Posts by Neil Barnes 6255 publicly visible posts • joined 18 Apr 2007
With 125 ASA film, the moon's correct exposure is of course 1/125th at f16.
Call me old fashioned, but I expect a camera to produce an image of the light that falls on its image sensor, not some statistical average of what you photographed ought to look like!
"Wait, that's not Mavis in the wedding photo!"
Every consumer car I've ever seen, the steering column has a shaft with splines on the end onto which the wheel fits, and a segment of threaded shaft which has a bloody great nut on it. Bolts seem to be reserved for attaching an aftermarket wheel to a boss, itself fastened using the forementioned nut.
Of course, I learned how cars worked in the old days, when there was a direct mechanical linkage to the steering rack... perhaps things have changed recently?
But either way, the lack of quality control implied here is... worrying.
Either Orwell: “If you want a picture of the future, imagine a boot stamping on a human face—for ever.”
or Huxley: "Violent Passion Surrogate. Regularly once a month. We flood the whole system with adrenin. It's the complete physiological equivalent of fear and rage. All the tonic effects of murdering Desdemona and being murdered by Othello, without any of the inconvenience."
It did strike me though that they have two critical components that aren't made in house: the combustible bits and the stuff they make the rocket parts from. And given the fun I'm having trying to source jelly bean electronics parts, they probably still have significant external dependencies there...
I am so doomed: not only won't I have a Ring device on the premises, I won't have Alexa and its ilk. I'm not overly happy about what my phone gets up to, either... largely because I don't (and in practical terms, can't) know. But I live in ex-Stasi land, and people haven't forgotten here.
Which is helpful if you've already found life in sufficient quantities to leave traces visible from orbit, or if you're looking for bacteria in a high earth desert, but is possibly not too helpful on, say, Mars or Europa where the local life, if any, may behave in completely different ways.
Mind you: if you sent the current Mars rovers trundling around the Atacama Desert - surely pretty Mars-like apart from the higher pressure and oxygen content of the atmosphere - would they actually detect microscopic life there? Or even the macroscopic tourists?
Anyhow, taking anything serious or for granted on the net without proper validating by you yourself is just asking for trouble.
Which reduces to the point I raised a few days ago: the only way you can be certain that it's me you're talking to is to meet me in person. A video or audio call is (or is close to) insufficient, and a text conversation on a forum such as this is even more so.
Proper validation has returned to seeking the original primary source of information; the internet will be terminally broken as an information source as long as these idiocies abound.
"No Dave, I'm not going to tell you what you asked; I'm going to tell you what the accumulated response of a billion similar questions was."
- build a cathedral
- ensure that it contains a long pendulum
- allow a passing polymath to observe the rate at which that pendulum swings
- calculate the expected rate (taking into account the local gravitational field) and note the difference between the expected and observed rates
- $
Or, more sensibly: what's wrong with UTC?
Well, you should be able to ask any search engine for exact search terms and exclusions and actually get a result on that basis.
Instead you get curated results that would rather sell you a holiday in Devon (or a holiday home!) than give you the information you are seeking.
Search machines - with CGPT or not - are sometimes (often) useful, but they do not have your interests at heart. Not if there's something that can be sold to you.
The only way now to talk to someone and *know* you are talking to them and not someone else (or some collection of statistical subroutines) is to do it in person, face to face.
Hmm. I wonder if that's why my employer now wants daily standup meetings with the video turned on? And that's why all the managers want back-to-the-office working?
Well, the magnetic field on a CRT from both the scan coils and the correction magnets on the neck are _ever_ so slightly stronger than the earth's magnetic field. They can pick up a magnetic field on the frame and the shadow mask; that's why colour screens always had a degaussing coil built in to operate when they were turned on. But the effect is of such a magnitude as to move the scanning electron beam from one colour phosphor dot to an adjacent one, and change the colour slightly.
I'm not sure about Europe, but in the UK there is no need to register a work for copyright; it exists as soon as the work is created in a physical form: when I print out a book, slap pigments on a canvas, or develop a photograph for example.
https://www.gov.uk/government/consultations/artificial-intelligence-and-ip-copyright-and-patents indicates that the government have thought about but at present:
For AI-devised inventions we plan no change to UK patent law now. Most respondents felt that AI is not yet advanced enough to invent without human intervention.
Analog was one of the original pulp magazines (though with several titles: Astounding Stories of Super-Science to begin with) and it's still going reasonably strong. It also accepts unsolicited submissions and to be honest I'd expect them to be an obvious target for this statistical vomit. Though I wonder about the chat-gpt training: if it is informed by the quality of freely available material, it won't be too hard for the editors to tell the difference. And noting that many of the early SF stories _are_ available, it's likely that the output will show the same cultural mores as the authors of the time presented. That should work well!
I think it was Analog who published a story in which a computer took over writing for an author, getting better and better until every book published was by this 'author'... book event horizon?
A quick search tells me that Meta has around three billion users as of last year, and that they had four and a half billion in income for 2022Q4 - so say, eighteen billion dollars income for three billion users, or six bucks each per year.
And they'd like users to pay twenty-four times that amount? And presumably still get the six bucks from advertisers (because a billion here, a billion there...)
And in the meantime, require said users to submit copies of government provided ID? Can we say 'hackers paradise'?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
