One Of LLVM's Top Contributors Quits Development Over CoC, Outreach Program

[mastermind]

In case you trust Snopes

Just stop. This is ridiculous.

"Not long after acquiring the reins of power, the Nazis banned the Social Democratic Party and sent its leaders and other leftists identified as threats to the National Socialist program to concentration camps."

Intra-species competition is always the harshest. Of course they could not tolerate other competing leftists. The same happened in Soviet Russia with all the "cleansings" of competing socialist/communist factions.

Edit: Sorry, the above picture on Snopes is actually outdated. See: Snopes CEO Accused in Divorce Proceedings of Embezzling Company Money to Spend on Prostitutes

[Palu Macil]

Just stop. This is ridiculous.


Intra-species competition is always the harshest. Of course they could not tolerate other competing leftists. The same happened in Soviet Russia with all the "cleansings" of competing socialist/communist factions.

Edit: Sorry, the above picture on Snopes is actually outdated. See: Snopes CEO Accused in Divorce Proceedings of Embezzling Company Money to Spend on Prostitutes

I'm sure your source is fine in this case, but I do find discrediting Snopes with a Breitbart link to be a bit ironic...

[Djhg2000]

I should have made sure I quoted you to be explicit about my response being to you...



It's funny you should use a quote so important to me. I did a lot of serious thinking about this quote during my four years as an infantry officer in the United States Army before my IT career where I now work on security-related software. I personally oversaw an information operation (propaganda) involving a newspaper and five radio stations in Afghanistan, and I thought a lot about the freedoms related to speech, usage of censorship, and when context changes some of these concepts. I realize not everyone here is an American, but I will only be able to speak with a deep understanding of my own nation's laws and customs.

In the United States, it is clear that freedom of speech should be protected from other nations or our own government encroaching. This is the most important context because those are the most powerful forces against which an individual would be powerless. However, the freedom changes based upon context when you look at other types of powers. The ultimate authority on interpreting this is the Supreme Court, and they have been very clear that freedom of speech is extremely important without being absolute. It doesn't bind individuals at all. I could tell you to leave my home if you say you think raisin cookies are better than chocolate chip, and that is not a violation of free speech. You could simply decide that you aren't my friend anyway. I have no real power over you. Employers have some regulation because they have power over you. You can look for a new job, so the power is not absolute, and so they can still tell you that you'll be fired if you say certain things. Academia is somewhere in the middle where they are an actual government institution (absolute power, or close to it, and the very topic of the freedom of speech amendment) but the Supreme Court has still allowed for restrictions on the freedom of speech.

As an aside, I don't really know enough to comment on the academic actions you witnessed. Investigating a complaint itself isn't suppression just as much as I don't think research should be suppressed even if it could be used by bigoted groups. There are a lot of other factors that could come into play, and in an academic setting there always seems to be a lot more going on than in other fields. I don't think those topics even get raised in a software setting.

Now we get to examine an open source community. They have almost no power over a collaborator (certainly less than an academic institution or employer) and therefore have all legal right to make specific choices in regards to speech they will accept or reject. Just like the US government at each level of federal, state, and local reserves more specific decisions to smaller entities (states rights and the equivalent state-by-state constitutions which vary) a community should be free to make very specific rules that shape the feel of the community. There aren't a lot of opportunities for people to bring up something that would get incorrectly interpreted as derogatory. A lot of people use phrases that make it sound like they need to put other people down and be angry and dramatic to get points across because a sea of idiots will otherwise overwhelm the project otherwise. I don't think this is a real problem. Rules should exist to fix real problems, not because of a flat understanding of an ideal that is far more nuanced in practice.

I have to disagree because freedom of speech is absolute. What you're talking about is the intent behind the speech, which is not inherently protected under free speech. They are two very different things and it's of outmost importance to recognize the difference.

For instance you can sometimes hear pretty bad threats being made online. The words formulating the threat are protected under free speech but the intention certainly isn't, so they could still get in trouble for the threat itself, but the words remain protected under free speech. If they weren't then it could be illegal to quote the threat. Or perhaps more relevant today, to store the written text containing the threat temporarily on a communications server.

What is happening in the world of academia today is the frightening development of the term "hate speech". This term implies the intent behind the speech can be nothing but malicious no matter the context. As soon as that label is applied the cited collection of words become forever tainted and along with them anyone directly or indirectly associated. The worst part is the social pressure to deal with this so called "hate speech", because if you don't then you are automatically presumed to agree with it. The consequence of that is the social acceptance of another very poisonous concept; "guilt by association".

"Guilt by association" is the idea that if you are guilty of something then anyone you can be associated with is guilty as well. This is in complete conflict with a basic cornerstone in a free society, namely that everyone is innocent until proven guilty. As an American I'm pretty sure you can agree with that one since it's effectively written into the constitutions fifth amendment. The two concepts are simply incompatible on a fundamental level, yet the former seems to be increasingly popular on social networks and propagating into academia.

Now the quote is indeed quite deep. To me it represents my inalienable right to express myself just as much as it represents your inalienable right to express yourself. There's nothing I can do to stop you but nor do I want to.

There's also a second quote I'd like to bring up about why censorship is flawed, but I cannot remember the exact quote and my web search skills are failing me. The gist of it is something along the lines of "when you give a liar the ability to freely express himself, he will eventually expose his own ill intentions". In my own experience this is definitely true. If you believe someone is trying to fool you then you let them speak until they fall into their own traps, at which point all you would need to do is to repeat their own words and their plans fall flat.

As to your point about the open source community I'd argue that while they are free to put up whatever demands they want, there's no reason why they should be immune to criticism by doing so. What they are trying to do is to impose limits on speech by their contributors, which to me is taking it too far. It's one thing to have guidelines (many projects have previously had those) but a whole other one to have hard rules that could jeopardize your future at any seemingly arbitrary moment. Again, I can tell you that it's hard to be truly productive in such an environment because of the added mental stress.

[audir8]

Traversing the entire tree of your unsupported conjectures, and their reasonable rebuttals, is NP-Complete. If I were you, I wouldn't expect people to address all of your statements, especially if you refuse to address even one of theirs to a satisfactory degree (or at all).

The same goes for you. I was defending this CoC: http://todogroup.org/opencodeofconduct/

and I think your statement proves you can get tired to defending against bullshit. I've only posted thrice including this, tired so easily?

[DMJC]

These political arguments are meaningless in the context of getting code written. Frankly all discussion of sex/sexuality should just be banned on open source projects. Unless you're working on a tele-dildonics project, or some sort of political software it should be completely unnecessary to even go there. That goes for slurs and insults as well as general discussion topics. Just ban all of it because the arguments for/against are too corrosive and divisive. Your LGBT/Hetero/Cis status doesn't mean shit to me, give me your code patches or shut up. Want to call someone gay? Ban-hammer. Want to talk about how much you like <insert gender here> Ban-hammer, and if you want to bring up the topic of discrimination outside of making a complaint backed by evidence? Big-giant Ban-hammer. Those should be the only rules governing discussion on these projects.

[Blahblah]

I think there is an apolitical issue with CoC's that is often ignored: they take informal discussions and place them into the realm of formality. For most FOSS projects, there are strong arguments for avoiding unnecessary formality:

1) When you have a formal CoC, you now need some sort of apparatus to enforce it. This both implies a non-zero time cost in management, and it also means that _somebody_ (or somebodies) need to be elevated above the other contributors to administer punitive actions. Creating a hierarchy in a group of only 10-20 consistent contributors, though, is bound to create in-fighting.

2) The ability to have candid, not-taken-seriously conversations reduces the pressure placed on those who contribute. Arguably, this makes the project more welcoming to newcomers than having a CoC.

3) The content inside the CoC itself is a maintenance effort, and depending on how litigious the community is, that can either be negligible or a source of constant frustration.

I could see justification for an official set of rules in a project that requires extremely rigorous coding standards, but in those cases, the CoC would be MUCH more than just "Don't be an asshole". It would be something like "Before you send your patch, make sure to run it through XYZ-testing-framework and record the blah blah blah...".

And before someone replies "Well the rules were written informally, so that means the conversations will stay informal."; it doesn't work that way. You could have a rule about _anything_, and you will eventually have some jerk trying to enforce it as a technicality in an unintended situation. Since someone is now technically in a position to enforce it, they are left with two options

1) Ignore the complaint, which means you are now playing favorites

2) Enforce the rule, and further push the community towards a tightly-controlled system

[k1e0x]

An actively sexist program based on a misguided view of the world, and a set of oppressive speech codes seem like perfectly decent reasons to part ways with the project. National speech codes are a major reason why I'm emigrating from my home country.

The problem with repressive codes of conduct is that they can be represented as "reasonable" and "something no decent person would disagree with", while repressing people severely. The scale of consequences for the most minor of missteps, as judged by your accusers, make most codes of conduct rather chilling. I know I actively avoid investing time in projects with strict codes of conduct, because people who contribute basically nothing can get you booted from a community you worked hard to create, just by feeling offended. Some codes even include restrictions on your public communications outside the project.

Even if you're capable of adhering to the code of conduct, it has an effect on your relationship to a project. It tells you that politics is more important than achieving the technical goals. This sort of politicking has already severely impacted the NodeJS upstream.

There also seems to be this idea going around that there's an underutilized population of enthusiastic, experienced female software developers (or talented aspiring developers) whose only barrier to success is sexism; I'd like to see just one shred of evidence that this is the case, and not just some academic/HR department fantasy. To me it seems like the goal of discriminating against 90% of the talent pool directly contradicts the goals of a software development project: to produce and maintain a high quality solution to a set of technical problems.

This is a good post and as you can see from all the up votes, people in general are in agreement with you. There is no need to pander you CoC to a vocal minority out of fear.

Diversity should be accepted and welcomed but it should not be the overall goal. Technical progress is the overall goal and that requires the best talent. (it even requires tolerance of assholes sometimes, because sometimes assholes are talented) Diversity as a goal is just bizarre.. everyone can't be a minority.

[fabtar]

I totally agree with Rafael, imho CoC is not the bigger issue but the LLVM association with Outreachy is.

When rules are about gender/race/perks and not about "persons" they are bad. Here in Italy, this attitude is slowly moving into the public too, i.e. here is a case with economic aids and discounts on the university taxes based on gender (for STEM, Italian article: http://www.ilsole24ore.com/art/notiz...?uuid=AEgpU2dD).

I think that discrimination should not be tolerated, for this reason kudos to Rafael for acting.

[cipri]

I am sorry, that you dont see the ideas behind coc.
Also in the academic world, if two candidates are for example of similar competence, the female has the priority. And it is good like this, to recover the inbalance.
I would recommend you to read about studies, that show that a group of homogen experts in one domain, can be less productive/innovative at solving problems than a team with people with different backgrounds. I am sorry, that you can also not imagine why bringing somebody from a different region , could help to spread the project of LLVM to also other parts of the world, bring in the end more contributors and more people interested in the project.

Sapere Aude!

cipri

[Holograph]

I am sorry, that you dont see the ideas behind coc.
Also in the academic world, if two candidates are for example of similar competence, the female has the priority. And it is good like this, to recover the inbalance.
I would recommend you to read about studies, that show that a group of homogen experts in one domain, can be less productive/innovative at solving problems than a team with people with different backgrounds. I am sorry, that you can also not imagine why bringing somebody from a different region , could help to spread the project of LLVM to also other parts of the world, bring in the end more contributors and more people interested in the project.

Sapere Aude!

cipri

This reply lacks substance and doesn't back up any of those claims or explain what religion would have to do with coding quality. If you'd like your point to be appreciated you would be well off backing up and explaining some of your statements at least.

(This is not hypocritical on my part because I'm not making any claims about what effects CoC has on the project - or on other projects.)