Google, Money and Censorship in Free Software communities
On 30 June 2019, a Debian Developer sent the email below to the debian-project mailing list.
It never appeared.
Alexander Wirt (formorer) has tried to justify censoring the mailing list in various ways. Wirt has multiple roles, as both Debian mailing list admin and also one of Debian's GSoC administrators and mentors. Google money pays for interns to do work for him. It appears he has a massive conflict of interest when using the former role to censor posts about Google, which relates to the latter role and its benefits.
Wirt has also made public threats to censor other discussions, for example, the DebConf Israel debate. In that case he has wrongly accused people of antisemitism, leaving people afraid to speak up again. The challenges of holding a successful event in that particular region require a far more mature approach, not a monoculture.
Why are these donations and conflicts of interest hidden from the free software community who rely on, interact with and contribute to Debian in so many ways? Why doesn't Debian provide a level playing field, why does money from Google get this veil of secrecy?
Is it just coincidence that a number of Google employees who spoke up about harassment are forced to resign and simultaneously, Debian Developers who spoke up about abusive leadership are obstructed from competing in elections? Are these symptoms of corporate influence?
Is it coincidence that the three free software communities censoring a recent blog about human rights from their Planet sites (FSFE, Debian and Mozilla, evidence of censorship) are also the communities where Google money is a disproportionate part of the budget?
Could the reason for secrecy about certain types of donation be motivated by the knowledge that unpleasant parts of the donor's culture also come along for the ride?
The email the cabal didn't want you to see
Subject: Re: Realizing Good Ideas with Debian Money Date: Sun, 30 Jun 2019 23:24:06 +0200 From: a volunteer To: firstname.lastname@example.org, email@example.com On 29/05/2019 13:49, Sam Hartman wrote: > > [moving a discussion from -devel to -project where it belongs] > >>>>>> "Mo" == Mo Zhou <firstname.lastname@example.org> writes: > > Mo> Hi, > Mo> On 2019-05-29 08:38, Raphael Hertzog wrote: > >> Use the $300,000 on our bank accounts? > > So, there were two $300k donations in the last year. > One of these was earmarked for a DSA equipment upgrade. When you write that it was earmarked for a DSA equipment upgrade, do you mean that was a condition imposed by the donor or it was the intention of those on the Debian side of the transaction? I don't see an issue either way but the comment is ambiguous as it stands. Debian announced a $300k donation from Handshake foundation. I couldn't find any public disclosure about other large donations and the source of the other $300k. In Bits from the DPL (December 2018), former Debian Project Leader (DPL) Chris Lamb opaquely refers to a discussion with Cat Allman about a "significant donation". Although there is a link to Google later in Lamb's email, Lamb fails to disclose the following facts: - Cat Allman is a Google employee (some people would already know that, others wouldn't) - the size of the donation - any conditions attached to the donation - private emails from Chris Lamb indicated he felt some pressure, influence or threat from Google shortly before accepting their money The Debian Social Contract states that Debian does not hide our problems. Corporate influence is one of the most serious problems most people can imagine, why has nothing been disclosed? Therefore, please tell us, 1. who did the other $300k come from? 2. if it was not Google, then what is the significant donation from Cat Allman / Google referred to in Bits from the DPL (December 2018)? 3. if it was from Google, why was that hidden? 4. please disclose all conditions, pressure and influence relating to any of these donations and any other payments received Regards, Daniel 1. https://www.debian.org/News/2019/20190329 2. https://lists.debian.org/debian-devel-announce/2018/12/msg00006.html 3. https://www.debian.org/social_contract
Censorship on the Google Summer of Code Mentor's mailing list
Google also operates a mailing list for mentors in Google Summer of Code. It looks a lot like any other free software community mailing list except for one thing: censorship.
Look through the "Received" headers of messages on the mailing list and you can find examples of messages that were delayed for some hours waiting for approval. It is not clear how many messages were silently censored, never appearing at all.
Recent attempts to discuss the issue on Google's own mailing list produced an unsurprising result: more censorship.
However, a number of people have since contacted community representatives privately about their negative experiences with Google Summer of Code. Here is one of the messages that Google didn't want other mentors to see, sent by one of the former Debian GSoC admins:
Subject: [GSoC Mentors] discussions about GSoC interns/students medical status Date: Sat, 6 Jul 2019 10:56:31 +0200 From: a volunteer To: Google Summer of Code Mentors List <email@example.com> Hi all, Just a few months ago, I wrote a blog lamenting the way some mentors have disclosed details of their interns' medical situations on mailing lists like this one. I asked the question: "Regardless of what support the student received, would Google allow their own employees' medical histories to be debated by 1,000 random strangers like this?" Yet it has happened again. If only my blog hadn't been censored. If our interns have trusted us with this sensitive information, especially when it concerns something that may lead to discrimination or embarrassment, like mental health, then it highlights the enormous trust and respect they have for us. Many of us are great at what we do as engineers, in many cases we are the experts on our subject area in the free software community. But we are not doctors. If an intern goes to work at Google's nearby office in Zurich, then they are automatically protected by income protection insurance (UVG, KTG and BVG, available from all major Swiss insurers). If the intern sends a doctor's note to the line manager, the manager doesn't have to spend one second contemplating its legitimacy. They certainly don't put details on a public email list. They simply forward it to HR and the insurance company steps in to cover the intern's salary. The cost? Approximately 1.5% of the payroll. Listening to what is said in these discussions, many mentors are obviously uncomfortable with the fact that "failing" an intern means they will not even be paid for hours worked prior to a genuine accident or illness. For 1.5% of the program budget, why doesn't Google simply take that burden off the mentors and give the interns peace of mind? On numerous occasions Stephanie Taylor has tried to gloss over this injustice with her rhetoric about how we have to punish people to make them try harder next year. Many of our interns are from developing countries where they already suffer injustice and discrimination. You would have to be pretty heartless to leave these people without pay. Could that be why Googlespeak clings to words like "fail" and "student" instead of "not pay" and "employee"? Many students from disadvantaged backgrounds, including women, have told me they don't apply at all because of the uncertainty about doing work that might never be paid. This is an even bigger tragedy than the time mentors lose on these situations. Regards, Daniel 1. https://danielpocock.com/google-influence-free-open-source-software-community-threats-sanctions-bullying/ -- Former Debian GSoC administrator https://danielpocock.com