If you're unhappy that bad people might use your software in unexpected ways, open source licenses were never appropriate for you in the first place.
Anyone can use your software! Some of them are very likely bad people who will misuse it to do bad things, but you don't have any control over it. Giving up control is how it works. It's how it's always worked, but often people don't understand the consequences.
People do not have perfect foresight, and the ways open source software is used has significantly shifted in recent years. As a result, people reevaluating whether or not they want to participate.
>Giving up control is how it works. It's how it's always worked,
no, it hasn't. Open source software, like any open and cooperative culture, existed on a bedrock, what we used to call norms when we still had some in our societies and people acted not always but at least most of the time in good faith. Hacker culture (word's in the name of this website) which underpinned so much of it, had many unwritten rules that people respected even in companies when there were still enough people in charge who shared at least some of the values.
Now it isn't just an exception but the rule that people will use what you write in the most abhorrent, greedy and stupid ways and it does look like the only way out is some Neal Stephenson Anathem-esque digital version of a monastery.
Open source software is published to the world and used far beyond any single community where certain norms might apply.
If you care about what people do with your code, you should put it in the license. To the extent that unwritten norms exist, it's unfair to expect strangers in different parts of the world to know what they are, and it's likely unenforceable.
This recently came up for the GPLv2 license, where Linus Torvalds and the Software Freedom Conservancy disagree about how it should be interpreted, and there's apparently a judge that agrees with Linus:
Inside open source communities maybe. In the corporate world? Absolutely not. Ever. They will take your open source code and do what they want with it, always have.
This varies. The lawyers for risk-adverse companies will make sure they follow the licenses. There are auditing tools to make sure you're not pulling in code you shouldn't. An example is Google's go-licenses command [1].
But you can be sure that even the risk-adverse companies are going to go by what the license says, rather than "community norms."
It’s a fair point that ai training makes enforcing licences more difficult than other situations. My point is that licence issues like this this aren’t really a technology issue it’s a company greed/legal issue because it’s always been the case.
People training LLM's on source code is sort of like using newspaper for wrapping fish. It's not the expected use, but people are still using it for something.
As they say, "reduce, reuse, recycle." Your words are getting composted.
When learning to play a musical instrument and practicing a new song, some common advice is to play it as slowly as you can stand, to learn the motions well.
Right, but then you need to learn the right motions well, motions that will make sense at the final speed. I suppose it's one of those things that are made easier by having a teacher.
> the main known applications of quantum computers remain (1) the simulation of quantum physics and chemistry themselves, (2) breaking a lot of currently deployed cryptography, and (3) eventually, achieving some modest benefits for optimization, machine learning, and other areas (but it will probably be a while before those modest benefits win out in practice).
...
> 2025 was clearly a year that met or exceeded my expectations on hardware, with multiple platforms now boasting >99.9% fidelity two-qubit gates, at or above the theoretical threshold for fault-tolerance. This year updated me in favor of taking more seriously the aggressive pronouncements—the “roadmaps”—of Google, Quantinuum, QuEra, PsiQuantum, and other companies about where they could be in 2028 or 2029.
...
> at some point, the people doing detailed estimates of how many physical qubits and gates it’ll take to break actually deployed cryptosystems using Shor’s algorithm are going to stop publishing those estimates, if for no other reason than the risk of giving too much information to adversaries. Indeed, for all we know, that point may have been passed already. This is the clearest warning that I can offer in public right now about the urgency of migrating to post-quantum cryptosystems, a process that I’m grateful is already underway.
Especially when you don't approach it that way; they made an HTML game in desktop Chrome then worked on getting it to work in various mobile browsers, including quite old ones.
The article gets to the right conclusion, develop mobile first. I'd put it more generally as develop for the most constrained devices you intend to support. If they had done so, it would have saved a lot of time (some of their expectations about APIs still would not have been met).
The point of all this is that PG&E sometimes needs to raise money. They do get money from customers paying their bills, but they often to raise money many years sooner than that. Selling bonds and selling stock are two ways of doing it, with different tradeoffs for investors. Stock might sometimes pay a higher dividend, but dividends aren’t guaranteed.
After the last bankruptcy, PG&E suspended dividends for six years, from 2018 to 2023.
No matter how good or bad management is, they’re still going to need to raise money. Bad management means more mistakes and then more money needs to be raised.
Bad management might get fired, but they’re not going to pay for their mistakes. The money needs to come from somewhere else.
Why should shareholders be taking on the risk of a city's power grid failing? Does packaging that risk as an investment opportunity somehow produce incentives to improve grid reliability and guide resources to be used efficiently?
PG&E emerged from their most recent bankruptcy with more debt than they entered with! This was largely because of the wildfire liabilities - $30 billion. In more direct terms, the lucky Californians are paying the unlucky Californians, and it was financed by bonds which used PG&E's existing assets as collateral (which for some reason weren't already collateralized.)
The best case for Californians overall would have been PG&E's debt and equity to go as close to 0 as possible, and all that extra debt have been used to actually upgrade the aging electricity infrastructure. Instead you are paying interest on past fire damage claims.
In 2018 PG&E had about $18b of long term debt, they now have just under $59b. Their outstanding shares also quadrupled. The bankruptcy didn't wipe out the equity, but investors got f'd hard if they thought they were acting conservatively. Would you accept a 1.25%ish dividend with the prospects of the stock going to 0 higher than it doubling in the next 10 years?
For all of the whiners about how utility investors shouldn't make any money, and possibly earn below their cost of capital, -- I spent some time looking at the utility industry over the last few years (including PG&E.) These are basically money pits which more money goes in than comes out over decades.
The other layer here is if the billions of dollars being borrowed are to build new infrastructure results in billions more in future liabilities to maintain everything. The first layer looked so bad I didn't go any further.
The petty dividend payouts utilities make just keep the equity investors from examining what they really own. Higher equity valuations let utilities borrow money cheaper than they really should be able to.
Functionally the whole thing looks like a ponzi scheme that perhaps could only happen with the 40 year run of ever shrinking interest rates. If the bond bull market is over then this utility ponzi scheme is going to blow too.
Bottom line, if investors were paying attention, your utility bills would be a lot higher. If utilities ever have a big problem getting Wall Street to keep funding their debt ponzi, they will be.
The alternative is the state owns the utility. Given how ugly the math is for utilities right now, I doubt it would be cheaper.
On the other hand, if the US is going 100% EV (AI datacenters or not), then there trillions of federal dollars are inbound and maybe utilities will be ok. One thing is for certain, the utilities, their investors (debt & equity) their customers, and the US states don't have the money to pay for all that has to be built.
Except that the state of California ended up on the hook for the first bankruptcy. The shareholders were the only ones who came out fine. The customers and the state got stuck with the bill.
Exactly what risk did they take on? A few missed dividends, and two years for the stock price to recover?
As for the second bankruptcy, the main result of that was that their customers ended up paying the bill for other customers whose houses were destroyed. But you are partially correct, the shareholders did take a haircut of a few percentage points from stock dilution. I wouldn't be too upset for them, the stock's now double what it was before the bankruptcy.[1]
California's cities wanted them to take a haircut of 100 percentage points, but that clearly didn't happen.
[1] For some reason, the wise stewardship of the shareholders and the board did nothing to mitigate the crisis that caused the company to get sued for 50 billion dollars. They were too busy squeezing dividends out of it to worry about liabilities. [2]
[2] And why should they? They aren't personally liable.
Just had to look this one up. PG&E's first bankruptcy was April 6, 2001. Based on the stock price decline prior to that, it looks like their shareholders thought everything was ok in November of 2000 and the stock was $27 (it bottomed out at $8.97 in April of 2001.) As of today, the stock is worth $15.97.
If we go back 30 years to 1995 -- and you invested $10,000 in PG&E and $10,000 in the S&P500, and reinvested the dividends -- today the PG&E investment would be worth $11,708. The S&P investment would be worth $201,420.
To put it in simpler terms, the PG&E investors look like gullible fools.
1. The stock recovered within 2 years and then shot to the moon.
2. You're not counting all the dividends they've siphoned out.
3. The reason it's at $16 today is because the company destroyed its own value... By prioritizing dividends over maintenance. Which killed a lot of people, destroyed a ton of property, with the damages exceeding the value of the firm. Yet, instead of being zeroed out, the shareholders are still there, still collecting dividends, and in a few years of guaranteed 10% margins, I'm sure the stock will recover.
They came out squeaky clean in 2001, keeping all their shares. (Major shout-out to the State of California for bailing them out, and shouldering all the subsequent liabilities from that adventure.)
The second one diluted them by 22%, by creating and giving new stock to the people who won lawsuits against the firm. The stock's up by more than 22% since the impact of the fires on the company was realized...
If they were zeroed out after either bankruptcy, I wouldn't be kvetching, but here we are...
> As for the second bankruptcy, the main result of that was that their customers ended up paying the bill for other customers whose houses were destroyed.
There's half of the major problems. If I walked around covered in gasoline every day and eventually walked past someone smoking, not a lot of people would blame the smoker for me getting engulfed in flames.
Yet build a wood house in a forest maintained for thousands of years by American Indians with fire, require universal electricity supply, and suddenly it's not the homeowner's fault at all. Everyone else should bail them out over and over again.
I wonder what it would look if you redid the benchmarks, testing against models that have reasoning effort set to various values. Maybe structured output is only worse if the model isn't allowed to do reasoning first?
For me, that’s kind of the point. It’s similar to how the characters in a novel don’t really exist, and yet you can’t really discuss what happens in a novel without pretending that they do. It doesn’t really make sense to treat the author’s motivations and each character’s motivations as the same.
Similarly, we’re all talking to ghosts now, which aren’t real, and yet there is something there that we can talk about. There are obvious behavioral differences depending on what persona the LLM is generating text for.
I also like the hint of danger in “talking to ghosts.” It’s difficult to see how a rational adult could be in any danger from just talking, but I believe the news reports that some people who get too deep into it get “possessed.”
I'm retired now, but I spent many hours writing and debugging code during my career. I believed that implementing features was what I was being paid to do. I was proud of fixing difficult bugs.
A shift to not writing code (which is apparently sometimes possible now) and managing AI agents instead is a pretty major industry change.
Anything you do with AI is improved if you're able to traverse the stack. There's no situation where knowing how to code won't put you above peers who don't.
It's like how every job requires math if you make it far enough.
Anyone can use your software! Some of them are very likely bad people who will misuse it to do bad things, but you don't have any control over it. Giving up control is how it works. It's how it's always worked, but often people don't understand the consequences.
reply