of your reasoning for discouraging recordings. I'm little concerned about "MUST NOT" for the same reason. (Are chairs legally responsible for determining if the conference room is bugged, kinda thing.) |
|
From an enforce-ability point of view, "Chairs MUST ensure that participants are not recording [...]" would not be reasonable. "Participants MUST NOT record [...]" would be. Whether we should go with MUST NOT or SHOULD NOT for participants, I'm less sure. Whether for scribing purposes, accessibility reasons, or other unforeseen reasons, there might be cases where it'd be appropriate. If we need to write this down, (which I am not 100% sure we do) I'd probably go with something like:
|
|
As an experiment, and following suggestion from one of the co-chairs, we have started publishing recordings from the Media & Entertainment IG calls. Our main reason for doing this was to increase the IG's reach to potential new participants. I share the reservations that @hober has mentioned here. The process we've used is to ask participants whether it's OK to record at the start of each call, and to not do that if anyone objects. In practice, no one has objected so far. I support having guidance in the process, and @frivoal's suggestion seems good, although I suggest making explicit that consent is required for publication/distribution of any recordings. |
|
I think I'd express it slightly stronger, something like Participants in a meeting should not make audio or video recordings, and must not do so without the explicit consent of of everyone who would be recorded, even if the recording is only for personal use of the person making it. Furthermore, separate explicit consent must be obtained before the recording is distributed to other people, or made public. |
|
I'd like to object to stating in the Process that recordings are simply not allowed. I respect the desire to not be recorded; I think any chair who wants to record needs to 1) clearly state the purpose of recording and the availability (public, Member, private-only-used-for-chairs-editing-the-minutes), 2) ask all participants for their permission, and do not record if anyone objects. This seems to match @chrisn's suggestion, and others I've talked to who have used recording as a tool (following those rules). |
Is anyone proposing that though? My text and David's variant say should not in the base case, not must not, and only say must not when you do not have consent. That's not "simply not allowed". |
|
Must not without consent is of course proper. "Should not" is a pretty
strong admonition to not attempt this as a matter of course.
…On Wed, Oct 9, 2019 at 9:56 AM Florian Rivoal ***@***.***> wrote:
I'd like to object to stating in the Process that recordings are simply
not allowed.
Is anyone proposing that though? My text and David's variant say *should
not* in the base case, not *must not*, and only say *must not* when you
do not have consent. That's not "simply not allowed".
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#334>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAD3Y6M23BL3TG4SW7WIFF3QNYENLANCNFSM4I5GH6QQ>
.
|
|
@cwilso, should not is "must not, unless you have a really good reason."
I think your (1) is the chair providing that really good reason to the room. So should not is the right formulation here. Unless you do (1), you can't. |
|
@hober that leaves a lot up to the determination of who is responsible for determining "what's a really good reason?" If it's the chair, okay, but I think we're in the same state we're in today? |
That's not the same state we're in today. Today, the process fails to give chairs any guidance here. |
@cwilso my reading of the proposal is that it's the whole group who determines if the reason is good enough and balances that against negative impacts. If anyone in the group is sufficiently concerned they can object, at which point there is no consensus for making the recording and the default of "not recording" needs to clearly apply. |
|
On Tue, 08 Oct 2019 14:17:57 +0200, David Singer ***@***.***> wrote:
I think I'd express it slightly stronger, something like
Participants in a meeting should not make audio or video recordings, and
must not do so without the explicit consent of of >everyone who would be
recorded, even if the recording is only for personal use of the person
making it. Furthermore, separate >explicit consent must be obtained
before the recording is distributed to other people, or made public.
I think this is in the right direction, but we need to think more
carefully.
I'm not sure that the specific technology details belong in the Process,
but we need some clear and well-known procedures.
For legal reasons, no recording should be made without the consent of all
participants.
A request for permission to record needs to include where/how the
recording will be stored and to whom it will be released for what purpose
- it is quite reasonable for people, on privacy grounds, to object to
cloud storage of a recording.
On the other hand, e.g. to facilitate auto-scribing for people who are
hard of hearing or who find spoken english difficult to understand, or as
an aide to post-meeting editing of the minutes for greater accuracy (some
groups do this), there are real use cases for recordings, including in
not-very-privacy-friendly systems.
We should probably end up with part of our participation, code of conduct,
and privacy policy noting our position on this.
|
@chaals please could you be more specific about the legal reasons? I have an antipathy towards vague "lawyers say its bad" arguments, and I don't think this is obvious, particularly since IRC scribing is already done and persisted and we already accept that IRC log != minutes. I think there could be an argument that if you're willing to persist an IRC log which may be almost verbatim, then there's no significant difference between that and an audio recording. I've never heard anyone request that no IRC log is kept, though obviously people do sometimes ask the scribe to take their hands off the keyboard for a while, which in my current mental model is equivalent to "please pause the recording". If there's a particular legal concern, there probably is a mitigation we should be aware of before finalising any process wording. |
|
@chaals indeed, can you indicate whether you think any changes need to be made to the (mercifully brief) proposed text? The question of shared storage is covered by the requirement to get separate permission if the recording is for more than personal use, for example |
|
The laws regarding recording of conversations are related to wire-tapping. In some places, you only need one party to know (or to have a warrant, in others you need full consent. I think - and I think it is the consensus - that we should require consent of all parties before recording a call. I don't think we should add any text to the process. I think this is an operational requirement that chairs need to be aware of and we need to make sure new participants are aware of, as we do with the Code of Conduct. And where the recording is stored is, a priori, relevant. I'm far less likely to give permission for recording (or more likely to simply drop off a call) where the sound is held in a cloud service that may subsequently be used for tracking my voice - or there is no clear archived promise that it will not be turned over to such a system. |
That falls under the category "is or made be available to others" as far as I am concerned. I don't think we should attempt to cover, or interpret, local laws. We do need norms of behavior, though. So, I don't think this is mere practice; members o fthe W3C need to be able to point at a policy that they can request be enforced. So, asking again for specific text; I would be OK adding a note that people have to abide by applicable (not necessarily 'local', as 'local' may not be well defined) laws. How is this? Participants in a meeting should not make audio or video recordings, and must not do so without the explicit consent of of everyone who would be recorded, even if the recording is only for personal use of the person making it. Furthermore, separate explicit consent must be obtained before the recording is distributed to other people, stored in shared or cloud storage, or made public. Note: there may also be applicable laws concerning recording. |
|
I'd like to push back on the "should not" part here. At the WebPerfWG we've been recording calls for the last year and posting them online (with explicit notice at the invitation to the calls as well as at the beginning of each meeting). We've seen multiple benefits to that practice:
I believe the transparency those recordings provide is extremely valuable. I'd hate to see that go away. While I understand that a "should not" means that chairs could still record while justifying their actions, I believe it sends the wrong message. Instead of encouraging openness and transparency by default, we would be encouraging the opposite. I don't think we should do that. |
|
I'd like to push back on your push back. Scribing is intentionally lossy: there is matter that can be said, or poorly said, that should not survive the meeting. Far from giving transparency, I think recording can and does have a significant chilling effect on conversation. It relays tone, for example: something said in a truly hostile or passive-aggressive way. So, I don't agree that recording encourages "openness and transparency ". |
|
I'm sure I don't understand you correctly, @dwsinger , because I just heard this as "recordings would have a chilling effect on the ability of participants to be hostile or passive-aggressive in the moment, and not have that recorded in the minutes." Which... I don't think is a good thing? |
|
It's the chair's job to manage hostile speech and passive-aggressive behaviour, not the existence of recordings. Everyone is (or should be) much much more careful opening their mouth, in case they mis-phrase, or come across wrongly in audio, if they are being recorded. It tilts the playing field to native speakers, people who are attuned to nuance, and badly away from people less comfortable talking. |
Is being mis-scribed significantly better? Again, from the WebPerfWG's experience, I saw multiple times where in the heat of scribing, some complex technical nuance was misunderstood and the scribe wrote down the opposite of what was actually said. |
|
Yes, I think we will only need to say something about the case when the transcript IS saved. Perhaps that retention of an automatic transcript of the entire meeting is subject to the same rules? Summarizing again: for the Process Written minutes are the formal record of decisions and their quality is therefore important; recording or transcripts can assist the accuracy and completeness. Recording for the chair/team-member/scribe to improve the meeting minutes (where the recording is kept to a few people and then discarded) is generally expected to be more acceptable than recording for the consumption of larger, or uncontrolled, groups, and/or perpetuity. for the Guide: If recording or retention of a transcript is intended by anyone, it is good practice:
|
Change to
We don't want to accidentally ban good scribing.
I'd suggest adding:
This is to make is so that sharing a recording to improve ascribing, of for the sake of a missing attendee, or for the accessibility needs of particular individuals is fine, but just putting it up on youtube is not. And with that in place, the second sentence of the second paragraph ("Recording for…") is no longer necessary. |
|
When we introduce the topic of Recording of Meetings, I would like us to provide verbiage which helps the community appreciate different points of view. Building on what @dwsinger previously wrote, here is a stream of consciousness: "Recording of W3C meetings can have powerful benefits for members of the W3C community but at the same time raises substantial privacy concerns. The benefits are both for those who attend the meeting, who do not have English as a first language or may not be familiar with W3C jargon, as well as for people who could not attend the meeting; especially if the meeting took place in a time zone which was not hospitable for them. On the other hand, there are many participants - while comfortable with scribing of meetings - are not comfortable with their verbatim words recorded and captured for posterity. In some jurisdictions there are also legal regulations about recording meetings. W3C seeks to have policies that balance these competing interests. Balancing them also depends on the means of recording (recording of full audio, automatic speech recognition with a text record of the output) and the duration with which recordings are kept (permanently, for a limited amount of time - after which they are discarded, used only by the scribe to clarify minutes, discarded immediately after the session (i.e. use the ASR only as a real-time aid). To balance these considerations, W3C encourages that groups look at the best balance point for their needs; respecting the strong considerations on both sides of the issue. In addition, W3C has the following formal policies:
|
|
I'm fine with the change to automated. I think that if a group says "we intend to take a recording to publish on YouTube" they are likely to get push-back, but I can conceive of meetings where a group agrees a recording could be useful. I have a slight preference not to ban anything, but rely on consent. The same goes for perpetuity. |
|
@jeffjaffe great text for the Guide! |
|
and yet again: Summarizing again: for the Process No-one may record a meeting, or retain an automated transcript, unless the intent is announced at the start of the meeting, and no-one withholds consent. If consent is withheld by anyone, recording/retention must not occur. The announcement must cover: (a) who will have access to the recording or transcript and (b) the purpose/use of it and (c) how it will be retained (e.g. privately, in a cloud service) and for how long. Note: Written minutes are the formal record of decisions and their quality is therefore important; recording or transcripts can assist the accuracy and completeness. Recording for the chair/team-member/scribe to improve the meeting minutes (where the recording is kept to a few people and then discarded) is generally expected to be more acceptable than recording for the consumption of larger, or uncontrolled, groups, and/or perpetuity. for the Guide: The W3C Process (linked) expresses W3C policy. Recording of W3C meetings can have powerful benefits for members of the W3C community but at the same time raises substantial privacy and other concerns. The benefits are both for those who attend the meeting, who do not have English as a first language or may not be familiar with W3C jargon, as well as for people who could not attend the meeting; especially if the meeting took place in a time zone which was not hospitable for them. On the other hand, there are many participants who, while comfortable with scribing of meetings, are not comfortable with their verbatim words recorded and captured for posterity. In some jurisdictions there are also legal regulations about recording meetings. W3C seeks to have policies that balance these competing interests. Balancing them also depends on
To balance these considerations, W3C encourages that groups look at the best balance point for their needs; respecting the strong considerations on both sides of the issue. If recording or retention of an automated transcript is intended by anyone, it is good practice:
|
|
The proposed process text in the comment above seems pretty close to the mark, as it enables us to do the right thing, but I would prefer of something stricter, that outright banned recordings meant for broad distribution / perpetuity, rather than merely note that they are less likely to be acceptable. I'd prefer stopping at recordings kept for limited time periods, made for the sake of identified individuals. Otherwise, we have a very high risk that there will be people who want to object to the recordings, but feel socially pressured not to (and possibly end up no participating at all as a consequence). The rule as proposed enables a few people to proactively object on behalf of others in every meeting, so I could live with it, but then we might as well ban in in the first place, which is what I propose to do. As for the Guidelines part, (in addition to the corresponding adjustment), I believe that "not comfortable with their verbatim words recorded and captured for posterity" understates the issue. Yes, some are reluctant to recordings due to "verbatim words being captured", but another important concern is those that worry about the increased risk of bullying or harassment. Specific proposal:
In case anybody needs a reminder of the motivation for that stricter stance, I'd like to refer you to the earlier #334 (comment) by @jensimmons. |
|
@fantasai pointed out to me that the above might be too strict, because it would not only apply to routine situations (at which it is aimed), but also at things like wanting to make a documentary about W3C, or a WG wanting specifically to make a collective video for some occasion, or that sort of things, which I don't feel we ought to be necessarily banning outright. With that in mind, replace my earlier proposal with this:
|
|
With regard to prohibiting routine recording of meetings:
I would much prefer we focus efforts on ensuring recordings and publications of these recordings happen only when all participants or would-be participants are comfortable with it; the push to having an advance warning in the agenda is good, and this could come with a suggestion that participants can contact privately the chairs/staff-contacts/an ombud if they feel having the recording put them at a disadvantage or at risk. Another angle might be to distinguish "presentations" from "discussions" (which is what we did most of the time for TPAC breakouts, and what has been done in recorded workshops afaik); but this feels challenging to capture in a hard-set rule, which is why I don't think prohibition is the right approach. |
|
In today's meeting, the sense was that my "balance of perspectives" was too long for the Process Document, which is why it needed to be relegated to the Guide. If the problem is length, I think a shorter version can be created for the Process Document by taking the opening section - specifically: "Recording of W3C meetings can have powerful benefits for members of the W3C community but at the same time raises substantial privacy and other concerns. The benefits are both for those who attend the meeting, who do not have English as a first language or may not be familiar with W3C jargon, as well as for people who could not attend the meeting; especially if the meeting took place in a time zone which was not hospitable for them. On the other hand, there are many participants who, while comfortable with scribing of meetings, are not comfortable with their verbatim words recorded and captured for posterity." |
|
yes, we need to handle recording portions (e.g. just the presentation) |
The WebTransport WG should also be added to that list. We have implemented recordings since our inception. We announce the start of recordings and ask for objections before commencing. We have had none to date and nor have we had any negative consequences, especially around harassment or bullying. We use the recordings to generate automated transcripts, so that the chairs can focus more on the technical nature of calls and less on being automated transcription machines. We do manually document actions, resolutions and issues being discussed. We avoid being greeted by omnipresent silence when asking for "can I have a volunteer scribe?". The potential abuses of manual scribing (selective recording, conscious and unconscious bias, subjective filtering) are numerous in nature and not much discussed in this thread as a counter-balance to the potential abuses of recording. A policy compromise between the two, in which we use modern technology to automate tasks, while protecting against the risks of privacy-loss, censorship and harassment incurred with any type of persistent digital documentation, would seem eminently desirable and achievable. |
|
@wilaw That's good information. Do you publish the recordings, or only the transcripts? |
We publish the recordings and the transcripts. All recordings are password protected. The password is made available to work group members upon request to the chairs. For someone who missed a meeting due to conflict or timezone, being able to revisit the complete meeting as it occurred is a useful capability. |
|
OK, we merged in @dwsinger’s proposed normative text per https://www.w3.org/2021/03/10-w3process-minutes.html#r01 with the following tweaks to address presentations etc. and to clarify that we're talking about audio/video recordings and not records such as the minutes:
Let us know if it needs further tweaking. The question of whether to restrict groups from taking such recordings even when consent is not withheld is still open. (Personally, I think we are not prepared to address this question for Process 2021.) Whether any non-normative guidance should go into /Guide or whether some of it belongs in a note in the Process itself is also still open. |
|
This is a candidate to close, as we have text in the proposed process |
dwsinger
added
Closed: Accepted
|
Looks good to me! |
frivoal
added
the
Commenter satisfied/accepting
frivoal
removed
the
Commenter satisfied/accepting
and others
During a group's meeting at TPAC this year I was very surprised when the chair intended to make an audio recording of the meeting, and objected. They seemed as surprised by my objection as I was by their intent to record the meeting.
One quality inherent to the practice of minuting a meeting is that participants are able to occasionally request that a statement not be minuted, or be minuted differently—this sometimes happens after the statement was made. Creating an audio or video recording of a meeting removes this option, and stifles participation.
I wonder what, if anything, the process should say on this.
The text was updated successfully, but these errors were encountered: