Requiem for GitHub
hintjens.com
hintjens.com
Ironically, that's why the money was invested in the first place, to be able to afford people like that, because the investors believe that creating a gamified system for money-driven people will result in a payout.
Github could have continued to grow organically instead of taking $100M. Since the $100M there hasn't been any significant new functionality on the web version of the site, and there have been fewer blog posts from the early team of people who many of us met and respected.
I assumed something was awry when Kneath left, he was a symbol of the culture at GH that was a legit, bootstrapped startup culture. He loved the product and evangelized it in a way that was clearly a labor of love.
The sexual-harassment stuff is unfortunate, but organizations of primates have things like that happen now and then, it's just rare that the victims decide to go public. This does not excuse it but it doesn't necessary say anything about the culture (I've seen female-founded, progressive firms tolerate very bad behavior from male employees).
Maybe Github should split into a corporate division and an open source division. The OSS division could make all of its work and decision making distributed and public, and the corporate division could do business development and attack new market opportunities.
We live in a time where moral hazard was effectively removed for people at the top of the economic food chain. The 2008 financial crisis should have wiped out the bad players, leaving them bankrupt and starting from the bottom to crawl their way back up. The bad bets, bad ideas and fraudulent practices that should have served as a cautionary tale to others instead became best practices. I don't think it is so much that money attracts the wrong kind of people, its that the wrong kind of people still have power because of crony capitalism.
Horvath had what she referred to as an awkward, almost aggressive encounter with another GitHub employee, who asked himself over to ‘talk,’ and then professed his love, and ‘hesitated’ when he was asked to leave,” TechCrunch reported in March. Hurt from my rejection, [he] started passive-aggressively ripping out my code from projects we had worked on together without so much as a ping or a comment,” Horvath told TechCrunch in an email. “His behaviour towards female employees … especially those he sees as opportunities is disgusting" (http://www.businessinsider.com.au/github-harassment-story-20...)
In case anyone actually cares about facts and evidence, rather than pushing some social agenda, here are the results of the independent investigation in to the matter: https://github.com/blog/1826-follow-up-to-the-investigation-...
Harassment does say something about the culture in which it happens. What you're pointing out, correctly, is that it's not exceptional for this kind of stuff to happen - that Github is probably not much worse than most other tech companies in this respect.
As you say, this is no excuse. Change has to start somewhere; we can't keep saying "oh, it just happens, you know, it's not worse here than anywhere else".
> Maybe Github should split into a corporate division and an open source division. The OSS division could make all of its work and decision making distributed and public, and the corporate division could do business development and attack new market opportunities.
I would like that, but how would it make investors money? If the answer is "it doesn't", then it won't happen.
And yet this single person's claim was enough to force out the CEO and tar GitHub with the "harassment" reputation.
We live in interesting times.
My default assumption is that when there are claims of harassment, there is some substance to them, even if the specifics are wrong. Bringing forward claims of harassment in the tech world turns your life (and that of the accused, if their name is publicized) into a shitstorm; people don't tend to do that without reason. But this is just a heuristic. It looks like Horvath's claims of harassment might be unsubstantiated in this case.
It's hard to tell, though, because most of the things she complains about (with the exception of the code-review stuff, which appears to be just false, which is worrying) is unverifiable outside of he-said/she-said. Note the noncommittal wording: "The investigation found no information to support misconduct or opportunistic behavior by the engineer against Julie or any other female employees in the workplace." No information to support suggests "neither evidence for or against". Which is natural. Harassment is hard to prove or disprove precisely because it usually doesn't leave evidence.
> there was an apparently thorough investigation by a reputed external investigator who found that the claims were false. [...] And yet this single person's claim was enough to force out the CEO
Specifically regarding the ousting of the CEO, let me quote the investigation you linked:
> The investigation found Tom Preston-Werner in his capacity as GitHub’s CEO acted inappropriately, including confrontational conduct, disregard of workplace complaints, insensitivity to the impact of his spouse's presence in the workplace, and failure to enforce an agreement that his spouse should not work in the office.
So, you think we are alle guilty of harassment, until proven innocent? I wonder what my fellow countryman Cesare Beccaria (https://en.wikipedia.org/wiki/Cesare_Beccaria) would think about this kind of mindset.
> Harassment is hard to prove or disprove precisely because it usually doesn't leave evidence.
But we live in a legal system that is based on proofs, not on suspects or claims.
> and failure to enforce an agreement that his spouse should not work in the office.
Anybody knows that it's safer to jump in the mouth of a T-Rex than doing something like that. I think this is the reason he resigned: to save his marriage.
But if we're interested in whether the industry itself has a problem, in whether a problem exists, then we should take women (or anyone else!) at their word when they say they experience harassment, absent strong evidence that they are lying or mistaken. Otherwise, we are violating the same rule: assuming they are guilty of lying until they prove themselves innocent!
You're right, and legal protections exist to protect individuals under these circumstances. It's a primary responsibility of Human Resources department to document such occurrences and justifiably fire bad actors when they occur so things don't devolve into 'he said, she said' chaos.
Instead there was no documented attempt at intervention. The woman in question reached out mass media to garner sympathy (a fireable offense in and of itself); 'he said, she said' was accepted as fact, and the man was asked to step down as a result.
"Better a murderer go free than an innocent be hanged, and so forth."
That's quite the slippery slope. The burden of proof exists to protect individuals from unjust accusation. Libel is a crime for a reason.
It's the legal and ethical duty of an organization to prevent from blindly taking sides in the absence of proof. Toxic, spiteful, aggressive, conniving, manipulative personalities aren't exclusive to any gender, race, creed, etc.
Considering the evidence after the fact, it seems that this was an organizational failure. As a result it set a bad precedent. GitHub lacks the ability to protect the well-being of its employees and worse, protection is exclusively granted to women.
To quote:
"Even so, we work in a world where inequality exists by default and we have to overcome that. Bullying, intimidation, and harassment, whether illegal or not, are absolutely unacceptable at GitHub and should not be tolerated anywhere. GitHub is committed to building a safe environment for female employees and all women in our community."
GitHub should be committed to building a safe environment for all employees. The resulting outcome proves that their neither willing nor capable of protecting the well being of their employees absent of institutional bias.
-----
As far as it concerns Tom Preston-Werner, he failed in his duty as the leader of the organization. From all outward appearances it seems like he's well aware of where things went wrong. Unfortunately, the damage is already done.
As much as we'd all desire to be friendly and personable with our colleagues, leadership should always maintain a degree of separation from those they lead. The military has clear rules about fraternization, well established businesses usually have similar rules. Startups are, well... startups. Their greatest strength -- to break common convention -- is also their greatest weakness.
It could be a PR and branding expense, to prevent a "Twitter Kills Ecosystem" event that drives away developers who have been evangelizing Github within large enterprises.
Preventing investor losses = making investors money.
It seems like your heart is in the right place but the quoted statement reads to me as a total non-sequitur. What model are you using to justify your calls for GitHub to hire 9x it's current number of employees?
What would these 4.5K new employees be doing exactly?
At the end of the day they provide a useful way to store and version control a code base, but the enterprise software market is very crowded and GitHub being in vogue could be a fad that gets eclipsed by something else, just like how HipChat was THE thing for teams until Slack.
Arbitrarily saying that "Google will birth AI and they need at least 1M employees to make it happen" is something that, in my opinion, sounds similarly vapid from any kind of quantitative or forecasting standpoint.
Would like to know how the figure was estimated, and how mentioning Yahoo is relevant in the least. They have a huge stake in Alibaba, and yahoos business model is a defacto investment company, with Yahoo standalone value [yhoo market cap - (yhoo stake in baba)*(market cap of baba)] around -15B last I checked. It's not even apples and oranges, it's like grapes vs dolphins - meaningless comparison
Edit: fixed my yhoo/baba math for clarity
Apologies OP
I have several open source projects on Github, but if they try to extract a recurring revenue stream from me, I'll move the projects to another platform.
Obviously, bitbucket could decide to do the same shortly after github (as I'm fairly certain they made free repos available to attract github's customers, and free private repos to differentiate). In the case that it does, I'm not sure what I'd do.
As much as github is about just hosting code, it's also become a place where people who code/contribute to open source gather -- it doesn't seem right now like someone would just run a gitlab instance big enough to rival what github has become. Maybe the better question is when someone will step up to create that service/meeting ground.
- githubs customer isnt opensource projects, its closed source. The masses moving away is a good thing
- as dependency on github increases, the larger of a liability for bigger companies. Google needs java for android, walmart needs npm for its servers, github likely has projects that are depended upon by bigger names though we would never know
- start ups and new comers likely hear about github before anything else. It is usually at the top for SEO. As this trend continues, dependence will only increase.
- are you able to break from github? At the moment, my community is the js community. There is a lot of crappy things happening in it but I cant stop using github, npm nor node. These are essential and convienient for me. Even if I agreed politically, realistically my workflow cannot change without waisting time.
I believe Im a libertarian intelligent coding machine like the rest of you. But this doomsday scenario sounds more like prayers than accurate predictions
The nature of git lets it migrate easily. (The rest of the infrastructure, not so much.)
It has many many users that's for sure but unlike proprietary instant messengers or social networks I don't see strong network effects.
Personally I'm not interested in another clone of GitHub.com. Let's try another model entirely. Something more distributed and federated. That way lies diversity and immunity to lock-in. Let's collaborate in an open fashion using open protocols and decentralized hosting.
IPFS seems neat as an object store. Sign your commits and throw them into IPFS and you're good? Given a persistent distributed object store, it seems less obvious how to publish projects, refs, and pull requests.
If you hear about "Rails", you want to be able to find the actual consensus Rails repository master branch in a trusted way. Some semantic data on https://rails.org/ maybe? Then the namespace for repositories is just the DNS namespace... But that's just one way of getting the data; any source you can verify should be fine. So a signed message by a trusted Rails maintainer published somewhere. I was imagining a kind of NNTP reflog with PGP signatures....
I would actually be quite happy to see GH's monopoly fade a little. I personally don't like this company, but mostly I don't like the idea of one repo manager to rule them all.
Apart from those three cases, I think it's still relatively easy to switch to something like gitlab.
All the alternatives are slow, clunky and ugly.