In a virtual neural net, when the output is close to what you want, you promote the inputs to the neurons that fired. When the output is not what you want, you demote them.
At a complete guess (I can only see the front page of the scientific papers, and probably wouldn't understand them if I went further anyway), they have some electrical or chemical means to reinforce neural links that have recently fired .
I'm fairly comfortable with claiming that worms don't suffer. I'm not sure what the boundaries of my own personal definition of "suffer" is, but I'm pretty sure an organism must possess a brain in order to experience it.
I think it would be helpful if you explained what you mean by "machine" in that case.
I'm trying to do it for you -- anticipating what I guess is your reasoning -- but I'm having an awful lot of trouble doing so without explicitly saying "unless it's alive". And that has the special problem that you then have to define "alive".
I'm more interested to know if others think it's less of a living being.
"It is still a challenge for scientists and philosophers to define life in unequivocal terms" (thanks Wikipedia)
I think you might be able to describe the test tube full of cells that's "piloting" the robot as alive. It's made of biological cells. Presumably it consumes nutrients.
Ethically, the most troublesome part is harvesting the cells from a rat foetus (which I suspect not many/. readers would object to.) From then on, it's at something like the level of a worm, if that.
In the case yes. But if it was possible t culture a full brain and in some way recover all memories from the dead brain to the cultured one?
It's a reasonably interesting thought experiment, but not a particularly new one (you don't even have to be particularly highbrow: Total Recall; Blade Runner; The Island).
Since you're speculating about almost completely sci-fi possibilities, why not just cure your terminal illness, or make an exact copy of yourself minus the disease (with some sort of molecular-level copier), or transfer your memories into a frozen clone of yourself, etc. ?
The cells are harvested from a rat foetus. They're grown in a special vessel, where they're in contact with an array of electrodes. They spontaneously arrange themselves into a neural network. The difficult part is training that network to do anything useful.
So there must be messaging back into the rat. So the robot is to some extent controlling the rat brain.
I'm pretty sure the rat is out of the picture. Probably dead (it doesn't seem worth the hassle to harvest braincells from a rat in a non-lethal manner).
From what I can tell, all that's going on is that they've constructed a neural network out of real neurons.
Wait, he used to be a cyborg and then decided a change of career was in order?!
Kevin Warwick is a fanatical self-publicist. He implanted a chip in his arm, which was able to read nervous signals and forward them to a computer, whereby he could operate robot arms etc. By virtue of that, he proclaimed himself a cyborg. You can buy his book about it, "I, Cyborg" if you really want to.
I think there are people who believe that there is some non-physical aspect to living things that separates them from machines. A "soul" for want of a better word.
Not me. I agree with you, we are (very complex) machines.
I suppose your reasoning is that if it works with cells from a rat's brain, it must have potential to work even better with cells from a human's brain, because humans are cleverer, right?
The thing is, there's not much difference between a rat's neuron and a human neuron, and both are very simple. In essence, they accept signals on their dendrites, and if the signals reach some threshold, they fire a signal from their axon, which typically is connected to the dendrite of another neutron.
I *guess* the advantage of using biological neurons instead of software or silicon is that it's easier to make/harvest vast quantities of them
But I can't see that human cells would be any better than rat cells, and just imagine the ethical objections from the God Squad!
How is this "device" with its CPU, memory, display, and I/O capabilities any less of a computer than an Apple ][ or a Commodore 64?
It's a classic matter of semantics. Just as people (understandably and justifiably) fail to agree on whether "Operating System" just means the kernel or whether it also includes/usr/bin/grep and X11, because they're thinking at different contextual levels, so people use "computer" to mean different things.
You could get quite wordy defining what "computer" means and what "device" means, such that the iPad is one and not the other. However I think we can all infer what kind of properties would make it a "computer" in that sense.
I would say that the only thing preventing the iPad from being a "computer" in that sense, is that Apple deliberately cripple it.
Perhaps "cripple" is too pejorative a word; you could argue that the "missing" features are features that confuse a certain class of user, and are thus better not included.
But if you were to leave the range of your router. Or go out of the country where you really don't want to be using your 3g radio.
My (offline, TomTom) satnav is pretty useless if I take it off the road. It's also pretty useless if I go out of the country, since it only has UK maps on it.
It remains pretty damned good at the purpose for which I bought it.
Likewise, if I buy a device for doing webby stuff around the home, it's entirely irrelevant how useful it is in other scenarios. You might as well complain that it doesn't work underwater. "But if you were to take it scuba diving"...
Just in case you weren't joking, "form" or "version" would probably have been a better word than "state". Let's assume English is not their first language.
ChromeOS is Google's official version, which presumably contains some closed-source components, and is unreleased.
Chromium OS is the fully open source version which you can get your hands on right now.
The guy was able to install Chromium OS on the iPad because (a) he could get his hands on the source and (b) compile it for any target architecture he chose.
For now if you have a machine capable of running fast on the web, you might as well have a more fully featured OS.
For now, you have no choice, since ChromeOS isn't finished.
I think/hope that by the time it's released as a consumer-ready product, it won't feel as if it lacks features.
It *will* run apps locally, it's just that those apps will be implemented in Javascript/HTML5 and the namespace for storage will be remote URLs.
I'm pretty sure that caching, Gears (or the features of Gears that made it into HTML5), etc. a great deal will work offline. That requires work on the part of the web app developers, but in the case of the Google apps at least, they'd be fools to release ChromeOS without making the apps offline-able.
Google's intention is that a non-technical user won't need to be aware that their "GMail" or "Word processor" window is really a browser.
(Note GMail works offline, after a fashion, already)
Well, it was kind of handy for a while when I was sending a lot of messages, and having a real keyboard was a boon.
Advantages over AIM/whatever
1. It definitely works where I live (UK) 2. The recipient sees my phone number as the sender 3. The recipient is not billed
I'm aware that US mobile phone billing is significantly different from the UK, but I'm not sure of the details. Here, it's always the sender that's billed (*) - which is one reason Twitter ceased to offer SMS-out service in the UK.
(* OK there are exceptions, but I'm talking about a consumer texting another consumer)
Are your friends particularly tech-savvy? I mean seriously, going to Android because of available emulators? I think you are under the massively broken assumption your friends remotely approximate the total user population.
I'm not sure about this. I see a lot of not particularly technical people buying flash carts and pirating Nintendo DS ROM images from the Internet. The people who make these products have made them pretty user friendly, so much so that I've spoken to people who are clever enough to do it, yet not clever enough to understand the legal implications ("It's not illegal at all; you can just get them for free off the Internet").
So it's vaguely realistic that non-technical people would ask "so, can I play my Sonic the Hedgehog ROMs on this phone?"
$60/mo sounds like a lot for mobile broadband. Maybe the US market is different from here.
I looked up MiFi after replying to your other post -- apologies, it is not a US-only product.
I dunno, it might be the circles I move in, but it feels as if almost everyone has a smartphone, and is therefore already paying for mobile broadband. If my home broadband were to fail (and let's face it, that's rare anyway), I'd tether to my phone. I'd probably turn off Windows Update and go easy on the YouTube/Spotify/etc. but if there was something really important I guess I'd bite the bullet and pay for the excess data -- but remember this is an exceptional situation.
I appreciate some people's smartphones don't do tethering, but hey, that's just thoughtless buying:) I have to tether over USB, but Android 2.2 can turn your phone into a WiFi AP.
I walk to work. Some public transport has free WiFi, and Google's expectation is that internet connectivity gets more and more ubiquitous over the next few years.
Some people don't have $1,440 for a two-year subscription to MiFi service in addition to what they're already paying for Internet access at home.
Sorry, I'm not in the US so I don't know what MiFi is.
I pay £23/mo for a two year contract. So that's less than US$1000 over two years, and includes an HTC Desire. If I'd bought the device outright the monthly charge would be less.
Admittedly this is capped at 500MB/mo, which is more than enough for everything except audio/video streaming. I count that as "effectively" uncapped.
An extra £2/mo would bump that up to 1GB/mo, and most iPhone users have no limit whatsoever on their mobile data.
As I said, lots of iPhone/Android apps rely on the Internet being available, and people don't tend to complain.
Note that ChromeOS is not intended to be completely at sea when the network is temporarily unavailable (think train routes that leave 3G coverage from time to time). Gears and HTML5 provide mechanisms for making web apps work offline, and Google want apps to take advantage of that.
Indeed, in terms of user experience it is a world of difference between the cirppled Google Docs word-processor interface and what OpenOffice Write can offer.
I hate webapps of today. They offer half of the functionality of real applications and the only "features" they have is that you have your data in an external server.
I think that's a fair statement. If you're using the advanced features of a local word processor, the current crop of web apps probably don't have them. As it happens, I have no use for those advanced features -- in fact they'd probably clutter up the UI without offering me anything I need -- but that's fine, different people have different needs.
I have Word and Excel installed on my laptop, but for personal use I prefer Google Docs word processor and spreadsheet. The features that are missing are features I don't need. I really value the ease with which you can share documents with others, and access the same document from multiple computers (my work PC, my home Mac, my phone, a friend's PC, a cybercafe).
But again, your needs are different to mine. Both kinds of app will continue to exist for some time to come.
Having said that, more feature-rich web apps will emerge. There will be competition, and consumers will demand more features. At the same time, browsers and network connections will get faster, UI libraries will get cleverer. I'm not sure how long it will take, but I suspect that in time Web apps will get as feature-rich as you want.
Let's get real -- this is a hacker demonstrating their own L33T SK1LLZ, and ChromiumOS's portability. Nobody in their right mind is suggesting that a typical end user would want to replace iOS with ChromeOS on an iPad.
What it does mean is that a hacker/developer who wants to try out ChromeOS on a touchscreen, and happens to own an iPad, can do so. If you want to develop for both platforms, well now you can test on both with less outlay on hardware.
I get ChromeOS. I don't get it on a tablet. It's obviously focused towards netbooks and pc's where you want an "instant on" option. It's designed to be used like a desktop browser...with a mouse or track pad. Tablets should have a mobile OS designed around touch screens.
I agree with this, and had kind of assumed that ChromeOS would have a tablet flavour with those kind of features -- multitouch zoom and drag, a good onscreen keyboard etc. The Wikipedia page on Chrome doesn't mention any of this -- but I would be very surprised if it was missing from the first official Google release of ChromeOS.
Slashdot Mirror
Therein lies the challenge.
In a virtual neural net, when the output is close to what you want, you promote the inputs to the neurons that fired. When the output is not what you want, you demote them.
At a complete guess (I can only see the front page of the scientific papers, and probably wouldn't understand them if I went further anyway), they have some electrical or chemical means to reinforce neural links that have recently fired .
I'm fairly comfortable with claiming that worms don't suffer. I'm not sure what the boundaries of my own personal definition of "suffer" is, but I'm pretty sure an organism must possess a brain in order to experience it.
In the sense I mean, yes.
I think it would be helpful if you explained what you mean by "machine" in that case.
I'm trying to do it for you -- anticipating what I guess is your reasoning -- but I'm having an awful lot of trouble doing so without explicitly saying "unless it's alive". And that has the special problem that you then have to define "alive".
I'm more interested to know if others think it's less of a living being.
"It is still a challenge for scientists and philosophers to define life in unequivocal terms" (thanks Wikipedia)
I think you might be able to describe the test tube full of cells that's "piloting" the robot as alive. It's made of biological cells. Presumably it consumes nutrients.
Ethically, the most troublesome part is harvesting the cells from a rat foetus (which I suspect not many /. readers would object to.) From then on, it's at something like the level of a worm, if that.
In the case yes. But if it was possible t culture a full brain and in some way recover all memories from the dead brain to the cultured one?
It's a reasonably interesting thought experiment, but not a particularly new one (you don't even have to be particularly highbrow: Total Recall; Blade Runner; The Island).
Since you're speculating about almost completely sci-fi possibilities, why not just cure your terminal illness, or make an exact copy of yourself minus the disease (with some sort of molecular-level copier), or transfer your memories into a frozen clone of yourself, etc. ?
I followed some links. http://journals.pepublishing.com/content/b31654739h7nk726/
The cells are harvested from a rat foetus. They're grown in a special vessel, where they're in contact with an array of electrodes. They spontaneously arrange themselves into a neural network. The difficult part is training that network to do anything useful.
So there must be messaging back into the rat. So the robot is to some extent controlling the rat brain.
I'm pretty sure the rat is out of the picture. Probably dead (it doesn't seem worth the hassle to harvest braincells from a rat in a non-lethal manner).
From what I can tell, all that's going on is that they've constructed a neural network out of real neurons.
Wait, he used to be a cyborg and then decided a change of career was in order?!
Kevin Warwick is a fanatical self-publicist. He implanted a chip in his arm, which was able to read nervous signals and forward them to a computer, whereby he could operate robot arms etc. By virtue of that, he proclaimed himself a cyborg. You can buy his book about it, "I, Cyborg" if you really want to.
I think there are people who believe that there is some non-physical aspect to living things that separates them from machines. A "soul" for want of a better word.
Not me. I agree with you, we are (very complex) machines.
I suppose your reasoning is that if it works with cells from a rat's brain, it must have potential to work even better with cells from a human's brain, because humans are cleverer, right?
The thing is, there's not much difference between a rat's neuron and a human neuron, and both are very simple. In essence, they accept signals on their dendrites, and if the signals reach some threshold, they fire a signal from their axon, which typically is connected to the dendrite of another neutron.
I *guess* the advantage of using biological neurons instead of software or silicon is that it's easier to make/harvest vast quantities of them
But I can't see that human cells would be any better than rat cells, and just imagine the ethical objections from the God Squad!
How is this "device" with its CPU, memory, display, and I/O capabilities any less of a computer than an Apple ][ or a Commodore 64?
It's a classic matter of semantics. Just as people (understandably and justifiably) fail to agree on whether "Operating System" just means the kernel or whether it also includes /usr/bin/grep and X11, because they're thinking at different contextual levels, so people use "computer" to mean different things.
You could get quite wordy defining what "computer" means and what "device" means, such that the iPad is one and not the other. However I think we can all infer what kind of properties would make it a "computer" in that sense.
I would say that the only thing preventing the iPad from being a "computer" in that sense, is that Apple deliberately cripple it.
Perhaps "cripple" is too pejorative a word; you could argue that the "missing" features are features that confuse a certain class of user, and are thus better not included.
A "feature" would be something that is non-standard.
I think your version of the English language is non-standard -- no offence intended, honest :)
"Feature" just means "something it has".
For example, my nose is a feature. A door is one of the features of my house.
But if you were to leave the range of your router. Or go out of the country where you really don't want to be using your 3g radio.
My (offline, TomTom) satnav is pretty useless if I take it off the road.
It's also pretty useless if I go out of the country, since it only has UK maps on it.
It remains pretty damned good at the purpose for which I bought it.
Likewise, if I buy a device for doing webby stuff around the home, it's entirely irrelevant how useful it is in other scenarios. You might as well complain that it doesn't work underwater. "But if you were to take it scuba diving"...
Then Google can introduce Chrome OS in a few years.
I'm not sure what the release schedule is, but I think mass market Chrome devices are more than a year away.
Until then, web apps designed for Chrome OS will need to be designed to run offline for hours at a time.
I expect that to be the case too.
Just in case you weren't joking, "form" or "version" would probably have been a better word than "state". Let's assume English is not their first language.
ChromeOS is Google's official version, which presumably contains some closed-source components, and is unreleased.
Chromium OS is the fully open source version which you can get your hands on right now.
The guy was able to install Chromium OS on the iPad because (a) he could get his hands on the source and (b) compile it for any target architecture he chose.
For now if you have a machine capable of running fast on the web, you might as well have a more fully featured OS.
For now, you have no choice, since ChromeOS isn't finished.
I think/hope that by the time it's released as a consumer-ready product, it won't feel as if it lacks features.
It *will* run apps locally, it's just that those apps will be implemented in Javascript/HTML5 and the namespace for storage will be remote URLs.
I'm pretty sure that caching, Gears (or the features of Gears that made it into HTML5), etc. a great deal will work offline. That requires work on the part of the web app developers, but in the case of the Google apps at least, they'd be fools to release ChromeOS without making the apps offline-able.
Google's intention is that a non-technical user won't need to be aware that their "GMail" or "Word processor" window is really a browser.
(Note GMail works offline, after a fashion, already)
The manual for the BBC Micro included some comforting words about nothing you could type in being able to do permanent damage to the hardware.
:)
However something like:
10 *TAPE 1
20 *TAPE 0
30 GOTO 10
... would burn out the cassette control relay if you left it long enough
Also you could fry a CRT monitor with a bad X modeline...
Well, it was kind of handy for a while when I was sending a lot of messages, and having a real keyboard was a boon.
Advantages over AIM/whatever
1. It definitely works where I live (UK)
2. The recipient sees my phone number as the sender
3. The recipient is not billed
I'm aware that US mobile phone billing is significantly different from the UK, but I'm not sure of the details. Here, it's always the sender that's billed (*) - which is one reason Twitter ceased to offer SMS-out service in the UK.
(* OK there are exceptions, but I'm talking about a consumer texting another consumer)
Are your friends particularly tech-savvy? I mean seriously, going to Android because of available emulators? I think you are under the massively broken assumption your friends remotely approximate the total user population.
I'm not sure about this. I see a lot of not particularly technical people buying flash carts and pirating Nintendo DS ROM images from the Internet. The people who make these products have made them pretty user friendly, so much so that I've spoken to people who are clever enough to do it, yet not clever enough to understand the legal implications ("It's not illegal at all; you can just get them for free off the Internet").
So it's vaguely realistic that non-technical people would ask "so, can I play my Sonic the Hedgehog ROMs on this phone?"
$60/mo sounds like a lot for mobile broadband. Maybe the US market is different from here.
I looked up MiFi after replying to your other post -- apologies, it is not a US-only product.
I dunno, it might be the circles I move in, but it feels as if almost everyone has a smartphone, and is therefore already paying for mobile broadband. If my home broadband were to fail (and let's face it, that's rare anyway), I'd tether to my phone. I'd probably turn off Windows Update and go easy on the YouTube/Spotify/etc. but if there was something really important I guess I'd bite the bullet and pay for the excess data -- but remember this is an exceptional situation.
I appreciate some people's smartphones don't do tethering, but hey, that's just thoughtless buying :)
I have to tether over USB, but Android 2.2 can turn your phone into a WiFi AP.
I walk to work. Some public transport has free WiFi, and Google's expectation is that internet connectivity gets more and more ubiquitous over the next few years.
Some people don't have $1,440 for a two-year subscription to MiFi service in addition to what they're already paying for Internet access at home.
Sorry, I'm not in the US so I don't know what MiFi is.
I pay £23/mo for a two year contract. So that's less than US$1000 over two years, and includes an HTC Desire. If I'd bought the device outright the monthly charge would be less.
Admittedly this is capped at 500MB/mo, which is more than enough for everything except audio/video streaming. I count that as "effectively" uncapped.
An extra £2/mo would bump that up to 1GB/mo, and most iPhone users have no limit whatsoever on their mobile data.
As I said, lots of iPhone/Android apps rely on the Internet being available, and people don't tend to complain.
Note that ChromeOS is not intended to be completely at sea when the network is temporarily unavailable (think train routes that leave 3G coverage from time to time). Gears and HTML5 provide mechanisms for making web apps work offline, and Google want apps to take advantage of that.
Indeed, in terms of user experience it is a world of difference between the cirppled Google Docs word-processor interface and what OpenOffice Write can offer.
I hate webapps of today. They offer half of the functionality of real applications and the only "features" they have is that you have your data in an external server.
I think that's a fair statement. If you're using the advanced features of a local word processor, the current crop of web apps probably don't have them. As it happens, I have no use for those advanced features -- in fact they'd probably clutter up the UI without offering me anything I need -- but that's fine, different people have different needs.
I have Word and Excel installed on my laptop, but for personal use I prefer Google Docs word processor and spreadsheet. The features that are missing are features I don't need. I really value the ease with which you can share documents with others, and access the same document from multiple computers (my work PC, my home Mac, my phone, a friend's PC, a cybercafe).
But again, your needs are different to mine. Both kinds of app will continue to exist for some time to come.
Having said that, more feature-rich web apps will emerge. There will be competition, and consumers will demand more features. At the same time, browsers and network connections will get faster, UI libraries will get cleverer. I'm not sure how long it will take, but I suspect that in time Web apps will get as feature-rich as you want.
There is a difference between "actively attempts to prevent" and "does not provide the means to".
There is. And exploits are workarounds to the former.
Let's get real -- this is a hacker demonstrating their own L33T SK1LLZ, and ChromiumOS's portability. Nobody in their right mind is suggesting that a typical end user would want to replace iOS with ChromeOS on an iPad.
What it does mean is that a hacker/developer who wants to try out ChromeOS on a touchscreen, and happens to own an iPad, can do so. If you want to develop for both platforms, well now you can test on both with less outlay on hardware.
I get ChromeOS. I don't get it on a tablet. It's obviously focused towards netbooks and pc's where you want an "instant on" option. It's designed to be used like a desktop browser...with a mouse or track pad. Tablets should have a mobile OS designed around touch screens.
I agree with this, and had kind of assumed that ChromeOS would have a tablet flavour with those kind of features -- multitouch zoom and drag, a good onscreen keyboard etc. The Wikipedia page on Chrome doesn't mention any of this -- but I would be very surprised if it was missing from the first official Google release of ChromeOS.