Hey Nat glad to see you here. A few days ago one of the biggest team collaborative games (Space Station 13) got banned on GitHub without a public explanation from GitHub staff, but some suspect it was because the code contained bad words and slurs. Do you know if this is why the project was banned, and will these new private team repos be subject to the same terms/rules?
Do you think the scale could be handled better if you informed repo owners 1: that their repo was disabled, and 2: why their repo was disabled?
Currently the owner has to contact support to know why it was disabled, our repo was disabled thursday at 5am pdt, we sent a ticket by 6am. We still don't know why it was disabled. Its tuesday. (edit: we did get a reply, vague comment about slurs, nobody's sure if its the nword word filter (so thats getting removed, ironically enough), or the comment from 2014 with a soft-a, (but it can go), or the fact that the meatball food item has a, umm, british name)).
Also, do you think the scale of content moderation would be easier if you tiered repo disables between can be resolved and can not be resolved, and in the former case provide the same 24 hours deadline that you provide line item dmcas, as well as provide access to the owner during any suspension if the 24 hours deadline is not met (That you also provide to line item dmcas)?
All of these unneeded trips to support has to be eating into the efficiency of things.
No, it’s not fair. Banning a repo should be taken as seriously as banning a book. Living in a country that is US where github HQ is hosted, freedom of speech should be prized and cared for dearly. For a commercial company, there should be only one reason to ban a repo and that is to abide with a law. For even that company should do everything in its power to prevent that or provide a viable lawful alternative. This should be taken so seriously that each ban should have been reviewed at CEO level. GitHub CEO saying he has no clue, it’s a scale issue and “mistakes are made” is not really acceptable.
I appreciate the idealism here, but the reality is that trying to run a business under the pretense of free speech absolutism can alienate an otherwise profitable market segment. With the loss of that market segment likely comes the grumbling of investors, to whom ultimately the executive management is beholden.
Grumbly investors beget grumbly board members, who then vote to oust executives to correct the profitability problem.
I think this is the most sensible answer here. My sibling comments are attempting to draw analogies to other types of censorship of minority groups which don't strike me as apt.
IMO you correctly summarized the forces they are dealing with. These people are just trying to make money. Idealism is problematic for the people invested in the company that aren't there for idealism, but money.
> can alienate an otherwise profitable market segment
How are you going to alienate/lose customers by not getting rid of customers? If anything, I'd argue the opposite; a platform that refuses to ban legal content is one that I find easier to trust (for a counterexample, see Google). It's not even like github-like companies are social networks where you can claim that one user's experience of the platform is made worse by another user's posts.
We all know that the most vocal on the left, who want to silence anyone who doesn't pander to their political ideals, pressure public companies, advertisers, etc. to 'cancel' those who refuse to go along - drop their advertising, cut off their servers, purge their DNS, ban their accounts, shame them relentlessly until they disappear.
Most US companies these days have no morals, and are easily influenced by these tactics due to greed and fear of being targeted themselves. Silicon Valley and the majority of the big tech companies seem to be especially vulnerable to this, probably due to their own employee demographics.
What many of these companies don't understand, possibly because they live in a relative 'bubble' surrounded by those who think similarly, is that there are a lot of us out there who not only disagree with this type of behavior, but will actively NOT use the services of any company who supports these types of tactics.
You aren't the customers in this situation. For every 10,000 of you who don't pay even pay GitHub the $7/mo for a subscription, there's a 3000-seat behemoth who pays $70k/mo for a GitHub Enterprise license.
You're principled minnows to that one profitable shark.
These companies understand profit, and that's where they derive their morality. I'd say it's probably more accurate that most US companies simply don't share your morals, not that they don't have morals at all.
Follow the money. This is a much more useful lens to analyze the situation than to consider the left/right political spectrum.
Sure, but that "lot of us" out there is a much smaller and usually much rowdier group of users that time and time again companies have been happy to wash their hands of. You're not profitable enough (and I'm not even getting started on the morality or ethics side of this).
I have assumed that many tech companies, especially in California and other liberal strongholds, hold this opinion. Like I said, they live in their insular bubbles, and imagine that the rest of the country is either deplorable and poor or they share their views.
Meanwhile, I work in a relatively conservative industry that also happens to have one of the largest budgets of any 'company' in the world. I have seen first hand when vendors were being evaluated for multi-million (or even billion) dollar projects, both Google and Github being crossed off the list without a second thought due to some of the publicly made political statements and actions of their executives and employees.
Why do people always feel the need to bring "the left" into this? Wanting to silence people who disagree with you has nothing to do with either the original definition of "left" or the parties considered "left" these days.
The same kinds of "censorship" that you talk about coming from "the left" can be found in extreme parts of every ideology. Conservatives (probably of the rich and christian variety) have pushed many platforms to completely remove all even slightly adult content (the latest example being Tumblr), all sides of the political spectrum have been pressuring sites like YouTube to the point where no political discussion from any side can be monetized...
This is not an issue of political sides - it's an issue of politics (and society) in general.
As for the part about companies not knowing about the people who don't approve of this behaviour: they do. They know exactly how many of us there are: not enough. Losing even a single big investor will make a company lose more money than if everyone who disagreed with them completely stopped using their services.
This comic is abused so much that I wonder if Randall would ever consider a follow-up poking fun at how it's wielded. It's meaningless in a normative, rather than legal, conversation such as this one.
You making the argument that to make some religious customers/investors happy, it's ok to mistreat LGBTs. After all, they are such minority segment and, you know, we are all here just for shareholder wealth maximization.
"Banning a book" colloquially means that nobody is allowed to read that book, it conjures images of book burnings and the gestapo searching your house for contraband. "Banning" a repo here means, "Github is not offering you free resources to develop your code. Fortunately, you're using a distributed source control management scheme so everyone has a backup. Please take it elsewhere."
Agree strongly with this. If a repo is public and gets banned, I think it's reasonable to expect that the community can know why, regardless of the rights or wrongs of the decision.
It seems reasonable to expect this, but it can fall down in practice for several reasons:
* Sometimes legal counsel provide advice that there should be no further response to the individual or organization. Often technical people don't understand this situation, but it doesn't change the merits of the legal advice. In smaller organizations a leader might take a chance in further engagement, if they think it's helpful, but it's unlikely a large organization would expose themselves to this risk.
* Breakdown in internal response processes. You'll find that many people are really uncomfortable in these situations (e.g. compliance team shut down service, but don't "own" the response.) Unless the legal team has written a response and instructions on how to deliver it, you will often see people in organizations avoid giving the response. Things get passed down as low as they can go which doesn't help because there is less experience with handling tough situations. Very often some poor person with support ends up having to give the response and they basically ignore it because they can avoid the situation. This isn't very professional of the organization, but it's a reality.
We're living with transparent juridical system and it works fine. Imagine that you could be thrown to jail without explaining a reason. That would be outrageous.
Yeah, criminals are always arrested and convicted. /s
It's a balance. With something as essential as human rights and personal freedom, people (tend to) err on the safe side. Online moderation can err on the other side, since consequences are relatively modest. If you get banned on GH, move to Gitlab or host your own, that's hardly a tragedy.
That is exactly what I do. I use self hosted solutions for my source code repositories. I just can't digest my code being handled by some other entity. Too important.
Amazing that you got downvoted for this. I pay for code hosting precisely because I want to see an ecosystem of code hosts, and monocultures are dangerous.
Well I've never downvoted a single post no matter how much I disliked it. Personally I consider this a kind of weakness and the whole system as promoting herd mentality. But whatever floats their boat.
Exactly. Screw around and try to game/skirt the law IRL and the risk is way too high that you'll goto jail anyway. There are usually no consequences for doing this online.
Compliance with the spirit is the objective. Sometimes the spirit and the letter differ for any number of reasons (many of which are completely reasonable).
People tend to get pretty upset when someone is very clearly complying with the letter while flying in complete opposition to the spirit, and it's not always an easy fix.
In that case, it sounds like the letter needs to be fixed. It's not fair to expect people to follow an ephemeral ideal of what the rules are rather than what they're told the rules actually are.
Like I said, it's not always that simple. When it's not, something less than 100% transparency allows one to look at the given particulars of a case and determine whether or not someone is simply trying to evade the spirit of a rule or not. It gives enforcement actors a little lee-way that they wouldn't otherwise have.
One of the worst things about engineers in general and HN specifically is we all pretend that law is executed like code, in a vacuum, idempotently based on the inputs. That's was, is, and will never be the case.
Abuse can be exposed and punished, and very often is.
That's why the letter of the law needs to be updated to better reflect the spirit. Imagine if police could arrest you, and keep you, without telling you why. That's something that society figured out a long time ago isn't healthy.
> Imagine if police could arrest you, and keep you, without telling you why. That's something that society figured out a long time ago isn't healthy.
The judicial system that backs it is a massive beast. If someone wants that level of assurances, they should be paying thousands of dollars for a github account. You get the level of perfection you pay for.
So far it's been mostly small / independent developers or organizations that were banned, and Github has Microsoft behind it, a $125bn / year revenue company with a legal team 1,500 strong (https://www.bizjournals.com/seattle/news/2019/12/02/how-brad...). I don't think fear of litigation is the issue.
The very first thing a corporate lawyer does is proactively prevent litigation through protective policies that specifically do NOT emphasize transparency.
Whoa, wanted to jump in here! SS13 is, in my opinion, one of the best games of all time when it runs well. Not very many people know about it.
I worry about the community dying and losing my favorite game, but have taken solace in the fact that the source will always be publicly available. If it was banned from GitHub, that's a major problem.
Is it? There are several GitHub alternatives, many completely free as well, and none of the source was lost unless all the maintainers and contributors also delete their local copies.
The alternatives don't have the mindshare that GitHub has when it comes to open source software. If the community around the game is already weak, moving to another provider will likely weaken it even more. The source won't be gone, but that's only half of what matters.
If it was the bad words/slurs, could that have been resolved by hiding them behind some basic string manipulation (ex. a caesar cipher)? I can see how GitHub wouldn't want a public repo to have objectionable words, but can't imagine the harm from obfuscating stored copy.
> I can see how GitHub wouldn't want a public repo to have objectionable words
I can't. Does GitHub really have nothing better to do than to play nanny cop because I used a naughty word in my code? Are brainfuck interpreters now off-limits? How about drivers for teledildonics hardware? Or libraries specifically for detecting and filtering swear words? Or maybe I just want to vent a bit in a comment every once in awhile because of some annoyance with the language or target platform or problem to be solved?
Fuck that and the horse it rode in on. We're all adults here (well, or possibly teenagers, but let's face it: they've probably already heard much worse at school).
If only we knew that 4 days ago when we first got banned, and not, well, 4 days later.
The issue is github works by report only.
You can do what ever you want in a github repo, but if you make a video game on github, and ban the wrong person, they can just go through your repo and look for ToS violations to troll you.
We are literally removing the in game chat word filter for the n word out of fear it could be used to git us banned again by somebody else mad their buggy pr got rejected or their character got banned in game for breaking the server rules
If that's official GitHub policy it's both unworkable and exceedingly ignorant of how people use the English language outside of the US. GitHub should have no business telling people how to write their source files.
