Mollamby, Conflicts of Interest vs Privacy

As the Mollamby affair has emerged, some people have rushed to defend the privacy of Chris Lamb and Molly de Blanc or dismissed it as mere innuendo without understanding the ethical issues.

What is the difference between innuendo and incrimination? Evidence.

Privacy is a valid consideration, but it is not the only one. I delayed publishing my own blog about the subject while weighing the privacy implications against the ethical issues.

Let's consider some of the evidence backing up the facts about Mollamby. Parts of the evidence have been redacted for the privacy of third parties but the material presented here accurately reflects the situation.

This is the opening comment sent by a student applying for GSoC in 2018 (Fact 2):

Date: 14 March 2018

I am [redacted/student name], ... from [redacted/country].
I’m [redacted/relationship] of [redacted/full name]

The student clearly identified a conflict of interest, giving the name of the other party and the type of relationship. The other party had also sent a similar email:

Date: 12 March 2018

... there are some students who might be interested
in [redacted/project]. Even my [redacted/relationship] has been ....

As they were honest and transparent from the outset, there is no question over their integrity and no need to discuss their identities.

This is the statement I made when agreeing to be a GSoC admin in 2018:

-------- Forwarded Message --------
Subject: Re: Google Summer of Code 2018
Date: Mon, 22 Jan 2018 08:41:49 +0100
From: Daniel Pocock <daniel@pocock.pro>
To: mollydb <deblanc@riseup.net>

On 22/01/18 02:25, mollydb wrote:
> I mmissed this on the application before! We need 2-5 administrators for
> the application. Who else wants to be one?
> 

You can use my name temporarily while looking for other people to help
you in this role.

... [redacted/name of other community] ...

However, I can't officially commit to help with the duties of an
administrator right now.

Regards,

Daniel

No volunteer is under any obligation to provide details of their personal life. This statement alone was made honestly and in good faith, that is what teamwork is all about.

A selection meeting was scheduled for 16 April 2018 and I was the one who reminded people about the conflict of interest (Fact 3). I was not a party to this conflict of interest. de Blanc both acknowledged and agreed with the way it was handled (Fact 6):

<pocock> yes, but [redacted] is not involved in the
  selection process because one candidate is [redacted]
<pocock> that could be one reason we are waiting
   until the last minute to confirm the selections
[redacted/other mentor acknowledgement]
<mollydb> nice responsibile decision making :)
<mollydb> thanks for being so consciencious

People had been reminded about it in a number of emails at each stage of the selection process, it wasn't sprung on people at the last minute. de Blanc had simply left the GSoC emails to other team members:

Date: 12 July 2018
From: Molly de Blanc 

As an additional note, I generally check my email once a week. For
anything immediate, -please- ping me on IRC as I'll be responsive there
(and can know to dive into my email).

When alerts were sent about the conflict of interest in March and April, other team members were unaware that de Blanc wasn't reading them.

Technically, it was a special case that was not strictly covered by Google's official rules. Given the huge effort volunteers make interacting with students, nobody had made the extra effort to seek Google clarification.

Now let's look at the complaint that Stephanie Taylor from Google sent to Debian on 13 July 2018 (yes, that was Friday the 13th):

Subject: Concerns around Debian GSoC students and conflict of interest
Date: Fri, 13 Jul 2018 08:23:36 +0200
From: Stephanie Taylor <sttaylor@google.com>
To: [redacted/private gmail addresses of all Debian GSoC admins]

Hello Debian Org Admins,

It has come to our attention that [redacted/position in Debian],
[redacted/full name], is the [redacted/relationship] of [redacted/name], ...

This is incredibly disturbing as the Debian folks have been valued
members of the GSoC community for many years and this threatens the
integrity of the program.

Taylor is complaining about conflicts of interest in Debian, this confirms Fact 7.

Who would investigate Taylor's complaint? Chris Lamb and Molly de Blanc. Mollamby.

Subject: Re: Concerns around Debian GSoC students and conflict of interest
Date: Fri, 13 Jul 2018 14:49:50 +0200
From: Molly de Blanc
To: Daniel Pocock

Just as a quick heads up, I'll be talking with the DPL later today to
get on the same page -- I know he also contacted Stephanie off-channel.

If you'd like to ping me on IRC, I can try to be online and accessible
(today turned into quite a busy day for me) at a time that works for you.

Cheers,
Molly

Notice that de Blanc does not mention her conflict of interest (romantic relationship with the DPL, Chris Lamb) in that email. Lamb never mentioned it either. Neither of them recused themselves (Fact 8). I was travelling that weekend and couldn't make time to join a hastily organized meeting. As boyfriend and girlfriend, they had a meeting without the rest of the Debian GSoC admin team. When the boyfriend is also the leader of the project and when the girlfriend's conduct is in question, is it any surprise that another volunteer is blamed and the girlfriend takes over the team?

That email is the smoking gun: two people at the very top of the free software ecosystem (Debian and OSI) using a volunteer as a scapegoat for a communication breakdown that one of them had been party to.

This farce is further compounded by the fact the original complaint was about conflicts of interest.

Mollamby hid their own conflict of interest while investigating a conflict of interest.

Is this a new style of disruptive leadership? Or is it simply good old fashioned cronyism?

Even this hidden conflict of interest may not be enough to justify discussing the relationship publicly. However, they have meted out severe punishments on other parties. de Blanc even went to FOSDEM and gave a talk boasting about demoting somebody and putting volunteers behind bars. If these people want to take on leadership positions and preach about harming other volunteers they also need to accept that their own conduct will come under public scrutiny. It is clearly not possible to talk about the way they both concealed and benefitted from a conflict of interest without also making their relationship a public matter. In this situation, the ethical transgressions heavily outweigh the concerns about their privacy.

What's more, I announced my resignation from the Debian GSoC team in August 2018, if people had not acted inappropriately after that, it is unlikely any of these facts would be under public scrutiny right now.

In a non-apology email sent by the new DPL Sam Hartman, Debian confirms there were conflicts of interest and that Debian is completely unprepared for these situations:

I regret that we didn’t have better tools for dealing with conflict of interest and hope we will develop those tools going forward.

...

The conflict of interest issue had no easy answer... There was not a
clear conflict of interest policy.  Sometimes in situations like that
you don’t have good options.

The GNOME community have also done an excellent job of reducing this complicated situation into a concise query to their own leadership. From the GNOME Foundation mailing list:

Nobody appears to be asking about Molly.

People are asking about you (Neil McGovern).  You and Lamb
both come from this Debian Cambridge grouping.
You are the Executive Director.

How long did you know that your new hire
was also your friend Lamby's girlfriend?

Please respond transparently, we would all like
to see this cleaned up so there will be
no discomfort or embarrassment at GUADEC.

It is interesting to see that a student applying to GSoC appears to be demonstrating more integrity than the leader of the Debian Project and the OSI board president combined.

Conflict of interest? OSI board meeting, Spring 2018, Microsoft, San Francisco