Think Different.
17,701 karma · joined November 12, 2010
Think Different.
The server at www.haproxy.com is taking too long to respond.
Not the best advert for a high availability proxy...
You won't be satisfied unless the actual word 'literal' appears? It appears in the first paragraph of the Stercoranism link, and again in the second paragraph.
"the 9th century Carolingian theologian Paschasius Radbertus... wrote an influential tract around 832 upholding the literal interpretation of Christ in the Eucharist". He supported the literal change, but said that the holy bits dissolved before becoming poo.
If you check out the Middle Ages section, it shows that the theological debate that it wasn't physical alteration started... at a date that is closer to us than the birth of jesus.
Check out Stercoranism [1] as well (which contributed to the above debate), whose whole basis is that the doctrine of physical change must lead towards normal digestive processes happening, and wondering if this turns the eucharist into, literally, holy shit.
At the end of that article is a bit of modern apologia stating that christ probably leaves as soon as the cracker hits your stomach ('but nobody knows precisely when'[2]). :)
[1] https://en.wikipedia.org/wiki/Stercoranism
[2] god does, after all, move in mysterious ways...
https://en.wikipedia.org/wiki/Transubstantiation
Plenty of people have believed it historically. Scroll down to the Catholic section and you'll see some of the mental gymnastics I talked about to claim it in modern times.
Protestants don't believe in literal transubstantiation; I'd say they probably trust their taste-buds more, but then again, compare Protestant/Anglo-German food against Catholic/Franco-Italo-Spanish food :)
> "questions that have no answer". They have more perhaps retreated to "god of the gaps"
I don't personally see a difference between these terms - they both mean claiming to have an answer for something that is unanswerable. If you have an answer for which the only proof is basically "just trust me", then it's not much of an answer. Russell's Teapot is a pretty clear example of this.
It's exactly the same with women's rights and minority civil rights: "Really, people would listen to them more if they just toned down and spoke politely". We already know that that doesn't work and just gets you ignored.
[1] https://www.washingtonpost.com/blogs/govbeat/wp/2014/07/08/t...
So where does the 'useful dialogue' start? I think it starts well before we just give moral authority to a bunch of very old men arguing about rules that only they made up. They should give us a better chain of authority than "since time immemorial".
Truth is that economies change slowly, and politicians have much less control over it than we all like to believe.
A colleague of mine had a son who needed to get seen there. They asked him to fax in the admissions forms. Who has a fax machine? He found somewhere and faxed them in, in a couple of jobs. Nope, can he fax them so they all come through in one continuous job. So he goes back and faxes them again. Nope, can he fax them so that the documents arrive in the right order? So he decides to go to a different children's hospital in the city.
I had a friend who worked in the QA dept of that hospital and tried to get processes improved. He said that any time you tried anything, the relevant head of department would say "If we make this change, children will die". Everyone around the table would know the lie for what it was, but the head of dept had the final say if it involved child health.
In short, faxes will still be going strong there for a while yet, and they're a big hospital. :)
At the time, he thought that schizophrenia would break up into four basic groups, based on positive/negative affect (does an episode have the person doing extra activity or withdrawing) and, I think, whether or not the person has hallucinations.
I haven't really kept in touch with the literature to see how his 4-way prediction went, but I have never seen anything to counter the idea that schizophrenia is more than a single disease. And, of course, classification of disease is a key element of treating it, so if schizophrenia is not being properly classified...
True, I wouldn't pay either $1 or $10 to generate a meme, but I wouldn't 'pay' $0 either.
I'd also discourage a patreon- or kickstarter-style offering for a B2B product, because it signals very clearly that you're not playing with the big fish.
This whole 'old people a dumb as rocks when it comes to tech' meme needs to die. Some old people have trouble with tech. Plenty of them don't.
What does this mean? An AI that reads text still has to decipher it. Show it a newspaper and it still has to analyse what it sees and convert it into something sensical; it doesn't have mystical powers that absorb the information.
Even if you're talking wire protocols, it still needs to decipher the incoming data. This series of 1s and 0s makes these ascii characters. This series of ascii characters makes this sentence. This sentence means 'foo'.
> The AI can even share trained neural ensembles or any computation device to perform a specific task
It was a massive coup when they designed a robot that could fold laundry. Not do the laundry, just fold it. How many other AIs has that been shared to?
I can only read the first paragraph of the article due to paywall, but it's making the case that 'computers are faster than humans' and they are... for certain tasks. Particularly ones with clear-cut rules. Machine learning and neural networks can do a lot of 'fuzzy' thinking, but they also make a lot of errors that fans just gloss over, errors that a human would never make (see the google image of two black people that got AI-tagged as 'gorillas' for one that made the news). Google did machine learning with their sketch program and it could tell if you drew a bicycle or a car... if you drew them in the stereotyped cartoonish side-on way. It had no idea what was going on if you drew it from a non-stereotyped angle. Fancy pattern-matching just doesn't compare to human understanding on that task.
> A man in the middle ages thinks of the future of a time with more castles, more knights, more kings.
A man in the middle ages also thought there were devils in the cemetery, that unicorns were a thing, and that alchemy was a way to get rich quick if you could just find out how... and all of these things turned out to be bupkis. He was never in danger from being abducted by devils if he took a shortcut through the church's side yard.
> Given a human and a strong AI (human level intelligence AI), the AI would prevail.
Until the AI has full control of its supply chain, from the tantalum mines in Africa, to the chip fabs in China, and the power plants in <locale>, it will not prevail. Parts break; it needs to replace and repair them. It needs 'food' (power) to think. The only real future where we have to seriously worry about a strong AI is one where it is being controlled by (or working with) a hostile party - ie: it's humans vs humans + AI, not humans vs AI.
Or if you mean on a personal level, depends where we are. If it controls the environment (doors etc), then it can just imprison me, and the strength of the AI (or even if it is an AI at all) is irrelevant. But if it can't do that, well... I can just go and flip the breakers, and wait for the power backup to drain.
Or, if you want an android-shape AI, keep in mind that AlphaGo was run on 1200 CPUs and 200 GPUs, and all it could do was play Go (albeit rather well). That's the kind of computational power required to do things 'the computer way', and requiring commensurate power and cooling. Moore's Law has already buckled, and chips are already being designed down to the atomic scale; there isn't much more miniaturisation to be done. Trying to fit a 'super AI' capable of all the things fans want into a form-factor of a human just ain't going to happen (kinda like alchemy).
We don't. We have massive parallelism in how we communicate. The aphorism "a picture is worth a thousand words" is based on that.
All in all, AIs beat humans on very specific problems, but they're worthless for most tasks - and an AI that's specialised to one task is very rarely suitable for another. You can get a human to do quite complex tasks and use initiative with only the simplest of instructions. Don't even need to be at expert-level. Where is the AI that can wash, dry, and fold the clothes, change the baby, make lunch and dinner, fix the leaky sink, and bring the dog in at night? Yet that's a very humdrum day for a human.
Cable internet is shared. Most of the NBN rollout (HFC) is on ageing cable internet puchased from another vendor.
s/some/most/. HFC is the predominant plan for most of the capital cities (I think not... hobart? or was it adelaide?).
Party A was in power, and having infighting problems. Party B was headed by an absolute arse of a man, who used the tactic "Say No! to EVERYTHING". Didn't matter how benign or banal, the opposition party would say no. A flagship policy like the NBN was particularly focused on - and if you look at interviews with the opposition communications minister at the time (now our prime minister), it was clear that he was toeing the party line and didn't think changing the policy was a good idea.
Fast-forward to the next exection, and Absolute Arse gets into power. Turns out he can't govern at all (despite 20 years on the front benches) and he definitely isn't going to backtrack on something he's attacked so routinely. He's kicked out by his own party. At this stage, the NBN has been gutted, and its doom is sealed.
In steps the comms minister as the new PM, but he is now beset by factionalism. He knows what a turkey the NBN is, but he doesn't have the political power to do an about-face on his party's policies.
In short, we're fucked. By the time any other party can get in, the NBN will be 'mostly complete'. The most annoying thing is that the folks who really need good internet is business, not residences (think graphic design, medical imagery, etc), and the party that fucked the NBN is the 'party of business'. Enterprises can already dictate their own network backbones, but small-to-medium businesses cannot.