Piper is very good here: It's an abuse of these LLM detection mechanisms to attribute to particular sources. Disappointing that @TuhinChakr didn't continue to engage.
@TuhinChakr@alexolegimas I do not believe you that "the darker glare" can be confidently attributed to a specific fanfiction story. I don't think that this is well established in the current literature. I understand you're not making a copyright claim, but you're making an attribution claim.
@tylercowen This is almost certainly about the release of ChatGPT 5.3 and Claude Opus 4.6. Given the timing of the tweet I assume it's not about model capability so much as either the alignment of announcements or the claim from OpenAI that they used 5.3 to debug it's own training.
Friends-
This is a tough note to write, but since a bunch of you have started to suspect something, I’ll cut to the chase: Last week I was diagnosed with metastasized, stage-four pancreatic cancer, and am gonna die.
Advanced pancreatic is nasty stuff; it’s a death sentence. But I already had a death sentence before last week too — we all do.
I’m blessed with amazing siblings and half-a-dozen buddies that are genuinely brothers. As one of them put it, “Sure, you’re on the clock, but we’re all on the clock.” Death is a wicked thief, and the bastard pursues us all.
Still, I’ve got less time than I’d prefer. This is hard for someone wired to work and build, but harder still as a husband and a dad. I can’t begin to describe how great my people are. During the past year, as we’d temporarily stepped back from public life and built new family rhythms, Melissa and I have grown even closer — and that on top of three decades of the best friend a man could ever have. Seven months ago, Corrie was commissioned into the Air Force and she’s off at instrument and multi-engine rounds of flight school. Last week, Alex kicked butt graduating from college a semester early even while teaching gen chem, organic, and physics (she’s a freak). This summer, 14-year-old Breck started learning to drive. (Okay, we’ve been driving off-book for six years — but now we’ve got paper to make it street-legal.) I couldn’t be more grateful to constantly get to bear-hug this motley crew of sinners and saints.
There’s not a good time to tell your peeps you’re now marching to the beat of a faster drummer — but the season of advent isn’t the worst. As a Christian, the weeks running up to Christmas are a time to orient our hearts toward the hope of what’s to come.
Not an abstract hope in fanciful human goodness; not hope in vague hallmark-sappy spirituality; not a bootstrapped hope in our own strength (what foolishness is the evaporating-muscle I once prided myself in). Nope — often we lazily say “hope” when what we mean is “optimism.” To be clear, optimism is great, and it’s absolutely necessary, but it’s insufficient. It’s not the kinda thing that holds up when you tell your daughters you’re not going to walk them down the aisle. Nor telling your mom and pops they’re gonna bury their son.
A well-lived life demands more reality — stiffer stuff. That’s why, during advent, even while still walking in darkness, we shout our hope — often properly with a gravelly voice soldiering through tears.
Such is the calling of the pilgrim. Those who know ourselves to need a Physician should dang well look forward to enduring beauty and eventual fulfillment. That is, we hope in a real Deliverer — a rescuing God, born at a real time, in a real place. But the eternal city — with foundations and without cancer — is not yet.
Remembering Isaiah’s prophecies of what’s to come doesn’t dull the pain of current sufferings. But it does put it in eternity’s perspective:
“When we've been there 10,000 years…We've no less days to sing God's praise.”
I’ll have more to say. I’m not going down without a fight. One sub-part of God’s grace is found in the jawdropping advances science has made the past few years in immunotherapy and more. Death and dying aren’t the same — the process of dying is still something to be lived. We’re zealously embracing a lot of gallows humor in our house, and I’ve pledged to do my part to run through the irreverent tape.
But for now, as our family faces the reality of treatments, but more importantly as we celebrate Christmas, we wish you peace: “The people walking in darkness have seen a great light; on those living in the land of deep darkness a light has dawned….For to us a son is given” (Isaiah 9).
With great gratitude, and with gravelly-but-hopeful voices,
Ben — and the Sasses
@econoadabsurdam Canberra likely doing some serious lifting here too — though it has a population much smaller than the Gold Coast these days.
https://t.co/TpVo3DRJST
@SDHamiltonVIC@Raf_Epstein Totally agree! Isn't Ukraine choosing to build more when they're in the middle of a land war a good signal they think they're defensible and useful?
I might have misunderstood your point.
Not an expert on the British (?) system and CfD, I work in the Australian energy space.
I'm not arguing that solar + batteries make total system costs go down when compared with, say, black coal generation (in Aus at least). I'm arguing that more Capex for a battery alone doesn't tell you system costs will go up, right? Other Capex might disappear elsewhere on the margin.
Say at midday the battery buys at $10. Then, at peak demand (6:30pm?) sells at $100 and this means NPV=0.
Surely the question for system cost is whether the alternative at 6:30pm is coal $50 or gas $150?
In other words, in this simple ststem adding a battery could (could!) reduce the Capex at 6:30pm as the peaking gas gets outcompeted, reducing system costs?
@7Kiwi@rtweed You're perfectly right about all those metrics.
But my point is that while the price to take NPV to zero may go up, the relevant question is whether the expected capture price goes up an equivalent amount (or less/more) with a battery. No?
This isn't true, because solar drives midday prices to zero, or below. Batteries are additional Capex that allow utilities or consumers to shift energy over time from that point to later in the day.
They're arbitrage machines.
Whether worthwhile or not is another argument. But (1) More Capex therefore (C) higher price is an invalid argument here.
@AlecStapp This is true if you want to understand construction capacity — but also worth bearing in mind that the investment dynamics (and overall system cost) for solar and wind change substantially as their own penetration increases, particularly out past their capacity factors.
I think it is wise to distinguish between nuclear energy (or propulsion) and weapons here. Baxter wanted us to be a nuclear weapons state and also build power stations.
Thankfully, those capabilities became technically separate in his time. When they did, we should have pursued nuclear energy, and only energy.
As @Dr_Keefer notes... accidental decarbonisation is the best kind.
I think 13% just the electricity price drop. They project very optimistic network utilisation (read: control over high demand appliances) seem to exclude Distribution upgrades. Prices go up even in their model if DER remains "uncoordinated".
The household energy expenditure definitely (!) exclude EV capital costs — drops there seem to be because of the petrol to electricity switch.
@themetresgained What was your setup prompt for it?
@gwern has written about it being a critical time to write online, precisely because it influences the output of 3-4 big machines that will steer a lot of thought. Probably true of the policy space....
@TMFScottP More interestingly, the bet seems to have moved the market he used substantially. It potentially should have corrected itself more afterwards: https://t.co/HcqOAxSGqv
@YoucanAxel@Dr_Keefer It's not particularly strange. The green wedge is about 10GWh per year now. This is close to the 16GWh per year 40GW of wind would produce at 40% capacity factor. Moreover, some of it would have been build when costs were higher.
It is certainly now a general revenue-raising tax, but was instituted to fund roads based on road-use, and remains a tax for that purpose.
The Productivity Commission thinks so, for instance, and cite declining excise revenues as they argue for a better tax that is more closely related to road-use as cars get more efficient (and now as they go electric).
"revenue from fuel excise — currently one of the largest sources of funding from motorists — has been lagging well behind growth in road use and the unit cost of building and maintaining roads" https://t.co/BuQzSsfKqN