>And YouTube also had those stupid challenges with everyone doing the same stupid shit before TikTok even existed.
And before the transistor, we had flagpole sitters[0] and dance marathons[1] and dozens of other memes, just in the 20th century.
This kind of thing is nothing new, and has been going on for as long we've been us. Now this is accessible to a larger and more varied audience, not just those who are nearby.
These large consultancies staff at a lot of places that aren’t big tech. While they certainly have some good talent the overwhelming reputation with body shops is that they place some pretty mediocre talent.
I like guns and cars, but not sports. How exactly is it performative? Both are engineering marvels and fascinating to watch videos about, and they also happen to be a load of fun.
I watch hours of videos on both with nobody else around and don't really talk about those topics with others much. So in the spirit of HN, I'm actually curious to know what about those interests is performative?
I mean the salaries are pretty terrible, and you most likely see multiple layers of contracting happening, each layer taking a cut. The final hourly rate for these engineers is abysmal, especially compared to what is probably be billed to the final customer.
I disagree because you dont just want "twitter for MMA" you want the whole network. So you want to be a good citizen and have instances federate with you. But most wont because of the nature of the content. So users would prefer platforms where they can follow all content from a single account.
> Suspending customers for using a fairly expensive subscription plan -- especially forfeiting an annual prepayment for a day or two of coloring outside the lines
they're being suspended for using a private api outside of the app for which the api was intended. If you make a clone of the hbo app, so that you can watch hbo shows without ads by logging in with your discounted ads-included membership, your account will also be suspended.
Granted. I guess, unless it's applied very aggressively, assessing the existence of rate limiting may require some sort of automation (and probably some heuristics – how much data points do you actually need? do you have to retrieve any data at all, while looking for a single signal? The article doesn't tell.) Same goes for lockout.
On the other hand, as mentioned already, all that's required is really looking for a return code and not for any data. Is accessing an API endpoint the same as retrieving data? Is there proof or evidence of intent of the latter? I guess, there remains much to be defined. Especially, if it's not so much about protecting reputation than it is about protecting data and ensuring trust, and the intent is to protect and secure this in the first place.
But the question is - why is the -p flag fine? It hits the same endpoints with the same OAuth token and same quotas.
Comments section here and on related news from Anthropic seems to be centered around the idea that the reason for these bans is that it burns tokens quickly, while their plans are subsidized. What changes with the -p flag? You're just using cli instead of HTTP.
Are the metrics from their cli more valuable than the treasure trove of prompt data that passes through to them either way that justifies this PR?
Cute that you think that's how it works. I guess you're also thinking everyone that voted for the current administration agrees with them on everything they do and voted them in exactly for that. I am at least glad you didn't say if you don't like how it works, move elsewhere.
I tend to err on the side of discretion as well. It's more professional.
Though over the years, I've learned to calibrate that discretion proportional to how much of a good-faith effort the counterparty involved seems to be making. If they clearly don't give a shit that they're incompetent, they can expect my megaphone to blare.
Interesting, though a lot of the UI seems broken. For my state I see some notice dates in the future (it's not explained why, if this is when the filing will be executed or if it's an incorrect filing date, as the column is just "Notice Date")
Some of the entries pull up a page that says "Failed to load company data: No company name provided in URL" from the state specific view (e.g, any link on https://warnfirehose.com/data/layoffs/california ). Has a vibe-coded feel to it.
I saw a lot of "Purchase dataset for city details" in places which was annoying. Wondering how much processing is being done on the base dataset to justify the pricing. Could you explain a bit on the normalization/cleaning process?
So you are saying that if at a certain point in parsing the only expected terms are 'a', 'b' and 'c', one should not put the corresponding parsed entry in a `char` (after checking it is either of these aka validating), and instead it should be put in some kind of enum type (parsed via `Alternative f`). Right?
Or browsing shelves in a bookshop. I've noticed I forget what I was doing ("prospective memory impairment") while looking for a good book. Also sometimes I annoy myself because I want to quit but I can't because I haven't found anything good yet. Whoops, where did the time go? So, ban bookshops.
> I have been using Emacs non-stop for >20 years and my .emacs is just 20 LOC. It's been shrinking, not growing.
Me too. I mean I'm using Emacs too, and it is 20+ years. I hate it deeply, and I cannot stop hating it because I cannot get away from it. I regret deeply choosing emacs 20+ years ago and spending 20 years to wrap my habits around it.
BTW my .emacs is still growing. I don't know how you manage to have 20 LoC of .emacs, I have a directory .emacs.d and a couple of dozen of files there. They are not large, some of them can be as small as 1 line. The last one I've wrote was dealing with indent of lua code. lua-ts-mode have some relatively simple rules that mostly work, but I was not happy with the result, there are some quirks that just are very inconvenient, and in some cases lua-ts-mode just fail to indent properly. So I fixed them to my taste. This one file is longer than 20 LoC.
Though, I should note, that LLMs make this much simpler. It is very simple to reverse-engineer what there is, and if you can explain the idea how to change the code, LLM can write all the elisp needed. It doesn't work out of the box, of course, and needs to be debugged, still LLM can save an hour or two.
> My goal is to bring it down towards 0 LOC.
You cannot. If you use lua you just cannot, because lua-mode uses indent of 3 spaces. Not 2, not 4, but three. So any lua sources you can find on github and try to edit will not be indented like lua-mode does. I cannot imagine what was going on the mind of the person who had chosen this value. The only possible explanation I have is something like "I want to be not like the others", but it doesn't seem right.
So you need at least to change lua-indent-offset (or lua-ts-indent-offset if you use treesitter), and it will be more than 0 LoC.
Tesla ‘Robotaxi’ adds 5 more crashes in Austin in a month — 4x worse than humans
https://electrek.co/2026/02/17/tesla-robotaxi-adds-5-more-cr...