Slashdot Mirror
Five Ideas That Will Reinvent Computing
prostoalex writes "PC Magazine looks at 5 ideas that will reinvent computing. IMAX-quality movies at home with new projectors, a mid-air mouse that requires no flat surface, a home quantum computer, a router-based peer-to-peer system, and a man-made brain all made the list."
How about the concept that software is a service and not a product.
Also, hasn't this existed for years now?
"A great democracy must be progressive or it will soon cease to be a great democracy." --Theodore Roosevelt
Most of these ideas look more like cool gadgets or specific applications to me.
Computing is everywhere now. I think a "re-invention" of it should probably be something that applies to the huge numbers of people who use computing as part of their everyday lives.
I was much more interested in these comments, which involve trying to fundamentally change the way in which we use our technology.
Peter
Although I must agree it is a very interesting and challenging project, I'm not convinced it will see much use in the biological research world. For starters there is one particular reason I have my doubts (from the article):
"The project is particularly daunting when you consider that modern neurology has yet to explain how the brain actually works. Yes, we know the fundamentals. But we can't be sure of every biological transaction, all the way down to the cellular level. Three years into this Cognitive Computing project, Modha's team isn't just building a brain from an existing blueprint. They're helping to create the blueprint as they build. It's reverse engineering of the highest order."
Although reverse engineering might seem as the perfect way to find out how something works from a technology point of view, this isn't necessarily true from a biological point of view. The thing is that when you reverse engineer a piece of technology you are completely certain about the underlying core principles because the technology you are reverse engineering is actually man made. With biology this is not the case.
I work in a research group that (amongst other things) tries to reverse engineer simple cellular pathways of a complexity scale that is that is far smaller than the actual function of the brain and even those attempts, although producing results, do not always conform with biological reality.
Furthermore, and I know this from experience, computer scientists and mathematicians tend to underestimate the actual complexity of such systems and the variability of biological systems within species and even within the same organism. It's not just a matter of mathematically connecting nodes in a neural net.
Let's be reasonable here, because the important aspects of the system biology of relatively "simple" biological systems remain largely elusive and difficult to simulate at the moment because not every core principle is known it seems a bit over confident to claim to simulate one of the most complex systems known without even having a complete rule book in your hands.
Nevertheless, an interesting project.
In Soviet Russia elephant rides you!
Bah, no mind-computer interfaces? no eye tracking devices?
both are more promising than the "3d mouse"
The "router-based peer-to-peer system" isn't all that revolutionary: the load-spreading system they describe is similar in many ways to a system of caching web proxies (good) mixed with Steam (evil). The article also describes a content-centric model of accessing data as opposed to a server-centric model, and that's kind of cool, but I don't have a whole lot of faith in that sort of thing right now.
What I THOUGHT they were talking about when I read "router-based peer-to-peer system" was ISPs and backbone services finally implementing multicast. Give any p2p software author a network where multicast actually works and you'll definitely see a revolution.
Most of these ideas are just gimmicks. One HUGE milestone only gets a footnote: non-volatile RAM.
Look at today's PC. Where is the bottleneck in 95% of all cases? The hard drive.
So, what could be the next killer feature? Non-volatile RAM (PRAM, FRAM, MRAM..). The immediate advantage is speed of course. But there is something much bigger.
Most of the time, loading a file is no longer necessary! Much of the boot time of today's OSes comes from loading stuff into RAM. This can be omitted with P/F/MRAM, reducing booting to device initialization. Also, suspend-to-disk comes for free.
Every single OS is based on the fact that there is a slow, but persistent memory (hard drive) and a fast, volatile one (RAM). They'd need a complete overhaul to fully exploit the new paradigm. Hell, almost all programs too. "Loading file to memory" is not necessary anymore, because the file already IS in memory! Thus, some sort of direct access is needed (unless the file is fragmented).
This sig does not contain any SCO code.
Sorry, but the article is just dumb.
How can you put quantum and organic computing on the same list as a hack to join up a bunch of projectors to make a larger screen and a fricking "beanbag" mouse that you wave about?
Most of the middle-class in any western country *can* affort to spend $12K for any damn thing they please. If it's worth it is another matter entirely. For 99% of the population that's gonna be a no.
Tech tends to fall like a lead-stone in price over time though, can you remember when a simple DVD-player was $3000 ? It's not that many years ago. You know, one of those sucky ones with no network, no divX, no mp3, no jpg, no video-cd compatibility and 10-second lag for layer-changing....
We used to have a $3000 0.8Mpix digital camera at work. Concluding that digital cameras will never appeal to the mass-market based on that would've been the wrong conclusion though....
I went to the soap homepage (http://www.patrickbaudisch.com/projects/soap/inde x.html) and watched the demo. DOA. The gyrations that guys hand has to make to control the mouse, and the speed of the cursor (I know, you can set that, but there's a limit to maintain precision) makes the propsect of using something like that for an extended period of time seem like a CIA torture technique. I use a "regular" optical moust with a wrist pad that has a wrist rest. It requires very little effort, and I can both zip across a 1920pixel screen and precisely pick points in CAD using the same settings. Plus I get three buttons and a scroll wheel (which, if you pan and zoom in good applications is a great movement saver).
In some ways it reminds me of a trackpad. Very cool looking and futuristic (back when they were first introduced) until you try to use it for anything, at which point it becomes a burden which slows down and degrades the accuracy of all of your pointing and selection operations.
Besides, once I get to eliminate my desk, end table, couch, and bed, where should I put my keyboard - or will they come up with a 60wpm on-screen soap-mouse-pick keyboard?
Is it just my observation, or are there way too many stupid people in the world?
Or you could remount a standard trackball mouse into a pistol grip: thumb moves the ball, and you have four buttons on the grip. You can rest your hand however you want, and you've got plenty of accuracy.
I've seen mouses like that. And I've heard good things about trackballs for gaming, though that was compared to joysticks, so I'm not sure how they stack up against regular mouses.
Actually, precision is one of the things I would be concerned about. Think about it - with a desk-bound mouse you can make relatively precise movements in two axes, as the third is constrained and the surface provides support for the device. With a hand held object, you must support it with the same fingers used to manipulate the device. Very few people have perfectly steady hands, which means decreasing the sensitivity to avoid shake - further aggravating the speed issue. Second, most of my mouse operations end in a click. My mouse doesn't move under the pressure required to register a click. It's one of the big problems I have with tapping a touch pad - enough sensitivity to allow useful mouse motion causes drag-clicks (i.e. - misses) with all but the most careful taps. That may not matter for a big Allow or Deny dialog, but for accurate cursor placement in graphics, dense text, or CAD, it's the kiss of prductivity death. Zoom and pan, I hear you cry? So I need to do an extra zoom/pan action to offset the click accuracy? How does that speed up my progress.
No, it's a fun looking device, but I think it may not be the mouse of the future unless a lot of other things change.
Is it just my observation, or are there way too many stupid people in the world?
First, build a fusion reactor that works. Then get Steve Jobs to put it in iPod form-factor. You'll have to send it in for refueling about every 18 months. lol
Actually, you _can_ already do this, but it's still in it's early days yet. When get get as far as seamless support for geographic virtualisation, then we _will_ have revolutionised computing - we're getting there now, but it's still not quite there yet. Cluster each of the objects for failover reasons, and you have a system that at least in potential, has 100% uptime. You can even migrate 'stuff' off sufficiently resilient hardware, replace the failing component, and migrate back seamlessly.
Extend this into remote sessions, such as citrix environments, but with the extra cool that you actually can move the OS instance to be 'close' to the user in terms of network topology and bandwidth.
Of course, the irony is that this isn't so very different from what mainframes were doing, back in the day. I guess things really do come full circle.
The things this guy lists? Meh. They're gimmicks, not revolutions.
"E: A man made brain? That's a revolutionary idea! With our deep understanding of the human psyche and physiological complexities, we could whip this problem in no more than 20 years. Why haven't we been working on this since the 60's?"
I think the idea was that they're working on a new processing paradigm (can't believe I actually used that word) to make computers friendlier to humans. I think the idea is we'll be able to tell computers what we want instead of giving them a literal list of instructions for what to do. For example, today we get our email by opening our email app, clicking 'get email', and we get a list of our messages to browse. Sometime down the road, instead we'll say something like: "Computer, what's in my inbox today?" And it'll say: "A friend of yours sent you an email with an amusing image you might like, but the rest of your messages are unimportant."
Maybe I'm wrong, but I didn't get the impression from the article that it was about building an emo computer.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
Oh for Christ's sake! ...but thank you for posting it.
With Pluribus, you can build a cineplex-quality image using a handful of ordinary, $1,000 PC projectors--in less time than it takes to pop the popcorn
A handful of $1000 projectors? Great idea - if your ame is Bill Gates or Lars Ulrich. For those of us who live on a paycheck and don't have unlimited sources of funding, this just ain't gonna happen. Next?
It's a bit like the Wii remote--only more accurate and far easier to use.
Why stop there? Why not just let me point at the giant Imax sized screen in my giant imax sized house with my bare finger? Don't these guys have any imagination?
Extreme Peer-to-Peer
<yawn>
The Man-Made Brain
How many beads do I have to string on my abacus before it becomes self-aware? After all, your computer is nothing more than a binary abacus with billions of beads. Your brain is electrochemical (with emphasis on the chemical). True thought is a chemical reaction. You can simulate anything, but flying your Microsoft Flight Simulator won't really make you move an inch.
This last one has been predicted as long as flying cars. Now, if they make computers out of biological substances I might change my mind about this.
If you want my turing machine, here it is. The original one was written on a TS1000 with 16K (that's kilo not mega) of memory in 1984. The PC version (1987, runs on DOS) is about 400k, but most of that is compiler overhead, it's not signifigantly different than the TS1000 version or the Apple IIe version.
Mine, "Artificial Insanity", is a turing debunker written on the premise that humans get tired, drunk, crazy, don't pay attenbtion, are smartasses, and have attitudes. So it does too. Warning: Its answers pissed one friend off so much he broke his keyboard typing back at it.
-mcgrew
[1] See for example D. Bacon Phys. Rev. A 70, 032309 (2004)