The keen observer will of course know that there's no such thing as "federal immunity"
The scary thing is that there is.. you should look up "sovereign immunity". The government has complete immunity, except where and how the law permits it to be held accountable. And while we have a constitution, defending those rights through the courts requires legislation to permit it. For the most part, federal law permits lawsuits against states that violate the constitution, but have permitted far less accountability for federal actions that violate the constitution.
For example, Section 1983 of the Civil Rights Act only permits individuals to sue state and local governments for rights violations. It can't be used to sue the federal government.
There's many court cases, dating back decades, tossing out cases against the federal government for rights violations. Look how SCOTUS has limited the precedent set by Bivens over the years, basically neutering it entirely.
Hold on, I'm not a sysadmin guy. Are you folks saying the server should not know what part of the world its in, that basically it should think it's in Greenwitch?
I would have thought you configure the server to know where it is have it clock set correctly for the local time zone, and the software running on the server should operate on UTC.
From a logging perspective, there is a time when an event happens. The timestamp for that should be absolute. Then there's the interaction with the viewer of the event, the person looking at the log, and where he is. If the timestamp is absolute, the event can be translated to the viewer at his local time. If the event happens in a a different TZ, for example a sysadmin sitting in PST looking at a box at EST, it's easier to translate the sysadmin TZ env, and any other sysadmin's TZ anywhere in the world, than to fiddle with the timestamp of the original event. It's a minor irritation if you run your server in UTC, and you had to add or subtract the offset, eg. if you want your cron to run at 6PM EDT, you have to write the cron for 0 22 * * *. You also had to do this mental arithmetic when you look at your local system logs, activities at 22:00:00 seem suspicious, but are they really? Avoid the headaches and set all your systems to UTC, and throw the logs into a tool that does the time translation for you.
The server does not "know" anything about the time, that is, it's really about the sysadmin knowing what happened and when.
1) Most software gets its timestamps from the system clock
2) If you have a mismatch between the system time and the application time, then you just have log timestamps that don't match up; it's a nightmare - even more so around DST/ST transitions
you've got it backwards - the server clock should be in UTC, and if an individual piece of software needs to know the location, that should be provided to it separately.
for example, I've got a server in my garage that runs Home Assistant. the overall server timezone is set to UTC, but I've configured Home Assistant with my "real" timezone so that I can define automation rules based on my local time.
Home Assistant also knows my GPS coordinates so that it can fetch weather, fire automation rules based on sunrise/sunset, etc. that wouldn't be possible with only the timezone.
Windows assumes computer clocks are local time. It can be configured to assume UTC. Other operating systems assume computer clocks are UTC. Many log tools are not time zone aware.
that's the difference between "aware" and "naive" timestamps. Python has a section explaining it in their docs (though the concept applies to any language):
Yes, that's exactly what I'm saying :). In fact, I've run servers where I didn't even physically know where it was located. It wouldn't have been hard to find out given some digging with traceroute, but it didn't matter. It was something I could SSH into and do everything I needed to without caring where it was.
Everyone else down-thread has clarified the why of it. Keep all of your globally distributed assets all running on a common clock (UTC) so that you can readily correlate things that have happened between them (and the rest of the world) without having to do a bunch of timezone math all the time.
Where did you get that from? I am currently using GPT-5 via OpenRouter and never added an OpenAI key to my account there. Same for any previous OpenAI model. BYOK is an option, not a necessity.
> {"id":"openai/gpt-5-chat","canonical_slug":"openai/gpt-5-chat-2025-08-07","hugging_face_id":"","name":"OpenAI: GPT-5 Chat","created":1754587837,"description":"GPT-5 Chat is designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.","context_length":400000,"architecture":{"modality":"text+image->text","input_modalities":["file","image","text"],"output_modalities":["text"],"tokenizer":"GPT","instruct_type":null},"pricing":{"prompt":"0.00000125","completion":"0.00001","request":"0","image":"0","audio":"0","web_search":"0","internal_reasoning":"0","input_cache_read":"0.000000125"},"top_provider":{"context_length":400000,"max_completion_tokens":128000,"is_moderated":true},"per_request_limits":null,"supported_parameters":["max_tokens","response_format","seed","structured_outputs"]},
If you look at the JSON you linked, it does not enforce BYOK for openai/gpt-5-chat, nor for openai/gpt-5-mini or openai/gpt-5-nano.
the media but also the llm providers actively encourage this to fuel their meteoric valuations that are based on the eminent value that would be provided by AGI replacing human labor.
the entire thing — from the phrasing of errors as “hallucinations”, to the demand for safety regulations, to assigning intention to llm outputs — is all a giant show to drive the hype cycle. and the media is an integral part of that, working together with openai et al.
pro is the $20/mo plan that they recently started allowing access to claude code.. but i’ve heard users hit the rate limit with a few queries.. so imo that sounds about right. the chat interface has its own limits separate from claude code.
Has to be wrong. I'm on that subscription as I wanted to reinforce my opinion that it's still shit for devs that actually have experience, like it was a few months ago.
While my plan didn't pan out, cuz it was way too effective, I can confidently say that I'm going through 3-6k tokens per prompt on average, and usually get around 3 hours of usage before I'm hitting the rate limit.
The limit is probably closer to 300k then <10k
Also the chat interface doesn't have a separate limit, once you hit it via Claude code, you cannot use the website either anymore.
Maybe it's a 7k limit per prompt? Dunno if I exceeded that before
I found I can hit the limit very quickly if I have it scan large quantities of code for analysis. If I try to be more surgical, and give it terse but accurate documentation and instructions, the budget lasts longer.
7k is literally nothing, even for a trial 7k of token is basically 1-2 files written that doesn't seem right if it is then i dont see why anyone would pay for that and not the 250 prompts/month from augment or one of the others
I hadn't even heard of augment, but Claude Code's UX is _mostly_ very nice (despite the problematic UX this particular project attempts to solve). So perhaps Claude Code has a better UI/UX?
I tried it with Roo Code (with 3.7 Sonnet, not Code). For agentic use you will probably hit the limit from your first prompt/assignment if it does some browser tool use.
if you work on a team most code you see isn’t yours.. ai code review is really no different than reviewing a pr… except you can edit the output easier and maybe get the author to fix it immediately
Reviewing code is harder than writing code. I know staff engineers that can’t review code. I don’t know where this confidence that you’ll be able to catch all the AI mistakes comes from.
I was about to say exactly this—it's not really that different from managing a bunch of junior programmers. You outline, they implement, and then you need to review certain things carefully to make sure they didn't do crazy things.
But yes, these juniors take minutes versus days or weeks to turn stuff around.
> if you work on a team most code you see isn’t yours.. ai code review is really no different than reviewing a pr… except you can edit the output easier and maybe get the author to fix it immediately
And you can't ask "why" about a decision you don't understand (or at least, not with the expectation that the answer holds any particular causal relationship with the actual reason)... so it's like reviewing a PR with no trust possible, no opportunity to learn or to teach, and no possibility for insight that will lead to a better code base in the future. So, the exact opposite of reviewing a PR.
Are you using the same tools as everyone else here? You absolutely can ask "why" and it does a better job of explaining with the appropriate context than most developers I know. If you realize it's using a design pattern that doesn't fit, add it to your rules file.
You can ask it "why", and it gives a probable English string that could reasonably explain why, had a developer written that code, they made certain choices; but there's no causal link between that and the actual code generation process that was previously used, is there? As a corollary, if Model A generates code, Model A is no better able to explain it than Model B.
I think that's right, and not a problem in practice. It's like asking a human why: "because it avoids an allocation" is a more useful response than "because Bob told me I should", even if the latter is the actual cause.
> I think that's right, and not a problem in practice. It's like asking a human why: "because it avoids an allocation" is a more useful response than "because Bob told me I should", even if the latter is the actual cause.
Maybe this is the source of the confusion between us? If I see someone writing overly convoluted code to avoid an allocation, and I ask why, I will take different actions based on those two answers! If I get the answer "because it avoids an allocation," then my role as a reviewer is to educate the code author about the trade-off space, make sure that the trade-offs they're choosing are aligned with the team's value assessments, and help them make more-aligned choices in the future. If I get the answer "because Bob told me I should," then I need to both address the command chain issues here, and educate /Bob/. An answer is "useful" in that it allows me to take the correct action to get the PR to the point that it can be submitted, and prevents me from having to make the same repeated effort on future PRs... and truth actually /matters/ for that.
Similarly, if an LLM gives an answer about "why" it made a decision that I don't want in my code base that has no causal link to the actual process of generating the code, it doesn't give me anything to work with to prevent it happening next time. I can spend as much effort as I want explaining (and adding to future prompts) the amount of code complexity we're willing to trade off to avoid an allocation in different cases (on the main event loop, etc)... but if that's not part of what fed in to actually making that trade-off, it's a waste of my time, no?
Right. I don't treat the LLM like a colleague at all, it's just a text generator, so I partially agree with your earlier statement:
> it's like reviewing a PR with no trust possible, no opportunity to learn or to teach, and no possibility for insight that will lead to a better code base in the future
The first part is 100% true. There is no trust. I treat any LLM code as toxic waste and its explanations as lies until proven otherwise.
The second part I disagree somewhat. I've learned plenty of things from AI output and analysis. You can't teach it to analyze allocations or code complexity, but you can feed it guidelines or samples of code in a certain style and that can be quite effective at nudging it towards similar output. Sometimes that doesn't work, and that's fine, it can still be a big time saver to have the LLM output as a starting point and tweak it (manually, or by giving the agent additional instructions).
Oh, it can infer quite a bit. I've seen many times in reasoning traces "The user is frustrated, understandably, and I should explain what I have done" after an exasperated "why???"
>And you can't ask "why" about a decision you don't understand (or at least, not with the expectation that the answer holds any particular causal relationship with the actual reason).
To be fair, humans are also very capable of post-hoc rationalization (particularly when they're in a hurry to churn out working code).
same. i ran a few tests ($100 worth of api calls) with opus 4 and didn’t see any difference compared to sonnet 4 other than the price.
also no idea why he thinks roo is handicapped when claude code nerfs the thinking output and requires typing “think”/think hard/think harder/ultrathink just to expand the max thinking tokens.. which on ultrathink only sets it at 32k… when the max in roo is 51200 and it’s just a setting.
I think you misread my comment. I wasn't asking for help. I get consistent good output from Sonnet 4 using RooCode, without needing Gemini for planning.
Edit: I think I know where our miscommunication is happening...
The "think"/"ultrathink" series of magic words are a claudecode specific feature used to control the max thinking tokens in the request. For example, in claude code, saying "ultrathink" sets the max thinking tokens to 32k.
On other clients these keywords do nothing. In Roo, max thinking tokens is a setting. You can just set it to 32k, and then that's the same as saying "ultrathink" in every prompt in claudecode. But in Roo, I can also setup different settings profiles to use for each mode (with different max thinking token settings), configure the mode prompt, system prompt, etc. No magic keywords needed.. and you have full control over the request.
The scary thing is that there is.. you should look up "sovereign immunity". The government has complete immunity, except where and how the law permits it to be held accountable. And while we have a constitution, defending those rights through the courts requires legislation to permit it. For the most part, federal law permits lawsuits against states that violate the constitution, but have permitted far less accountability for federal actions that violate the constitution.
For example, Section 1983 of the Civil Rights Act only permits individuals to sue state and local governments for rights violations. It can't be used to sue the federal government.
There's many court cases, dating back decades, tossing out cases against the federal government for rights violations. Look how SCOTUS has limited the precedent set by Bivens over the years, basically neutering it entirely.