Do you think a company in 1930s Germany should have ethically refused to provide software that was used in concentration camps? [In fact, there was a bit of "IT" then, used for such, but it was provided by IBM. But to make the analogy closer, let's imagine a hypothetically Germany company].
("Companies" don't do anything by themselves, so I guess the question is if the decision-makers in such a company should refuse to sell software to the German government for such purposes, and if the employees should try to pressure the decision-makers to).
If we agree that in that case the ethical choice is to refuse to supply the software, and that it would in fact be unethical to sell software for such a purpose...
Then we already agree that there is some case where a company should refuse to provide services for 'political' reasons, even to a government agency of the country it's located in.
So it's no longer a question of if a company should ever be "expected"to do this -- but if they should in this case, if this particular scenario is such an example. People can disagree on that, can think that obviously this is unlike the Nazi example, that this example does not rise to that level. I'm not trying to insist that this is definitely a "Nazi-like" example.
But once we agree there is at least one such case, it's not a categorical dispute about whether business decisions should be "politisized" ever -- it's a debate about the particular ethics of the specific situation we (or github) finds themselves in, if this example is one that requires us to ethically refuse cooperation or not. Very particularly. I think that is a fine debate to have. I think the debate about whether a company should ever do this sort of thing is not so much, because really we should all be able agree there are some lines that should not be crossed there, there are some cases where, yes, a company should be expected to refuse service to it's own government, once we examine the historical examples that are obviously beyond the lines.
If we agree that there are any circumstances where a company is ethically required to refuse selling it's software to a national government, then we can look at the particulars of any specific instance and decide if we think that's such a case, but we don't need to argue about whether it's appropraite to "to expect that a _federal agency_ will be denied service from a private entity," becuase we've all agreed that there are at least some circumstances where it is.
What is in fact inappropriate is to argue that a company should never be expected to deny service to a government, no matter what they do.
I've posited the same argument around violence. Most people, in my experience, wouldn't fault a Jew for violently resisting the Nazis. In fact, many resistance groups are held in very high regard. So immediately the questions necessarily moves from "is violence sometimes allowed?" to "when is violence allowed?". Which, if nothing else, is very much a more interesting discussion to have.
However that is a dicussion that is, at most, tangential to this one.
I think its appropriate to refuse to work with nazi germany, apartheid south africa, khmer rouge, etc. But I don’t think ICE is anywhere near those regimes.
If GitHub staff equate them, then I question the logic of any organization that makes those comparisons. Mainly because it they aren’t operating rationally then perhaps next is DEA, NRA, non-GPL contributors, etc.
I don’t see any good where companies try to work or not work with specific organizations based on very niche boycott campaigns.
I think there's a moral component and a practical component. The latter becomes very difficult when taking a stand against a government entity in your home jurisdiction, since the offending government may put you out of business or something.
For example, in Nazi Germany (which I agree is a useful example to reference), a company that refused service to the Nazis would have had its executives murdered and been taken over by the state within a day. (My guess based on my recent read of The Rise and Fall of the Third Reich, but I may be off).
On the other hand, refusing service to North Korea (sanctions aside) is a clearly ethically correct thing and offers little practical risk for companies based in the West.
I'm even betting that it is a small vocal minority that is against it.
So, how should GitHub proceed in your opinion?
Or perhaps the checks are very common and don’t usually result in public letters and whatnot.
ICE hasn’t even been accelerating in their harm to people. If anything they’ve plateaued or declined. I’m not sure how to best measure fascism in ICE, but long term internment, death, and mistreatment is probably a start. The death rate has dropped, despite many more detentions [0]. I tried to figure out some metric around number detained, but I think that may also be influenced by number of crossings and not straightforward to risk of becoming Nazi Germany.
How do you think we can better measure this risk within ICE?
[0] https://www.cato.org/blog/8-people-died-immigration-detentio...
IBM literally sold them information retrieval technology they used to keep track of concentration camp inmates. So clearly the important decision-makers did not agree.
I guess it is like any other question of what an organization should do. Those with the power to make decisions will decide based on some combination of their own ethical standards or (more likely) what think they is "good for the business", where PR as well as employee morale are components of that. The employees without the power to decide directly can organize to try to convince those with the power to make decisions of the correct ethical choice, or of what's good for the business, or to try to change the calculus of what's good for the business by effecting PR and employee morale etc.
I mean, this is kind of just a description of how human organizations or collective decision-making works....
I personally think that what ICE is doing is absolutely immoral and unethical, putting people, including asylum seekers, into (yes I think it's appropriate language) concentration camps, without a trial or hearing or access to a lawyer, in unsafe conditions (covid makes this even more extreme), separating children from parents, etc. If we looked for an external arbiter of this, I think it also clearly violates international law and agreements on the rights of migrants and refugees, so that could be another argument, don't sell software to organizations that will use it to violate international law.
I personally wouldn't at this point call for github (or anyone) to avoid business with the federal governmetn entirely -- just to avoid your products being used for the programs that are violating international human rights. That is, avoid doing business with ICE, for sure. Maybe with DHS in general, or particular programs/units in DHS.
I can't make you agree. People disagree, this is part of human life. But if I were in github, I'd be working to convince other co-workers of this, and to convince decision-makers leaders of it, as those in the article are presumably doing (I still can't read the article because paywall, so I'm only guessing as I think most commenting are!) This is how humans in organizations work.
However pretending that a vocal minority trying to push an agenda isn't bypassing democratic decision making is ludicrous.
Of course decision making in companies is not democratic, but if as a partisan employee you're trying to impose your opinion instead of trying to implement some kind of democratic process to consult other employees anonymously then you are no different than any authoritarian movement.
As you note, neither github nor hardly any company has any kind of "democratic decision-making" at all. It's just not how companies work in America. So I don't understand hwo you can be "bypassing" something that doesn't exist.
When the executive and other decision makers at a company make decisions with no democratic decision making whatsoever, do you think that is "no different than any authoritarian movement"? Personally, I think saying it's "no different" is a bit much, it's a differnet in a bunch of ways -- but I think it's not great, I think we should work to put democratic decision-making in all companies. Do you agree?
In the meantime though, we don't have that. So what are people who want to have impact on the decision-making supposed to do? Even when there is democratic decision-making, it's considered normal to try to convince your fellow-decision makers of things. I don't understand where you are drawing the line between allowable ways to effect decision-making and "like an authoritarian movement". I mean, nobody's threatening anyone with violence, are they? What means of persuasion or pressure are according to you allowed, and what means are not?
It doesn't help that I don't think either of us has actually read the article, because it's behind a paywall? So I actually have no idea what methods or persuasion or pressure they are using. Do you know more than me? If so, feel free to tell me (ideally with a link to another article so I can get it from the source), and explain why you think those methods have crossed the line into "authoritarian"?
Since I don't know specificcally what they are doing, I can't really defend it specifically. Like, if they were beating people up who didn't agree with them, I'd definitely agree that's something authoritarian movements do! (I still don't think I'd agree it's "no different", there are always differences, that's a kind of lazy thing to say, "no different"). But I don't see any reason to think or assume they are doing that? Do you have more information than me? I'm confused why you are assuming they are using unethical methods, or even what you think those methods they are using are. Are you saying just that going to the press makes it "no different than any authoritarian movement", but if you just talked about it quietly inside the company that would be okay? That would seem an odd distinction to me.
If they want their opinion to be taken seriously, they should setup democratic vote among employees to collect votes anonymously about the issue at hand.
Proceeding otherwise is simply trying to convince colleagues but also removes the possibility for other employees to have their voices heard anonymously.
Fortunately, in fact, plenty of organized people have been taken seriously, and have effected change, through other means.
I think this is an important distinction that needs to be resolved in some legal means. many people have differing opinions on this, so having some systematic process is useful so we don’t end up wasting energy on a very loud minority.
This is a complex issue and it’s something that companies should be good at. In this case, having some systematic way of making this decision public is helpful to me in planning who to purchase services from. Some generic statement that the decision was “based on company values” (that can change at any time) makes it hard to predict future decisions.
This ambiguity over how products are run is a big part of how I struggle to predict what Google products will continue or be, arbitrarily to me, discontinued.
By possibly bypassing the majority's opinion. Also called authoritarianism. Putting pressure in a non-democratic way is silencing people's opinion and similar to censorship.