> Old server nginx converted to reverse proxy We wrote a Python script that parsed every server {} block across all 34 Nginx site configs, backed up the originals, and replaced them with proxy configurations pointing to the new server. This meant that during DNS propagation, any request still hitting the old IP was silently forwarded. No user would see a disruption.
What was the config on the receiving side to support this? Did you whitelist the old server IP to trust the forwarding headers? Otherwise you’d get the old server IP in your app logs. Not a huge deal for an hour but if something went wrong it can get confusing.
> i presume they wont let you “manage all your AI spend in one place” for free.
Of course they will. In return they get to control who they’re routing requests to. I wouldn’t be surprised if this turns I to the LLM equivalent of “paying for order flow”.
shivers? as in it frightens you? i believe there is no way around tokens being prices like gasoline at the gas station - it changes every hour. Any other system means you are either over- or underspending.
They can route between models but you pay the standard rate for whichever model is selected (plus 5% fee). Afaik all current model providers have fixed prices per tokens which don't vary depending on, say, demand or hardware availability.
And also completely meaningless as a credit rating in the context of creditworthiness specifically means the ability to repay. And they can always print dollar bills to do so.
Now whether that $1 in 20 years will buy anything is an entirely different story.
That's not how it used to work. That's still not how it works in my country. I buy my bread from a specialized shop, my cheese from another, and my fresh produce from yet another. I know people who only buy their meat from a butcher (I do it sometimes, but not always).
I understand your point of view. But in cities of all sizes, it's easier to not have to do that. For example in NYC, a medium size city, you can easily go do your shopping in multiple places, and not at the same time.
Economic viability isn't what led to "wide availability and inventory". No, it's imperialism. It's exploitation of the Global South. It's paying slave wages through subsidiaries in West Africa to cocoa farmers while making sure those countries stay poor, for example.
We also wage economic war on our our anointed enemies like Cuba and then use the inevitable result of that economic warfare as a reason why our system is good.
> The act of typing isn't the hard part - its understanding what's going on, and why you're doing it. Using AI to generate code is only faster if you try and skip that step - which leads to an inevitable disaster
It’s more than just typing though. A simple example remembering the exact incantation of CSS classes to style something that you can easily describe in plain English.
Yes, you could look them up or maybe even memorize them. But there’s no way you can make wholesale changes to a layout faster than a machine.
It lowers the cost for experimentation. A whole series of “what if this was…” can be answered with an implementation in minutes. Not a whole afternoon on one idea that you feel a sunk cost to keep.
> It’s more than just typing though. A simple example remembering the exact incantation of CSS classes to style something that you can easily describe in plain English.
Do that enough and you won't know enough about your codebase to recognise errors in the LLM output.
It also means you're so helpless as a developer that you could never debug another person's code, because how would you recognize the errors, you haven't made them yourself.
imo a question is, do you still need to understand the codebase? What if that process changes and the language you’re reading is a natural one instead of code?
> What if that process changes and the language you’re reading is a natural one instead of code?
Okay, when that happens, then sure, you don't need to understand the codebase.
I have not seen any evidence that that is currently the case, so my observation that "Continue letting the LLM write your code for you, and soon you won't be able to spot errors in its output" is still applicable today.
When the situation changes, then we can ask if it is really that improtant to understand the code. Until that happens, you still need to understand the code.
You do have a point but as the sibling comment pointed out, the negative eventuality you are describing also has not happened for many devs.
I quite enjoy being much more of an architect than I could compared to 90% of my career so far (24 years in total). I have coded my fingers and eyes out and I spot idiocies in LLM output from trivially easy to needing an hour carefully reviewing.
So, I don't see the "soon" in your statement happening, ahem, anytime soon for me, and for many others.
What happens when your LLM of choice goes on an infinite loop failing to solve a problem?
What happens when your LLM provider goes down during an incident?
What happens when you have an incident on a distributed system so complex that no LLM can maintain a good enough understanding of the system as a whole in a single session to spot the problem?
What happens when the LLM providers stop offering loss leader subscriptions?
AFAIK everything I use has timeouts, retries, and some way of throwing up its hands and turning things back to me.
I use several providers interchangeably.
I stay away from overly complex distributed systems and use the simplest thing possible.
I plan to wait for some guys in China to train a model on traces that I can run locally, benefitting from their national “diffusion” strategy and lack of access to bleeding-edge chips.
> What if that process changes and the language you’re reading is a natural one instead of code?
Natural language is not a good way to specify computer systems. This is a lesson we seem doomed to forget again and again. It's the curse of our profession: nobody wants to learn anything if it gets in the way of the latest fad. There's already a historical problem in software engineering: the people asking for stuff use plain language, and there's a need to convert it to a formal spec, and this takes time and is error prone. But it seems we are introducing a whole new layer of lossy interpretation to the whole mess, and we're doing this happily and open eyed because fuck the lessons of software engineering.
I could see LLMs being used to check/analyze natural language requirements and help turn them into formal requirements though.
> But it seems we are introducing a whole new layer of lossy interpretation to the whole mess (...)
I recommend you get acquainted with LLMs and code assistants, because a few of your assertions are outright wrong. Take for example any of the mainstream spec-driven development frameworks. All they do is walk you through the SRS process using a set of system prompts to generate a set of documents featuring usecases, functional requirements, and refined tasks in the form of an actionable plan.
Then you feed that plan to a LLM assistant and your feature is implemented.
I seriously recommend you check it out. This process is far more structured and thought through than any feature work that your average SDE ever does.
> I recommend you get acquainted with LLMs and code assistants
I use them daily, thanks for your condescension.
> I could see LLMs being used to check/analyze natural language requirements and help turn them into formal requirements though.
Did you read this part of my comment?
> Take for example any of the mainstream spec-driven development frameworks. All they do is walk you through the SRS process using a set of system prompts to generate a set of documents featuring usecases, functional requirements, and refined tasks in the form of an actionable plan.
I'm not criticizing spec-driven development frameworks, but how battle-tested are they? Does it remove the inherent ambiguity in natural language? And do you believe this is how most people are vibe-coding, anyway?
Yes, and your comment contrasts heavily with the reality of using LLMs as code assistants, as conveyed in comments such as "a whole new layer of lossy interpretation. This is profoundly wrong, even if you use LLMs naively.
I repeat: LLM assistants have been used to walk users through software requirements specification processes that not only document exactly what usecases and functional requirements your project must adhere to, but also create tasks and implement them.
The deliverable is both a thorough documentation of all requirements considered up until that point and the actual features being delivered.
To drive the point home, even Microsoft of all companies provides this sort of framework. This isn't an arcane, obscure tool. This is as mainstream as it can be.
> I'm not criticizing spec-driven development frameworks, but how battle-tested are they?
I really recommend you get acquainted with this class of tools, because your question is in the "not even wrong" territory. Again, the purpose of these tools is to walk developers through a software requirements specification process. All these frameworks do is put together system prompts to help you write down exactly what you want to do, break it down into tasks, and then resume the regular plan+agent execution flow.
What do you think "battle tested" means in this topic? Check if writing requirements specifications is something worth pursuing?
I repeat: LLM assistants lower formal approaches to the software development lifecycle by orders of magnitude, to the point you can drive each and every single task with a formal SRS doc. This isn't theoretical, it's month's old stuff. The focus right now is to remove human intervention from the SRS process as well with the help of agents.
> Yes, and your comment contrasts heavily with the reality of using LLMs as code assistants, as conveyed in comments such as "a whole new layer of lossy interpretation. This is profoundly wrong, even if you use LLMs naively.
Most people, when told they sound condescending, try to reframe their argument in order to remove this and become more convincing.
Sadly, you chose to double down instead. Not worth pursuing.
This is not correct. CSS is the style rules for all rendering situations of that HTML, not just your single requirement that it "looks about right" in your narrow set of test cases.
Nobody writing production CSS for a serious web page can avoid rewriting it. Nobody is memorizing anything. It's deeply intertwined with the requirements as they change. You will eventually be forced to review every line of it carefully as each new test is added or when the HTML is changed. No AI is doing that level of testing or has the training data to provide those answers.
It sounds like you're better off not using a web page at all if this bothers you. This isn't a deficiency of CSS. It's the main feature. It's designed to provide tools that can cover all cases.
If you only have one rendering case, you want an image. If you want to skip the code, you can just not write code. Create a mockup of images and hand it off to your web devs.
Eh, I've written so much CSS and I hate it so much I use AI to write it now not because it's faster or better at doing so, just so I don't need to do it.
> But there’s no way you can make wholesale changes to a layout faster than a machine.
You lost me here. I can make changes very quickly once I understand both the problem and the solution I want to go with. Modifying text is quite easy. I spend very little time doing it as a developer.
> It lowers the cost for experimentation. A whole series of “what if this was…”
Anecdotal, but I've noticed while this is true it also adds the danger of knowing when to stop.
Early on I would take forever trying to get something exactly to whats in my head. Which meant I would spend too much time in one sitting then if I had previously built it by hand.
Now I try to time box with the mindset "good enough".
> For any nonviolent offenses against the United States which they may have committed or taken part in during the period from January 1
2014 through the date of this pardon (JAN 19, 2025).
> Now, Therefore, I, Gerald R. Ford, President of the United States, pursuant to the pardon power conferred upon me by Article II, Section 2, of the Constitution, have granted and by these presents do grant a full, free, and absolute pardon unto Richard Nixon for all offenses against the United States which he, Richard Nixon, has committed or may have committed or taken part in during the period from January 20, 1969 through August 9, 1974.
Not quite as long, but much more significant. (No violence exception, the criminal was the President, and they were crimes against the entire country, not some random drug/tax charges.)
The real embarrassment is how little effort there's been to limit/reform the pardon system since then.
Pardons have valid uses, but it's wild that a single person can unilaterally pardon donators, family members, former presidents, etc, without needing so much as a simple majority confirmation vote in the House or Senate.
The questionable pardons that we've seen over the last few years (and the Nixon pardon) are just the tip of iceberg in terms of how badly they could be abused.
I'd imagine it won't be long until we see a president issue a preemptive pardon to themself at the end of their term, because there's nothing in the constitution that says they can't.
Isn't that the whole point of all these pardon things? To reduce incentives to usurp power to avoid responsibility by providing less destructive for the political system ways to avoid responsibility.
Or concretely, would the Israeli wars end sooner if Netanyahu was pardoned of all crimes? Would Kim Jong Un consider giving up his position if he could be pardoned, or at least credibly believe that he could live a life in luxurious exile? I don’t know the answer to either of those questions, but I do think letting some people get away with crimes with witness immunity can make it much more difficult for criminals to organize as the optimum move is to defect before anyone else does. Which is why I think elite blackmail focuses on unforgivable deeds.
> Would Kim Jong Un consider giving up his position if he could be pardoned, or at least credibly believe that he could live a life in luxurious exile?
The kind of despot that sends assassins against people in exile is unlikely to choose it themselves.
Hmm. I feel like it isn’t over the history of the US and there was a period where US governance tended toward an ideal but the last 50 years have reverted to the norm. E.g Oliver North
So exactly when was that? Before 50 years ago, “Separate but Equal” was the law of the land as decided by the Supreme Court, laws against interracial marriage and laws against “sodomy” (homosexuality) were also upheld by the Supreme Court.
There is absolutely no point in US history that the US was “ideal”.
My still living parents grew up in the segregated south.
> It’s in the Constitution. There isn’t that much anyone can do.
We have modified the constitution before. It is not easy, sure. But, presidential pardons are being abused so thoroughly that it does warrant people making the effort to change things.
If we put 3x caught pot dealers in for life, a corrupt President can certainly rot alongside them.
The more powerful you are, the more significant the penalties for abuse of that power should be, because the damage you can do is correspondingly large.
Don’t think the three strikes law you are referring to (3x nonviolent = life in prison) exists anymore. As it was generally regarded as a cruel mistake. Likewise with hanging politicians for lying…
"Time to move on" is used only when someone in power is guilty. Happened with Nixon. Mitch McConnell basically said the same thing about Trump after J6 insurrection. And I think Garland believed the same thing when he did not move fast enough to investigate Trump. People believe in law and order when rich and powerful face the same consequences as the common man for their crimes committed and not when they are let off the hook.
I do think Garland made a massive mistake. Nixon resigned; Trump did not. Nixon largely disappeared, as most former Presidents do during their successor's term. Trump was still communicating crimes and clearly intended more.
I'm drawing a kind of fine and possibly meaningless distinction here. I think Ford made the best decision he could at the time. Garland had the benefit of hindsight: he saw the way the corruption had become far deeper than the President himself. Garland should have known better.
Fox News was founded by Roger Ailes with the explicit intent to prevent another Nixon situation. Not the "criminal President" part, mind you; the punished (Republican) President part.
Preemptive meaning they hadn't yet been convicted. Nixon was pardoned by Ford in this manner (for "all offenses against the United States" between Jan. 20, 1969—Aug. 9, 1974). Carter preemptively mass-pardoned draft dodgers, etc.
That's when I learned you can be pardoned for future crime since the expiration was end of day. There's nothing stopping a president from signing blanket future pardons with a 100 year expiry. I'm amazed there wasnt any discussion when it happened.
Look at what the Trump administration has done with the DOJ pursuing unwarranted indictments against anyone Trump doesn't like. All getting thrown out so far. And you lead with questioning why one of his constant targets would pardon his family?
The bigger question is why this isn't more outrage at the GOP attempts to find something on Biden or Clinton. They have been wasting tax dollars while Coomer "investigates" for something that he has never been able to prove.
I'd have pardoned everyone around me given that constant sustained and terrible attack.
All the while the Trump grift machine continues without so much as a blink.
No, it was right to consider the possibility that Trump would violate the norms here. Letting the President right unaddressed wrongs is the entire reason the pardon power exists.
His own current Chief of Staff has similar concerns, and grand juries seem to be taking the same position; that these are just revenge.
"Susie Wiles, the White House chief of staff, told an interviewer that she forged a “loose agreement” with Mr. Trump to stop focusing after three months on punishing antagonists, an effort that evidently did not succeed. While she insisted that Mr. Trump is not constantly thinking about retribution, she said that “when there’s an opportunity, he will go for it.”"
The only reason he’s not constantly thinking about retribution is because he spends most of the time with his brain idle, or thinking of his next grift.
> It's not a rule, it's a law passed by Congress and signed by the President in 1978. You can't just ignore it.
They’re not ignoring. They’re saying they think the law itself is unconstitutional.
From the article:
>> In a sweeping new memorandum from the Office of Legal Counsel, the DOJ claims the Presidential Records Act is unconstitutional. The department’s edict, which is already facing legal challenges, argues that a president’s records are private, rather than public, property.
The general rule is that even facially unconstitutional laws are usually enforced until a judicial ruling against them. see e.g. all the people who did prison time for municipal handgun prohibitions until District of Columbia v. Heller (2008) and McDonald v. City of Chicago (2010).
"This is unconstitional because Trump doesn't like it" is not a very strong argument. The position he's holding is called "Public Office" (not private office) for a reason.
What was the config on the receiving side to support this? Did you whitelist the old server IP to trust the forwarding headers? Otherwise you’d get the old server IP in your app logs. Not a huge deal for an hour but if something went wrong it can get confusing.
reply