Interesting. For projects that don't need a "community" we've been using Amazon CodeCommit.
We don't like to use Github anymore because of the politics there. Example: We had a project involving software filters that included this sentence in the description:
> Phase is related to time, but a pure time delay does not involve any phase shift. A pure time delay or "group delay" is constant with frequency. Phase shift varies with frequency and can advance or retard as the frequency changes.
We got a note from a woman who works at github who was apparently scanning open source projects for "offensive" words. Our project was flagged for the use of the word "retard".
We removed our project and never used github again.
Just to be clear, you're saying: "If large company x hires a single incompetent person, that is a good reason to never deal with said company again." I think that's the parents interpretation of OP and hence the "seems like a disproportionate response." (I agree it was absurd to flag the content).
That's oversimplification. Even if it is just her own personal incompetence, if GitHub just hired her and let her loose as some kind of independent repo ombudsman to scold users, then yeah, that's a terrible management decision that warrants reconsidering your relationship with them.
The point is the same though. Resting a project's livelihood on a company that employed obviously incompetent process/humans/scripts is a risky move, and removing that livelihood from that company makes total sense.
It doesn't sound incompetent to me, there will always be false positives on that scale. Did they remove the OP's repo, or simply send them a message? If a legitimate email gets caught in your spam filter, do you switch email providers? Github and their process/humans/scripts don't sound like the incompetent ones in this scenario. I would be more concerned that my livelihood depended on such an over reactionary company.
"Hey boss,I tried cloning the Github repo of project X, but it says it can't be found, and I noticed our org page on Github is gone."
"Oh yeah, they sent us a message about the use of the word 'retard' in when referring to phase shifting."
"Oh, so they deleted our organization?"
"No."
"Oh? Well why can't I view it, weren't the projects public? What did you say to them?"
"I didn't say anything, I moved us to a different host."
You're injecting a lot of assumptions into your narrative.
My point is simply: if Github is flagging content based on perceived personal morals (not ever being allowed to use the word "retard," even in legitimate contexts), some choose not to associate their work with them.
Even then, they still wrote the automation. It would be a prudent move to involve human review before emailing someone and accusing them of making a bigoted slur.
When there are numerous alternatives to GitHub why would I continue to use them? Clearly they were not putting the time and effort in to both train their employee and review the content (simply reading the sentence would give context), why bother dealing with them?
Also why as a code hosting service would they be getting involved with the politics of the content of projects they host?
Absurd being the key word. This isn't a simple misunderstanding we're talking about, it's a company employee using company time and resources (i.e. sanctioned by the larger company) for something that can charitably be described as misguided.
It raises questions about the internal company culture, their commitment to pragmatism (doubly important for a source code hosting site), and their priorities. Any developer could have told this person that a string match does not equal a linguistic match.
Small signals matter; they're usually all you get out of large orgs.
People who represent themselves as employees of a company represent the company. If the company permits incompetent employees to behave like this on their work profiles, that's not an unreasonable deal-breaker.
I actually assumed that this message was automated – "We found on objectionable word in your project, please don't use it."
It just seems that this problems is remarkably easily solved with a simple "We are using a technical term correctly, maybe you should consider adjusting your automated emails."
I do get the impression that people generally seem to be all to eager to take offence, and it strikes me as a bit ironic to flounce off in a huff as a result of a badly-judged automated process!
I find it worse. Github provides a great service and certainly has provided an incalculable value to os, but at a project level, it doesnt provide nearly enough value to dictate word choice, etiquette or style. Yes, the service is free, but github needs high caliber os projects more than they need it. It dosnt even matter who has an edge, automating spam at a user who says a naughty word is silly.
Idk, i dont understand why they would even care. Unless it reached a very intolerable bar, i dont think this is necessary
Go read up on Github user: CoralineAda [1][2] Then consider that GitHub hired her. I would not be surprised if the this automated script and her employment were directly related.
I'm not terribly surprised. GitHub is, after all, a company which changed its logo to celebrate a highly political U.S. Supreme Court decision and which has had issues with left-wing racism (http://www.businessinsider.com/github-the-full-inside-story-...).
I really don't want this to be a witch hunt (not seeing that here, just wary of pointing out individuals)—this is exactly the type of thing a business should take responsibility for as a whole. This was a process failure, IMHO.
I think you guys missed my point. In pointing out one individual, I was pointing to the organization as a whole. Github supports this kind over-sensitive social-activism. Example A: dragnet language-police script. Example B: hiring controversial, social-activists.
For better or worse, the appearance of coming from a real person seems to be the trend in marketing automation. I think many people, especially outside of tech, believe they are interacting with a real person most or all of the time, when in reality, they only are during exceptions to common requests. But people like interacting with real people more than they like seeing "automated message blah blah".
Related: The company is slipping my mind right now, but one startup changed their email name in automated messages from male to female and saw improved metrics as a result.
It's quite disappointing to see people defending a company that hires bigoted political activists to crawl users' projects in order to harass users for perceived political transgressions.
And what level of access do these on-staff political activists have to private user information? What measures, if any, does GitHub have in place to prevent one of their political activist employees from using privileged information to go after perceived political enemies or even just retaliate against users who protest being harassed?
It's quite disappointing to see Hacker News users use hyperbolic language and throw out nonsensical arguments.
What is "bigoted" about trying to avoid the user of the word "retard"? I'm not saying that what Github is doing is wise or sensible, but I'm struggling to see how it is bigoted.
> What measures, if any, does GitHub have in place to prevent one of their political activist employees from using privileged information to go after perceived political enemies or even just retaliate against users who protest being harassed?
I would imagine the same as any other company. There's nothing that would indicate otherwise, aside from your own biases against Github.
It seems like the action from the github employee was disproportionate. That sounds like a very risky environment to host your product on. The response seems to match the proportion of the action.
Perhaps to a comment from a random user, this would be disproportionate. But when it is a GitHub employee sending these messages, someone needs to communicate the issue to them in metrics they will understand, like projects leaving.
Assuming Github's ToS applied here, it's unlikely that the project owners would have succeeded in keeping the content as is on github by arguing for it. This is regardless of the legitimacy of the content. Being on Github isn't critical for every project, and we need more projects that have their primary location on Bitbucket, Gitlab, etc. instead, in order to break the Github monopoly by making it normal that it's fine to use a different site for a project.
It cannot be healthy to have all our eggs in one basket (Github), regardless of how featureful and nice Github is, especially so given the availability of viable options. To that end, we should not complain when a project, say, evil-mode is on bitbucket, and similarly not ask a project to open shop on github.
I mean, if we don't work towards a fully distributed project model ala fossil-scm (via things like git-appraise), we should at least use alternatives when we can, and not strengthen Github's monopoly.
Github does have the lead in terms of exposure and discoverability. To take advantage of that, I've left placeholders in all my projects that are elsewhere now:
https://github.com/caseysoftware/marvel-php
If you find it on Github, you can find out how to get started but everything else is on Gitlab.
we're not talking about anonymous web forums, we're talking about a git repository. It isn't "silly" to be bothered by random github employees scanning and flagging keywords
How about it's offensive that a company that bills itself as an open source hosting company even employees (incompetent) morality police? Really, scanning projects for naughty words?
Looks like a very big sign that more warnings will come later, and you'll have to spend time fighting such things now, and start worrying about your data be cut from the site without previous notice.
Going away is the correct response, fighting the decision is never useful. One might wait a bit to see if it's a fluke, but that will certainly depend on the GP's evaluation of the company.
I stopped using Github when they hired blatant racists as part of their 'social outreach' team. The infestation of politically-correct politics drives talent that can't be bothered with such cruft away.
I was there to code. Not worry about if I hurt some oversensitive person's feelings for referring to master/slave copies.
We're all on Github to write code, but I'd argue you are dangerously close to throwing the baby out with the bathwater there. It seems perfectly reasonable for someone to say "hey, this term you use might be offensive/makes me uncomfortable, please consider changing it." It's also perfectly legitimate for you to say "I don't want to do that".
I just kind of wish that everyone would be a little less sensitive about the prospect of other people being offended by things that they are not. Political correctness, in the sense of "try not to say things that are offensive" sense, is probably not the worst idea in the world!
It's "the worst idea in the world" because it perpetuates the idea that certain words or thoughts are taboo.
This doesn't mean people stop having them.
This just means that they are labelled/judged if they have them.
Guess what happens then? People keep it to themselves. And then they vote for Trump. Why? Because, he is a ridiculous loudmouth who is very out of touch with most of reality. But, he doesn't feel the need to be censored. And to people who haven't spoken out due to fear of reprisal, this is an admiral characteristic. He gives them validation.
>It's also perfectly legitimate for you to say "I don't want to do that".
At which point you get harassed by a mob, spammed with issues that clutter up your project, and get emailed death threats.
Github is allowed, of course, to choose which projects they decide to host. Demanding the removal of the word "retards" [0] or a project would be shut down drove at least one developer away.
So no - that isn't "perfectly legitimate" for that community. "I don't want to do that" is not good enough.
I think it's reasonable if only so you can avoid the overhead of needing to deal woth censors. There's also a lot of overlap between open source and free speech, so I can understand a natural distaste for censorship in general.
To olay devil's advocate, what's wrong with offending people who read your source code? You can't harass people through source code in a meaningful way because you can't force them to look at the code through natural interaction with GitHub itself, if they see a project that is insulting, they can just ignore it.
It's not like a social media site where you can ensure your victims' feeds will be polluted with your insults.
Maybe if they had some allegiance to or belief in Github fixing the issue, talking to employee, retraining them, it might be worth. But when there is another project that is just as easy to use, it often just easier to drop it and switch.
Also that interaction with Github, can be interpreted as a pointer to larger systemic issue (for better or for worse) which the customer doesn't believe will be solved with a simple email.
Given no other information I would agree. But given other information both about Github and about the kind of processes that seek you out to tell you that you are being offensive, I understand. Maybe this was a one time thing, maybe it will be another donglegate. If it is easy enough to move, why wait and find out?
A customer can choose to take their business elsewhere for whatever reason, at any time. I mean you can speculate as to whether it was disproportionate or not, but the only determinant here whose opinion matters is the paying customer.
Typically US, blocking the use of such words. Everybody says 'fuck' in the US, constantly, but whey you use it on TV or post it online, all hell breaks loose. It's just a word folks, get over it. What is the worst that could happen when you see 'fuck' or 'retard' somewhere?
I'm also curious on how many downvotes I will get for this, because the polite way is to use f*ck or something, right?
It's a political act to publicly declare something along the lines of, "I quit because of their politics."
Fortunately, people like me see such declarations — "Hmm, I'd better now reactivate my paid GitHub subscription, and stop pushing people to competitors. Their attitude is improving!"
This was also one important reason why I am switching. The price, the enjoyment of hacking are primary reasons, but the politics pathway bothers me as well.
Yes. She indicated that it doesn't matter how the word is used, some will see it as aggressive.
As she appeared to be a non-technical person mysteriously employed by GitHub as some consolation to some organization (IIRC it was called the "Ada Initiative") I realized that github is no longer a technology company. It's a Social Justice company. Good luck to them--I don't even disagree with the basic premise--but I like my git repo companies to be worried about git.
Some of the commenters here assumed that the message was automated, you're making it abundantly clear thats was not the case. In fact she defended her position. Did you try to escalate it? This is absurd.
[I'm not OP] Does one have to always make sure not to mention gender in English? I'm from Poland and in Polish you mention it all the time unless you specifically go out of your way not to do so. Before I've seen people criticized for translating "programistka" as "female programmer" and there was that proposal for some NLP library the other day, to drop any functionality detecting gender... And doing that to Polish (or any Slavic) NLP library would make the software useless: emperor/empress,prince/princess,sorcerer/sorceress - it may sound like fantasy to many but this is the reality for language with grammatical gender. I'm not saying there are no biases against women in Eastern Europe, but no one here would find mentioning gender offensive or inappropriate, in fact the opposite is true... I also don't remember such rule being mentioned in my English course.
It's not usual to avoid any mention of gender in English; some people avoid it for political reasons. But when you're translating a gender-marked word into an English equivalent which is not marked for gender, it's not idiomatic to add another word to indicate gender unless it's relevant to the topic at hand.
English lost most of its grammatical gender centuries ago. Since the 1960s and 1970s the parts that remain have become politically fraught in some circles, but usage varies by community. Unless you're deliberately wading into that fight, I'd stick to the way you were taught.
I see, recently I'm seeing people being called out on thoughtlessly mentioning the gender. The 'female programmer' was a lazy translation, agree, but I would've also used 'woman' to describe person who wrote the that e-mail if I know that for a fact that woman was the author and question by kmonsen really got me thinking if that would have wronged someone...
On a side not, I do miss that added context in English, especially in novels, when it's hard to guess just based on the names of the characters. It makes it easier for me to imagine them and evaluate interactions between them.
Perhaps she signed-off her email/post/comment with a female name? You know, it is perfectly reasonable to assume gender when the name is obvious. E.g. the odds of a person called "Samantha" being male are miniscule compared to it being a female. Because in both cases, "it" is a person we are talking about.
We can't just erase gender. People think in gender, and they consequently speak it. I don't think twice about mentioning "some guy" commented on my post, for example. Just the other day I decided to be even and refer to a hypothetical person by using "s/he", and even got called out on it.
I'm confused by your question. A lot of organizations have similar concerns. Do you believe that all discussion of diversity is equally nonsensical, or is there something about Github's stance that strikes you as especially so?
I'm in an asian country where racist is (mostly) not a thing, so why people cares about those rates is a mystery for me. In my thought, if there are 2 candidates for a leader role, and they decided "oh, we need more women in leader rank, we should chose her", it is just wrong. Race/gender shouldn't be even considered as a factor in those situations.
so why people cares about those rates is a mystery for me.
If someone rolls a die 10.000 times, and 75% of the time it lands on a six, then instead of "not caring", many people will care. Particularly those that are not betting on the six. Though the people betting on the six have a vested stake in "not caring".
In other words, it's a matter of bias.
If talent is not correlated to race/sex, then we can expect that the number of people of various races/sexes that are represented in the industry should roughly match the general makeup of the population (ie it's a mostly random distribution). Instead we see that this is not case. So either, talent is correlated to race/sex or there is a bias somewhere in the process.
> If talent is not correlated to race/sex, then we can expect that the number of people of various races/sexes that are represented in the industry should roughly match the general makeup of the population (ie it's a mostly random distribution).
Employer hiring should match the candidate population, not the general population. This is a subtle but very important difference when we also see, for instance, gender bias in graduation rates of STEM fields. (I'm not claiming that this accounts for all bias.) Really, when you say "there is a bias somewhere in the process", it must be remembered that the "process" being spoken of is the entire process that leads to a person being a valid candidate, from schooling to prior experience. (It's hard to get hired as a senior if you can't get hired as a junior first.)
First of all, there is the moral case: Certain groups have been kept away from positions of power & wealth to such an extend that taking remedial action is warranted.
Secondly: there are obviously some differences in the behavior, knowledge, and culture of different groups and it is beneficial for a company to include those and profit from it.
But... you can't have gone through more than 14 years of life without having heard all thread a few times, can you?
So you're saying there would be more mentally challenged people committing changes to open source projects, if we didn't use the word "retard" in a scientific sense referring to digital filter algorithms?
I think that's a bit of a simplistic way of looking at it. There's an easy explanation of why this approach exists though, if you agree with the following statements:
- Race, gender, sexual orientation, disability etc. do not generally affect an individual's skill as an employee
- People from certain groups are under-represented in the industry
- Under-representation generally reflects an underlying problem somewhere, because one would expect various groups to be represented broadly in line with wider society
- One of the problems that can cause under-representation is an existing bias in an industry, which can discourage other members of a group
- One of the approaches to solving this is to help ensure that those groups feel comfortable in the industry
- One of the approaches to helping them feel comfortable is to ensure that communities are inviting, open-minded, and aware that things they say or do may upset or discourage other people, even though they are do not find those things offensive themselves.
It's about observing a trend which seems wrong for various reasons, and attempting to help solve it. I don't doubt for a second that many people go completely overboard on this issue, but outright rejecting any attempt at using, say, gender-neutral language in open-source projects just seems to go completely in the other direction.
We don't like to use Github anymore because of the politics there. Example: We had a project involving software filters that included this sentence in the description:
> Phase is related to time, but a pure time delay does not involve any phase shift. A pure time delay or "group delay" is constant with frequency. Phase shift varies with frequency and can advance or retard as the frequency changes.
We got a note from a woman who works at github who was apparently scanning open source projects for "offensive" words. Our project was flagged for the use of the word "retard".
We removed our project and never used github again.