I haven’t seriously used Python in over 15 years, but I assume the comparison is against using a preforking server with 1+ process per core.
The question is whether 1+ thread per core with GIL free Python perform as well as 1+ process per core with GIL.
My understanding is that this global is just a way to demonstrate that the finely grained locking in the GIL free version may make it so that preforking servers may still be more performant.
Amusingly, it's also inspired by the life of the writer's mixed race black father, who in the 1700s became the highest ranking black general in the west not to be surpassed until the American generals in the 1990s and supposedly attracted the jealousy of Napolean (See "The Black Count").
This is one of those I should have known better moments. But I have never put together Alexander Dumas the author being related (son of) to Thomas-Alexander Dumas of Haiti/France despite them both being French, mix-race and overlapping in time. Learning about Thomas-Alexander Dumas in context of Haiti and the French revolution vs the context of reading Alexander Dumas I just never thought of it.
Over 85% of people sentenced with 2 years or more await their appeal in jail, at least initially.
I personally agree with you that shouldn’t be the case, but given Sarkozy made his entire political career about being tough on crime and harsher mandatory sentencing, I’d be appalled if he received any sort of special treatment.
> I'll have to read more into this, but it says he just "conspired" to do it, whatever that means.
The court couldn't prove beyond reasonable doubt that the money was used for his campaign.
However they were able to prove beyond reasonable doubt that he knew what his subordinates were planing and that he did nothing to stop it.
In France conspiring to commit a crime is punishable, regardless of whether the crime actually happened or not.
That's a law that has been crafted by Sarkozy's own party.
> The solitary confinement part is quite harsh
The solitary part isn't a punishment, but to ensure his safety. They even went as far as to allocated another cell for the two full time police officers of his security detail...
Also the upside is that he has a cell for himself, something a lot of prisoners would love to have given the over prison occupancy in France is 137% (and up to 200% in some specific prisons).
> As the nonprofit steward of this infrastructure, Ruby Central has a fiduciary duty to safeguard the supply chain and protect the long-term stability of the ecosystem. In consultation with legal counsel and following a recent security audit, we are strengthening our governance processes, formalizing operator agreements, and tightening access to production systems.
It took less than two weeks from this statement for them to put out an incident report from them forgetting to change the password on the infrastructure they took from the previous maintainers. I can't say I'm shocked that this didn't actually result in people's confidence in their ability as steward to provide long-term stability for the ecosystem.
Ruby Central has been the entity responsible for the infrastructure hosting rubygems.org the entire time. Literally since the beginning of rubygems.org. Any hosting bills, contracts, or agreements are in the name of the Ruby Central corporation and always have been, as far as I know. Any "previous maintainers" were working as contractors or employees of Ruby Central, if they were working on infrastructure.
The (open source) source code for rubygems and bundler, the libraries that rubyists use in their apps to manage gem dependencies, are potentially another story.
But the infrastructure, to have passwords to it, for rubygems.org, has been Ruby Central since the beginning of rubygems.org without any break. I don't know why people receiving checks from Ruby Central as contractors would think they had a personal right above Ruby Central to the infrastructure that Ruby Central has been running since long before they received those checks. Them thinking they did is sketchy.
Again, the open source source code, I agree, is another matter with other considerations. It has had many maintainers and contributors over time, including periods where development was not coordinated by Ruby Central. And all the code is owned by it's authors, and licensed MIT-style. But you're talking about passwords to infrastructure...
Genuine question: how do you take something which you have already been paying for?
They removed other maintainers access to their AWS account, and one of them had allegedly taken a screenshot of the root password from a password manager and logged in a few hours later and changed the root password to lock the legal owners out. Most of the community has turned on the maintainer who did that, it was extremely childish behaviour.
> They removed other maintainers access to their AWS account, and one of them had allegedly taken a screenshot of the root password from a password manager
Inaccurate:
> Ruby Central also had not removed me as an “owner” of the Ruby Central GitHub Organization. They also had not rotated any of the credentials shared across the operational team using the RubyGems 1Password account.
> I believe Ruby Central confused themselves into thinking the “Ruby Central” 1Password account was used by operators, and they did revoke my access there. However, that 1Password account was not used by the open source team of RubyGems.org service operators. Instead, we used the “RubyGems” 1Password account, which was full of operational credentials. Ruby Central did not remove me from the “RubyGems” 1Password account, even as of today. https://andre.arko.net/2025/10/09/the-rubygems-security-inci...
Ruby Central didn't realize that they hadn't actually revoked any access to the previous maintainers (and that they didn't have the updated root AWS credentials) until two weeks later when André notified them.
They keep on using buzzwords. These Ruby central guys never maintained a single gem used by many people in their life. I have no idea what they are writing, but it feels as if AI is writing their statements. Even then it is of such a poor, repetitive quality that even AI may just accidentally write better "summaries". People lost all trust in Ruby Central - there is no way for them to win back trust here.
IMO it would be better to start from a clean slate; dissolve Ruby Central and bring back the community with a new policy, rules - but that's not going to happen. Ruby Central went the corporate way and that's it. It would just be ironic if, say in 10 years, gem.coop proves to be much more successful whereas Ruby Central still writes the same AI-generated text ("we care for the community even if everyone is now elsewhere already").
Afaik many of the people who were on board to help start gem.coop have stepped back after the recent controversies with Andre Arko, at this point I don’t think it will ever be anything more than a ruby gems mirror
Yes, the need isn't exactly the same. `load_async` use case if for known slow-ish queries, hence for which you want actual parallelization on the server.
Since that discussion on the forum, I talked more about pipelining with some other core devs, and that may happen in some form or another in the future.
The main limiting factor is that most of the big Rails contributors work with MySQL, not Postgres, and MySQL doesn't really have proper pipelining support.
re: MySQL, strictly speaking, that isn't true; proper pipelining was introduced starting in MySQL 5.7, ten years ago. However it requires using the newer "X Protocol", which isn't widely supported by third-party drivers, nor is it supported in MariaDB. So adoption has been poor.
> at which point the only claims made are fairly vague allusions to there being more to the story with even a hint at what that might be.
The goal of my post was mostly to provide "character evidence".
It's not for me to relay accusations made by others that I can't substantiate myself. Some other people did that previously and that is what caused that massive controversy.
> The grievances against Shopify seem pretty legitimate based on the only knowledge we have as outsiders.
My whole post is about how these allegations are horseshit.
Your post isn't about them being horseshit though; you say that you consider them to be, but pretty the only information in it is entirely unrelated. To me, it basically sounds like you're saying "those people are lying, and I think something different happened but I won't say what it is". Maybe I'm unusual, but that just doesn't convince me at all. I don't know how to decide whether something is believable if no one will tell me what exactly it is that I'm supposed to believe.
I did read that article before seeing your response here. I honestly don't feel like it does much to change my perception of the events that led up to it. My understanding of the claims that you are describing as horseshit are that someone who maintained gem and bundler for years got intentionally pushed out after Ruby Central was threatened to have their funding revoked from Shopify if they didn't take over those packages and remove him. I had never heard of this maintainer before, but I have used bundler and gem before, so my perspective is that even if he was a problem and there was an argument that he should be removed, having one third party threaten another into removing him by forcing the change in ownership of the tools used by the entire community is an extremely myopic way of doing it. Doing an improper job of it that gave him an opening to potentially exploit his continued access is exactly the sort of thing that explains why you shouldn't go about forcing changes like this without adequate transparency and community consensus; instead of improving the security for the community, now a bunch of people who had never heard of the parties involved with this conflict need to be worried about the collateral damage. If you think someone is dangerous, it would make sense to be prepared for this sort of thing after you escalate your conflict with them.
In the absence of any other explanation about what actually happened, the only accounting of the events paints the change in ownership as at best reckless and irresponsible. I'd love to be wrong, but without anything concrete to explain why I shouldn't trust this, I can't differentiate between the reality we're in and one where the accusations are correct and the responses to them are being made in bad faith, and the simplest explanation is that it's because they're the same.
> after Ruby Central was threatened to have their funding revoked from Shopify
So you take the original allegations at face value, even though they only rely on second hand reporting of anonymous testimonies, yet you don't want to consider my post even though the standard of proof is the same.
You didn't include the actual context of the sentence you quoted, which is that I'm summarizing my understanding of the allegations. Do you not agree with me that those are what the allegations are, or do you think my ability to summarize them somehow implies that I must agree with them?
The problem with your post is that you're asking people to believe something without telling them what it is. I'd be more than willing to consider an alternate explanation of what happened but so far no one has been willing to share one. Regardless of your reasoning for withholding it (and the reasons of the others who apparently have knowledge of it), no one is going to be convinced of anything of anything by just asking people to trust blindly. At the end of the day, people are not going to believe there's some secret truth that explains everything about how Ruby Central and Shopify were acting if l in good faith; they'll need to be told what actually happened, or they'll quite understandably trust the people who don't seem to be trying to hide something.
Funny because Airbnb was one of the examples I had in mind when I wrote this (but granted it's been a long time since I heard about the state of their infra, so might be outdated by now).
I’m in the same boat, unsure if this is outdated. I left Airbnb five years ago and haven’t followed up to ask anyone what the current state of the tech is.
The question is whether 1+ thread per core with GIL free Python perform as well as 1+ process per core with GIL.
My understanding is that this global is just a way to demonstrate that the finely grained locking in the GIL free version may make it so that preforking servers may still be more performant.
reply