Which is an AI/LLM thing: "an open standard and open-source framework introduced by Anthropic in November 2024 to standardize the way artificial intelligence (AI) systems like large language models (LLMs) integrate and share data with external tools, systems, and data sources" (Wikipedia).
It's possible to synthesise hydrocarbon analogues of petroluem-based fuels. The problem to date has been that this isn't cost-competitive with petroleum, though the difference is narrower than you might expect. Most famously, a Google X Project attempted this and succeeded technically, but the economics were unfavourable: Project Foghorn: <https://x.company/projects/foghorn/>. Both Germany and South Africa have performed synfuel production (from coal) at industrial scale since the 1930s / 1950s, respectively. Using non-fossil carbon is largely the same chemistry; the process does in fact scale.
Fischer-Tropsch and Sabatier process can both operate with scavenged CO2. There's been some work since the 1990s utilising seawater as a CO2 source, with CO2 capture being far more efficient than from atmospheric sources.
Whilst hydrocarbons have numerous downsides (whether sourced from fossil or renewable sources), they are also quite convenient, exceedingly well-proven, and tremendously useful. In some applications, particularly marine and aviation transport, there are few if any viable alternatives.
Flipside: drastically changing a brand's sizing standards has repeatedly driven me away from long-time favourites.
This happened to me several times from the 1990s through the aughts. Literally between one shopping session and the next, the same style of clothes (tops, bottoms) which had fit perfectly no longer did, resulting both in a set of returns (of clothing) and non-returns (of myself, for future purchases) to those stores. As someone who generally dislikes the shopping experience, additional and insurmountable frictions such as these are absolutely fatal.
More recently (as I've just commented) it's the widespread adoption of stretch fabrics in non-athletic wear. I may want stretch in some of my workout clothes. I don't want it in my street clothing.
I'd noticed the near-universal adoption of stretch fabrics recently, and greatly dislike it. I hadn't considered that this is an inventory optimisation method, though that absolutely makes sense.
This has been a long-unfulfilled promise of technology for literally decades.
I remember multiple iterations of various made-to-measure clothing initiatives, dating to at least the 1990s. Levi Straus offered this in its San Francisco Union Square flagship at one point, as did other clothing producers. One of the few extant references I can find is a 16 year old Tripadvisor question with answers indicating that the option existed many years earlier than that: <https://www.tripadvisor.com/ShowTopic-g60713-i30-k3392413-Ma...>
More recently, in the late aughts / early teens, there was something of a flood of made-to-measure selvage jeans merchants, though I believe these have also largely ceased to operate. When I checked in on them, prices were far higher than off-the-rack options, not by percentages but by multiples, as in 3--5 times as expensive. I suspect that these to remained niche.
The broader lesson is probably that mass-produced, mediocre-fitting clothes simply offer vastly greater economies of scale that no element of automation or technology can overcome.
For men (speaking from experience) and likely women having tailored clothing made in a lower-wage region (Hong Kong, India, and Thailand seem to be the usual suspects) may remain an option. Or the self-provisioning option via a sewing machine, as noted in TFA.
Closer to HN's interests: similar failures of promises for individually-tailored technologically-mediated options to emerge, unless those serve the interests of advertisers or other mass manipulators seems to be a profoundly persistent tendency. In light of present trends (AI / LLM) I'm calibrating expectations accordingly.
Update: Levi Strauss apparently does still offer bespoke jeans, under the "Lot No. 1" label. Pricing begins around $USD500, and the option is offered only in SF, NY, and London.
Erosion, water damage, freeze-thaw cycles, foundation settlement (these are literally mountains of stone built on sand), vandalism and other intentional damage, floodplain evolution.
That last is particularly noteworthy as the Nile famously floods on an annual basis, and that itself is the basis for Egyptian civilisation as those floods created what is still one of the most fantastically productive breadbaskets of the world. Ancient Rome relied on Egypt for grain, and even today demographics data and more vividly night-time satellite light-pollution imagery reveal the Nile as a highly-populated ribbon within a sea of darkness and desolation.
The pyramids have withstood multiple risks for many thousands of years. Despite their simple and rugged overall architecture, that remains impressive.
I've had repeated comment and email exchanges with you over the years over whether or not an explicit "killed" or "dead" indicator on accounts and their posts/comments should exist. I understand the reasons against this, and arguably they'd be more relevant in the case of detected bot accounts (the indicator would be yet more training data, assuming feedback training).
I've also been experimenting with my own indicators for specific accounts based on my own criteria and interactions which I've found useful (applied through my own HN tweaks). E.g., if I see an explicit mod note that an account has been banned, I can mark it as such myself, sparing confusion.
How HN can implement a Voight-Kampff test becomes an increasingly relevant question.... One of several HN needs to consider with increasing urgency.
more important than an indicator it would probably be useful to somehow disable vouching for a killed account. i don't know if it is possible to set the flag counter to some high number so that vouching simply has no effect or to an invalid number like -1
That's antithetical to how HN has operated in the past. Vouching for deads is fair when the account is an actual human, and happens to post valid content. I do this occasionally myself (I read with "showdead" on), though not especially often.
Accounts killed for spamming AI content seem to me to violate that premise, and an unvouchable kill does seem appropriate, especially where it's not immediately evident to the casual reader that an account was killed for posting AI content.
I'm thinking of how I'd like to indicate such accounts myself, and am leaning toward adding a robot emogi via an ":after" CSS rule.
i think we are actually agreeing. i am not talking about making kills unvouchable in general but i am suggesting how an unvouchable kill could be implemented without to much effort. the unvouchable kill should of course be only applied to appropriate cases, it's not meant to replace the regular kill.
Yes, we are in agreement here. I was simply noting what HN's past policy and rationale have been.
LLMs change the calculus somewhat in making automated bot-posting far more viable. It's clearly already a problem. I suspect that moderation policies will have to adapt to this. There's also the fact that such a change would make AI-banned discernable from normal bans, in that AI-banned accounts would not have vouchable comments. If explicitly noting AI banning isn't adopted on the basis that this would provide information to either the AI or its operator of the fact / nature of the banning, the absence of a vouch option would reveal the fact regardless.
(A relatively small example of changes we'll see induced by LLMs in the larger world as well. Interesting times....)
my suggestion would not remove the vouch option, it would just make it ineffective. people using it would not notice. the system would still indicate that you vouched for a message.
Rather famously in at least the case of Google and others, with government funding:
"Google’s true origin partly lies in CIA and NSA research grants for mass surveillance" (January 28, 2025)
The intelligence community hoped that the nation’s leading computer scientists could take non-classified information and user data, combine it with what would become known as the internet, and begin to create for-profit, commercial enterprises to suit the needs of both the intelligence community and the public. They hoped to direct the supercomputing revolution from the start in order to make sense of what millions of human beings did inside this digital information network. That collaboration has made a comprehensive public-private mass surveillance state possible today.
The Massive Digital Data Systems (MDDS) ... program's stated aim was to provide more than a dozen grants of several million dollars each to advance this research concept. The grants were to be directed largely through the NSF so that the most promising, successful efforts could be captured as intellectual property and form the basis of companies attracting investments from Silicon Valley. This type of public-to-private innovation system helped launch powerful science and technology companies like Qualcomm $QCOM +1.61%, Symantec, Netscape, and others.
The Internet itself (particularly its precursor, ARPANET), was also government funded, as was development of the World Wide Web (CERN). Oracle, the database company, grew out of the CIA's Project Oracle.
Which is an AI/LLM thing: "an open standard and open-source framework introduced by Anthropic in November 2024 to standardize the way artificial intelligence (AI) systems like large language models (LLMs) integrate and share data with external tools, systems, and data sources" (Wikipedia).
<https://modelcontextprotocol.io/docs/getting-started/intro>
<https://en.wikipedia.org/wiki/Model_Context_Protocol>
reply