GitHub is in a tough spot. From what I've heard they've been ordered to move everything to Azure from their long standing dataceners. That is bound to cause issues. Then on top of that they are using AI coders for infra changes (supposedly) which will also add issues.
And then on top of all that, their traffic is probably skyrocketing like mad because of everyone else using AI coders. Look at popular projects -- a few minutes after an issue is filed they have sometimes 10+ patches submitted. All generating PRs and forks and all the things.
That can't be easy on their servers.
I do not envy their reliability team (but having been through this myself, if you're reading this GitHub team, feel free to reach out!).
> Look at popular projects -- a few minutes after an issue is filed they have sometimes 10+ patches submitted. All generating PRs and forks and all the things.
I think this is a really important point that is getting overlooked in most conversations about GitHub's reliability lately.
GitHub was not designed or architected for a world where millions of AI coding agents can trivially generate huge volumes of commits and PRs. This alone is such a huge spike and change in user behavior that it wouldn't be unreasonable to expect even a very well-architected site to struggle with reliability. For GitHub, N 9s of availability pre-AI simply does not mean the same thing as N 9s of availability post-AI. Those are two completely different levels of difficulty, even when N is the same.
yeah this is indeed a good insight. Back in the days, who would expect so many bots to "review" code and leave overly verbose comments under every PR in a popular repo?
Some may want to come in here leaving snarky comments about how they shouldn't vote for an administration that doesn't believe in climate change. But I will give a concrete example:
This administration fired thousands of Forrest Service and BLM employees at the start of the administration last year. Those workers were the ones that were responsible for the maintenance of these lands and for the fire lookout programs.
Maybe they couldn't have prevented this fire, but it's pretty clear these fires are much worse today because of those firings last year.
I’m not sure if these fires are correlated with the staff reductions in BLM. In the interview with a resident she doesn’t mention anything about that, just that last year was very wet which provided a lot of fuel and this hot dry spring has turned it into a tinder box.
BLM provides fire lookout services that were severely cut. They also perform prescribed burns, which they no longer have staff and budget for, which would have reduced the dry grass.
I seriously doubt ranchers were politically against OPM being used for fire prevention funding. Agriculture industry is well known to be highly socialist politically when it comes to agriculture subsidies.
Of course they weren't but look at how Nebraska voted in the last election. Other than the cities, it was mostly GOP, the party that explicitly said they would cut funding for those services if elected (it was in Project 2025).
"Other than the cities." Omaha recently hit the 1M mark, which includes that near part of Iowa. Lincoln has around 300k. So causally for a state total of 2M, around 700k dispersed. Compare this with the fires and controlled burn issues in Colorado, or the 'red' part of Colorado, compare population. But we like smoking pot in Denver/Boulder, so we will pretend there aren't more Trump voters, regular issues with fires (under Obama anyone) etc, N is for narrative. I didn't vote for Trump and I have experience seeing many perspectives in both of these states. 700k people and 2 electoral votes didn't throw the election. "Trump won statewide in Nebraska, as well as in the 1st and 3rd Congressional Districts, picking up four electoral votes. Harris carried the Omaha-based 2nd Congressional District and its electoral vote." I suppose Boulder voted for its fentanyl problems as well.
People are doing this now. It's basically what skills.sh and its ilk are for -- to teach AIs how to do new things.
For example, my company makes a new framework, and we have a skill we can point an agent at. Using that skill, it can one-shot fairly complicated code using our framework.
The skill itself is pretty much just the documentation and some code examples.
Isn't the "skill" just stuff that gets put into the context? Usually with a level of indirection like "look at this file in this situation"?
How long can you keep adding novel things into the start of every session's context and get good performance, before it loses track of which parts of that context are relevant to what tasks?
IMO for working on large codebases sticking to "what the out of the box training does" is going to scale better for larger amounts of business logic than creating ever-more not-in-model-training context that has to be bootstrapped on every task. Every "here's an example to think about" is taking away from space that could be used by "here is the specific code I want modified."
The sort of framework you mention in a different reply - "No, it was created by our team of engineers over the last three years based on years of previous PhD research." - is likely a bit special, if you gain a lot of expressibility for the up-front cost, but this is very much not the common situation for in-house framework development, and could likely get even more rare over time with current trends.
> Isn't the "skill" just stuff that gets put into the context? Usually with a level of indirection like "look at this file in this situation"?
Today, yes. I assume in the future it will be integrated differently, maybe we'll have JIT fine-tuning. This is where the innovation for the foundation model providers will come in -- figuring out how to quickly add new knowledge to the model.
Or maybe we'll have lots of small fine tuned models. But the point is, we have ways today to "teach" models about new things. Those ways will get better. Just like we have ways to teach humans new things, and we get better at that too.
A human seeing a new programming language still has to apply previous knowledge of other programming languages to the problem before they can really understand it. We're making LLMs do the same thing.
Yes and no. How does a human learn a new language? They use their previous experience and the documentation to learn it. Oftentimes they way someone learns a new language is they take something in an old language and rewrite it.
LLMs are really good at doing that. Arguably better than humans at RTFM and then applying what's there.
The norcal/socal divide caused by the river is funny to me. I grew up in LA, then moved to the Bay Area for college. In LA we never really talked about where our water comes from. But we were always 'in a drought' and always taught to conserve water.
My wife grew up in the Bay Area, and was told the same.
But her family is from Sacramento. Up until about 15 years ago, everyone in Sacramento paid the same for water (based on square footage of your home). There were no water meters. So they didn't conserve. They ran the sprinklers in 100 degree heat for hours, they washed sidewalks with water instead sweeping, and all the other things.
But when the meters came, her Uncle blamed SoCal for "stealing his water". He complained every month when the bill came about how he has to pay more now because of SoCal.
Owens valley, where LA "steals" water from, is on the eastern side of the Sierras.
NorCal, including Sacramento, is on the western side of the Sierras.
So unless they planned on pumping the water over/under the mountain range that surrounds it in every direction except for towards LA, that water was never available for any NorCal city to use.
The California Aqueduct delivers water from the western Sierras through the Central Valley and to Los Angeles. This is likely what NorCal refers to when they say SoCal is 'stealing our water'.
SoCal does, yes; about half the water going through the SWP from NorCal, or ~75% if you include Bakersfield/Kern as part of SoCal (though most would consider it Central Valley).
But SoCal isn't only LA. LA itself gets a bit less than half of their water from MWP, which manages the water from the SWP and the Colorado. About the same amount it gets from the the eastern Sierras. These are supposed to drop to ~10% of LA's water supply as recapture/recycling projects complete.
Or computed the other way around, LA only has rights to ~20% of the water managed by MWD. Of course water supply, distribution, and rights are all blended and traded around all the time, but generally speaking it's not "LA" using up that water from NorCal, the consumption is significantly more from the cities and farms that came after.
tl;dr: Urban water use is tiny. In NorCal, the vast majority of the water flows unimpeded to the sea. In the Central Valley, most water is used for agriculture. Agricultural water use in any one of the 3 major basins in the Central Valley is more than all urban areas in California combined. Unsurprisingly, urban use is the primary one in the SF and LA areas, but the absolute totals are very small compared to total CA water supplies.
Not just agriculture but highly water intensive agriculture like almonds. Also I read that a lot of laws about water in some US states contain so many grandfathered clauses that few people 'control' a lot of water use, not sure how much.
Owens valley is basically dried up from the water that LA takes. It's interesting as you drive in the towns in the Valley and you see all the LA Department of Water and Power offices over 200 miles from Los Angeles. The courts had to force the LA DWP to quit taking too much water from the streams that feed Mono Lake as it was in danger of drying out.
Yep, Owens valley is basically an environmental disaster created by LA. So in the grand scheme of things, buying water from NorCal is better than stealing from the Owens valley through antiquated water rights.
But really, California (and really the entire Western US) needs a water rights governance overhaul. Right now the focus is all on urban water use, which is practically negligible compared to the agricultural water rights usage.
I disagree I drive through there every winter and the lakes are very large. The ecology of the valley is dry but nowhere near as dry as say the mojave just to the south.
they are saying that LA takes water from sources which would otherwise drain into the sacramento and san joaquin river delta. The video from this post mentions the California State Water Project which takes water from the Feather River (Oroville Dam) and distributes it along the Western edge of the central valley South to Bakersfield where it is then pumped over the mountains both towards Los Angeles and further East to San Bernardino and Riverside. It provides way more water to SoCal than the two Los Angeles-specific aqueducts from the Owens Valley on the Eastern side of the Sierras.
The CWP is designed for robustness, on top of delivery. Those aqueducts you're pointing to that feed into the municipal portion of the Inland Empire are frequently empty because the IE has it's own (mostly) self-sufficient water store (the San Bernardino Mountains). They exist in case there is a point in which those regions need water fed in. You can literally just drive down to them at pretty much any time through the year and see that they're dry.
Additionally, if you're focused on the 6% (out of 11% total) water allocation that goes towards supporting the infrastructure of 22million people over the 50% that goes into non-optimal agriculture (almonds, for instance) in-between the two...then you're missing the forest for the trees, my friend.
Yes, Norcal spent decades wagging fingers at SoCal about this. There were books like Cadillac Desert.
Meanwhile, San Francisco drinks clean glacier water that a valley in Yosemite was destroyed to provide this and they refuse to repurpose a downstream damn that has enough capacity to do it.
Can you clarify what you mean by: “they refuse to repurpose a downstream dam”
California has insufficient water storage to meet demand, it’s not like we have huge dams lying around that we leave empty when there is water available to fill them.
You might be referring to Don Pedro dam - but we are already filling that up (modulo what we need to keep empty for flood control). SF has some contractual right they could possibly exercise to water in Don Pedro but that doesn’t magically result in California’s water supply being held constant if we stop storing water in the Hetch Hetchy. If SF gets the Don Pedro water, that means someone else that was going to get it is deprived.
Now, you could argue that the state can get by with lower storage because ag needs to consume less or more groundwater recharge or whatever, but that’s a different question.
There is and have been entire plans that address this, including in the last real attempt that almost went anywhere. You add things like groundwater recharge. I have groundwater recharge in the basin I live in. But today, in 2026, reductions elsewhere already exceed the capacity of Hetch Hetchy by an order of magnitude.
California has rewilded the Trinity River, resurrected Mono Lake, and has protected a lot of its special places. Voters have voted to tax themselves for state parks. I know that the environment, especially aside from climate issues, just aren't sexy right now. But it still matters to me. I remember the Southern California skies in the 1970s. It's not perfect now, but the improvement despite the increase in people and cars is something to celebrate too.
The only people who oppose this are the rate payers in the place that externalized the cost of the Hetch Hetchy on the world.
Some of the environmental word today might just shout slogans and prattle on with pseudoscientific babble, but there have been a lot of serious people in these efforts, just like with the other cases I mentioned.
Crystal Springs isnt anywhere near Yosemite if that is what you are referencing. That being said it supposedly was gorgeous and almost as amazing before being filled with water
Yeah but the Sierra Club enthusiastically believes that infill housing is evil because a shadow falls on a corner of a park used by underserved minorities. So one must assume that the environment is not anywhere near the top priority for them.
I grew up in Sacramento and I remember when my parents were had a flat rate water bill. Those were the good ol' days!
It frustrates me how everyone moralizes water use rather than accepting that free markets allow for people who are simply willing to pay for it. For example, if you live in Sacrmanto and don't have a pool, you're just doing it all wrong (in my opinion, of course).
I watched my friend's family farm in Modesto flood their fields to irrigate them. No meter, just a valve off the canal and they pay a flat rate. So it offends me that my shower head is legally required to restrict it's flow. Or that neighbors decide that a pile of rock in the front yard is "better for the environment" as it radiates heat on a 105°F day...
In the last 10 years, driven a lot by school shootings, the tide shifted and parents started fighting schools about letting their kids keep their phone "so they can be contacted in emergencies". The schools gave up fighting with the parents.
Laws like this give the school cover to confiscate the phones and say "talk to your congressperson if this bothers you, my hands are tied".
It's more about job seeking than anything. If you jump on a fad early, and it turns out to be the winner, when you're looking for work you can say you have X years of experience with it, which will be a few more than most of the other candidates.
It also shows a passion for learning and improvement, something hiring managers are often looking for signals of.
But of course it's a trade off. This rewards people who don't have family or other obligations, who have time to learn all the new fads so they can be early on the winners.
Does anyone else remember when someone ported kill to the DOOM engine? So you could fire up DOOM and kill processes using different guns for different kill levels?
I will not assume any liability for damage caused from running this code. Especially if you are running it as root. In fact, we both know that this will cause damage to the system, and that's why you want to try it. You have been warned.
What you're talking about has nothing to do with Waymo at all though. It's ostensibly off topic here. You're talking about car culture in general.
Yes, I blame the parents or the adults that were supposed to supervise the child (but not the child). I teach my kids not to run into the street. I also watch them like a hawk near streets because kids are dumb.
I agree with you that we have too strong of a car culture. But we do. So until that changes, we need to teach our kids and adults to be vigilant.
But while we do that, I'd still rather have Waymos around than human drivers.
reply