Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[smjjames]

You know, I just saw that commercial with the insufferable little hipster brat talking about "what's a computer" and it hit me: we now get to say Apple is hostile to homosexuals!

Alan Fucking Turing, prosecuted and chemically castrated for being homosexual is responsible for the concepts behind the computer I am typing this on, and which you are reading this on, so fuck Apple entirely.

Huh? Are you talking about that commercial that used to air many years ago with the younger hipster guy representing MACs/Apple and the older guy representing PCs/Microsoft?

[Starver]

Sounds like something else. Certainly the "I'm a PC/I'm a Mac" campaign (as fronted in the UK by a local duo Mitchell & Webb, IIRC) doesn't mentally associate itself with the above complaint in my mind.

I don't suppose there's a linkable demo of the offending material, is there? Just to royally annoy us all?

[smjjames]

Yeah, it's the only Apple ad that I know of which remotely looks like what Max is describing.

[Reelya]

Nah, I think he's referring to some recent ad on TV where another kid asks what this kid is doing on his computer (iPad I guess: i kind of filter ads out: I'll check next time it' on whether it's an Apple ad), and he asks "what's a computer?" which is implying that because you got the i-version of something it has transcended normal categories of stuff. It sounds like standard Apple levels of meaningless fluff.

And btw, the old "Hi I'm a Mac and I'm a PC" was bullshit. The PC should have been a long-haired gamer with black jeans and a metal t-shirt, talking about all the bitching games you can get and mods that he's done on his rig. Mac gaming is a sad joke of an idea.

[Folly]

RIP Net Neutrality.

[redwallzyl]

2010-2012.

Edgy

[Magistrum]

His name starts with ISP, it's a trap!

[Reelya]

http://www.news.com.au/technology/online/hacking/grinch-bots-may-steal-christmas-by-snatching-up-high-demand-toys-regulators-warn/news-story/1d0b631d216257b33d26eaf45a98dfd1

This is a particularly scummy use of tech. Some people are using shopping bots that buy out all the high-demand Christmas toys, then they scalp them on ebay for vastly inflated prices. Of course it's not a viable long-term strategy: if it becomes expected every year then they'll just start making more toys.

[wierd]

Dutch national consumer service, Consumentenbond, warns people that want to buy a new phone not to buy the Samsung Galaxy S7, S7 Edge, or J3 2016.

From march 2018 onwards, Samsung will stop providing updates for these models, which will make their software age fast, and make the phones less secure.

The director of Consumentenbond, Bart Combée says it's "absurd".
"Samsung terminates support of 3 of it's phones less than 2 years after they were put on the market, and won't look back, while those phones are expected to be sold on the market way beyond their support termination date.
As far as we are concerned, Samsung has an obligation here that they can't just walk away from.


So yeah. Don't buy Samsung unless they change their crazy ways.

Forget planned obsolescence, just say fuck it.

Uhm... Use LineageOS instead? Because-- yeah, they are the new cyanogenmod, and they offer WEEKLY images for that series of phone?
https://download.lineageos.org/herolte

[Tack]

Hey, AI guys.

There's a game. It's called 'Universal Paperclips'.

Here's links to the Game and the Topic in OG.
Definitely wanna discuss it with someone, but don't really want to jump into 'spoilers'.

Warnings:
It is a clicklike game and so therefore it's both addictive and time consuming. It's better than the rest of the genre but will definitely eat a few hours.
The good news is that there is an endgame, and dedicated people can reach it rather quickly.

The 'Lategame' section has a research which will start playing music. If you're playing this at work, maybe turn the sound down.

[Tack]

So computers could never be “open goal” oriented?
That’s perhaps more worrying in its own way

[wierd]

Isn't that an expected consequence of the halting problem?

EG, the computer does not even know (and CANNOT know) if it can even continue processing, let alone have absolute certainty that its process will reach the desired outcome.

[Reelya]

I think the "paperclip maximizer" as it's commonly related has been mutated into a completely different argument. In the original version, there's no construct about not knowing how many paperclips it has made, it's making a completely different point altogether than the bastardized version that spread around later. This was the wording of the original version:

https://en.wikipedia.org/wiki/Instrumental_convergence

Suppose we have an AI whose only goal is to make as many paper clips as possible. The AI will realize quickly that it would be much better if there were no humans because humans might decide to switch it off. Because if humans do so, there would be fewer paper clips. Also, human bodies contain a lot of atoms that could be made into paper clips. The future that the AI would be trying to gear towards would be one in which there were a lot of paper clips but no humans.

Notice that there's no mention of 100,000 or 1,000,000 paper clips here, unlike the later version which spread. And there's no mention of the AI not realizing it's already made "enough" paperclips, because that concept is completely missing from the original formulation.

The entire point here is that "weak AI" could be really dangerous, because it has the "blind machine" aspect of traditional technology, along with the self-learning / growth capacity of AI. e.g. if you get in the way of an automated machine, it could fuck you up. Now, add in the capacity for self-learning and growth into that machine, and it's quite possible that a system designed to maximize something-or-other could go off the rails if allowed to grow by itself. This is basically the point:

https://wiki.lesswrong.com/wiki/Paperclip_maximizer

The artificial general intelligence (AGI) won't revise or otherwise change its goals, since changing its goals would result in fewer paperclips being made in the future, and that opposes its current goal. It has one simple goal of maximizing the number of paperclips; human life, learning, joy, and so on are not specified as goals. An AGI is simply an optimization process—a goal-seeker, a utility-function-maximizer. Its values can be completely alien to ours. If its utility function is to maximize paperclips, then it will do exactly that.

So, it could be extremely "clever" from one point of view, but being extremely clever at achieving stuff does not in any way imply the need for conscious self-awareness. Basically it would an an alien intelligence with 100% task-focus at the exclusion of all other values.

[Reelya]

<snip>

[Reelya]

if you read the whole section, he's basically listing things that could go wrong with a badly-designed AI that you can just assume a human wouldn't do. While they're unlikely, it's still worth thinking about.

https://books.google.com.au/books?id=1mMJBAAAQBAJ&pg=PT123&lpg=PT123&dq=if+the+AI+is+a+sensible+Bayesian+agent,+it+would+never+assign+exactly+zero+probability+to+the+hypothesis+that+it+has+not+yet+achieved+its+goal&source=bl&ots=Pnq3RWyaid&sig=Lr3jiE8m9yV1HRgd_8lZ_Gmc6Ak&hl=en&sa=X&ved=0ahUKEwi3o46O267YAhUKzLwKHXEJCKcQ6AEIKTAA#v=onepage&q=if%20the%20AI%20is%20a%20sensible%20Bayesian%20agent%2C%20it%20would%20never%20assign%20exactly%20zero%20probability%20to%20the%20hypothesis%20that%20it%20has%20not%20yet%20achieved%20its%20goal&f=false

The concept of how a machine could "know" that it's task is completed is definitely worth thinking about. What if it's data bank was corrupted and the "counter" for how many had been produced was faulty? what about faulty sensors?