If you're still interested in coding, learn Scala! You'll leverage your existing JVM knowledge, and have exposure to a few currently-hyped trends: Functional programming, actor-based concurrency, etc.
Sure, you can do the same with a tank of gasoline and a spark too. We all know how dangerous Lithium chemistries can be, but luckily there's progress being made. Care to comment on the substance of my post?
Let's see, the Newton died in 1998; at that time a current desktop CPU was roughly 200MHz and 500 MIPS. Something that could fit the power, heat and price profiles of a handheld was more like 50MHz and 50 MIPS. The MessagePad 2100 had 4MB of RAM, for god's sake. Imagine how much modern software you could run with 50 MIPS and 4MB! Arguably, the Newton was running software that was way ahead of its time, albeit hamstrung by having to run on hardware that was merely OF its time.
Handwriting recognition is hard, but what if you could afford to spend a billion instructions per second on it? What if you had a 4 GFLOPS CPU that consumed 15W fully loaded? What if you could fit 1GB of memory and a couple amp-hours of battery in a handheld? Hey, guess what, we can do that now.
Obviously you're right that Canada has no monopoly on public-sector health research. My comment was intended to show that while the premise of this post's headline, namely that the possibility of profit is necessary to motivate research into new drugs and medical procedures, is pretty much the prevailing attitude in the US, it's a different story elsewhere.
I would never argue that corporations should be coerced into doing research that's primarily in the public interest rather than their own, although it's easy to imagine an effective incentive program that could push the balance a bit. I do argue, however, that research that is unlikely to result in any commercial gain, but nonetheless shows promise for benefiting the public, needs to be funded, and that's a job for strong, well-funded public and nonprofit sector agencies.
Thanks for your response, but if you had read my comment more closely, you would see that I suggested that two different classes of organization should:
investigate whether the proposed drug is safe and/or effective as a treatment for various types of cancer (i.e. cancer research agencies, with no profit incentive), and if so:
produce pharmaceuticals to deliver the drug (i.e. generic manufacturers, who make a profit that's somewhat limited by natural competition, rather than inflated by monopoly rents)
...where we believe that governments have a responsibility to set policy for, and even fund, public health initiatives that are not necessarily advantageous to any given industry sector or corporation.
The research in question was funded by a Canadian federal government agency, and I'm certain that oneortwo well-funded, non-profit and/or public sector agencies will step up to the plate to study whether the proposed treatment is safe, and if so, some smart non-intellectual-property-driven and yet profitable organization will market it.
I haven't had cable or broadcast TV (where I live you get about 2.5 channels without cable) in about ten years, but I've watched plenty of movies (and TV shows on DVD) at home during that time.
Initially, my cable service was included in my apartment rent, then they reduced my rent (!) by $22 and unbundled the cable service. The cable company continued the service for free for at least six months, then finally shut it off after receiving no response to their exhortations and incentives. I could easily afford it, but during the cable era, I'd found that my interests were broad enough that I could usually find something marginally interesting to watch, and without TV I had a lot more time.
Fast forward ten years, and we just gave away our TV; it hadn't been turned on in months. If we want to watch video content, we use a 17" laptop or 20" widescreen LCD monitor. Between prompt DVD releases and, uh... alternative means of acquiring content, I just don't feel the need to "tune in," and I certainly don't miss advertising.
I don't plan to get another TV or cable when my son is "old enough" to watch either; I've decided that I want him to learn his culture from people I like, rather than the random jerks who've managed to stumble into the limelight.
We've all heard factoids about the number of hours of TV per day, hours of advertising per year, number of murders, etc. that the average child sees. I, for one, welcome^W don't plan to have an average child.:)
Good post, although as you probably know, NiMH (and NiCd) cells both produce a nominal 1.2V, not 1.5V, so you'd need 23 cells. Nominal LiIon voltage is 3.6V, so can use a 1/3 fewer cells to achieve the same voltage (mobile phones today typically use a single LiIon cell, whereas three NiMH or NiCd cells in series would be required).
On the downside, it was only quite recently that (ironically) Sony discovered how they could be used safely at all in consumer applications.:)
I've downloaded and installed OpenCyc 1.0, it works fine (after quite a long initial startup delay and with enough swap) on a 2GB machine.
I've been playing with it for a couple of hours, and I have a question.
I've created the following constants for my cats, their sibling and parents:
#$Comet-TheCat
#$Rocket-TheCat
#$Packet-TheCat
#$Mama-TheCat
#$GhostDad-TheCat
I've asserted (#$isa [cat] #$Cat) about all of them.
I've asserted (#$biologicalMother [cat] #$Mama-TheCat) about Comet, Rocket and Packet
I've asserted (#$biologicalFather [cat] #$GhostDad-TheCat) about Comet, Rocket and Packet as well.
I even created #$ConceptionOfKitties, asserted (#$isa #$ConceptionOfKitties #$BiologicalReproductionEvent), (#$parentActors #$ConceptionOfKitties #$Mama-TheCat) and (#$parentActors #$ConceptionOfKitties #$GhostDad-TheCat).
So why can't Cyc infer that (#$siblings #$Comet-TheCat #$Packet-TheCat)? Is it a limitation in the public subset of the ontology, or some more fundamental issue with my data?
The title of this book is disingenuous! XML and PHP each come with a huge helping of nonsense that you can't get rid of. Together, they're... *shudder*
I remembered reading a fawning article on a tech recycling outfit in Wired or somesuch back in the iDolatry era (1995-7?) and wondered if it was the same one you've been on about. I was a few pages deep into the Google results for "california PC recycling" before I spotted the word "Australia.":)
BTW, next time you're bored of your job, drop us a line.:)
This thing, from http://www.infrant.com/, can be had for around $600 US without drives, which is not much more than it would cost to build your own. It lets you start with two mirrored drives in RAID 1, then add another drive or two, and it transparently migrates to RAID 5. Pretty cool.
Note I have no financial interest in Infrant, I just want their products.:)
It's not just game programmers who are "weird." Most people, including programmers, just aren't very good at doing things they haven't already done before. And just because you can write code and 95% of the population can't doesn't mean you have the talent or skill to handle any coding task thrown at you.
You and I have been comfortable coding for multithreaded environments for awhile, but the game industry hasn't been forced to feel that pain yet, and Gabe Newell is pissing into the wind about it when he should be spending all his time trying to figure out how to make best use of the hardware we're going to get rather than the hardware we wish we could get.
AMD, Intel and IBM would love to continue providing the exponential improvements in single-threaded, in-order execution we've been enjoying for thirty years, but it doesn't look likely this trend will continue even at geometric rates into the foreseeable future.
Of course, particular JVM implementations (in this case the outdated and increasingly unpopular MS VM) are subject to the same rules as any other software. My point is that the design of Java, including the verifier, security manager and default applet policy, provide an effective (but of course not perfect) prophylactic against most common types of mobile code security problems.
It's not a bad idea per se to automatically download and run stuff from the Internet, but any software designed to do so had better be designed and implemented properly. The dozens (hundreds?) of "cross-site scripting" bugs that have surfaced in popular browsers in the past few years are evidence that this is rarely done well. Java's 10 year old sandbox design has been quite successful, and Flash has followed a substantially similar design.
Unfortunately, code signing, as currently implemented and (mis)understood by users, is an all-or-nothing proposition. There are certainly legitimate uses for privileged mobile code, but most users don't really read or understand security warning dialogs, they just think "I just clicked the Start Game button, and now it's asking me if I really want to Start the Game. How stupid."
Marimba actually came up with a good partial solution ages ago. When their framework loaded and executed a Java app, the framework would closely manage exactly what resources could be exploited by the app. Each application's ability to read and write files was restricted by default to its own tiny corner of the filesystem, and the amount of space it could occupy with its files was constrained as well.
Note that Java's security manager infrastructure has allowed these sorts of fine-grained controls since 1.2 (circa 1998), but no one to my knowledge has yet found a way to effectively communicate to a user:
what resources a given piece of mobile code will want to exploit;
what the risks of running it might be;
some assurance that the code is published by someone they trust;
...While maintaining some degree of user-friendliness. It's a tough problem.
MSIE's concept of local policies set according to centrally defined security zones was a step in the right direction; it's too bad its development stalled when the Browser War was "won."
Art is a pretty amazingly nerdy guy. That site is about the impressive but ill-fated "Mark I" glider.
In the time since the site was put together he's built a remotely-piloted submarine and has been working on Mark II, which will probably feature a much better camera system, a modular "mission bay" and most importantly considering the way Mark I left this world, some awareness of the height of the terrain in the flight area.
Most likely he'll manually enter landmarks like the mountain that likely put paid to Mark I, but if anyone knows of any conveniently and freely accessible topographical data, especially of the us Northwest and Western Canada, please let me know and I'll pass it along to Art.
Slashdot Mirror
If you're still interested in coding, learn Scala! You'll leverage your existing JVM knowledge, and have exposure to a few currently-hyped trends: Functional programming, actor-based concurrency, etc.
I logged in just to say the same thing. I love this story.
Unfortunately I don't think the whole Australia/Utopia aspect of it is very likely... The dystopian part seems like where we're headed though.
Maybe there's some way to set up a trillion-dollar escrow fund though... I'd pledge my $1k.
Sure, you can do the same with a tank of gasoline and a spark too. We all know how dangerous Lithium chemistries can be, but luckily there's progress being made. Care to comment on the substance of my post?
Let's see, 1 Ah == 1000 mAh
NiMH AA batteries typically have at least 2000 mAh.
The LiPoly cells used in phones often have 1500 mAh.
Mind you, you can't draw one amp out of any small battery for any length of time, much less an hour, but the units are convertible.
Let's see, the Newton died in 1998; at that time a current desktop CPU was roughly 200MHz and 500 MIPS. Something that could fit the power, heat and price profiles of a handheld was more like 50MHz and 50 MIPS. The MessagePad 2100 had 4MB of RAM, for god's sake. Imagine how much modern software you could run with 50 MIPS and 4MB! Arguably, the Newton was running software that was way ahead of its time, albeit hamstrung by having to run on hardware that was merely OF its time.
Handwriting recognition is hard, but what if you could afford to spend a billion instructions per second on it? What if you had a 4 GFLOPS CPU that consumed 15W fully loaded? What if you could fit 1GB of memory and a couple amp-hours of battery in a handheld? Hey, guess what, we can do that now.
Obviously you're right that Canada has no monopoly on public-sector health research. My comment was intended to show that while the premise of this post's headline, namely that the possibility of profit is necessary to motivate research into new drugs and medical procedures, is pretty much the prevailing attitude in the US, it's a different story elsewhere.
I would never argue that corporations should be coerced into doing research that's primarily in the public interest rather than their own, although it's easy to imagine an effective incentive program that could push the balance a bit. I do argue, however, that research that is unlikely to result in any commercial gain, but nonetheless shows promise for benefiting the public, needs to be funded, and that's a job for strong, well-funded public and nonprofit sector agencies.
...where we believe that governments have a responsibility to set policy for, and even fund, public health initiatives that are not necessarily advantageous to any given industry sector or corporation.
The research in question was funded by a Canadian federal government agency, and I'm certain that one or two well-funded, non-profit and/or public sector agencies will step up to the plate to study whether the proposed treatment is safe, and if so, some smart non-intellectual-property-driven and yet profitable organization will market it.
I haven't had cable or broadcast TV (where I live you get about 2.5 channels without cable) in about ten years, but I've watched plenty of movies (and TV shows on DVD) at home during that time.
:)
Initially, my cable service was included in my apartment rent, then they reduced my rent (!) by $22 and unbundled the cable service. The cable company continued the service for free for at least six months, then finally shut it off after receiving no response to their exhortations and incentives. I could easily afford it, but during the cable era, I'd found that my interests were broad enough that I could usually find something marginally interesting to watch, and without TV I had a lot more time.
Fast forward ten years, and we just gave away our TV; it hadn't been turned on in months. If we want to watch video content, we use a 17" laptop or 20" widescreen LCD monitor. Between prompt DVD releases and, uh... alternative means of acquiring content, I just don't feel the need to "tune in," and I certainly don't miss advertising.
I don't plan to get another TV or cable when my son is "old enough" to watch either; I've decided that I want him to learn his culture from people I like, rather than the random jerks who've managed to stumble into the limelight.
We've all heard factoids about the number of hours of TV per day, hours of advertising per year, number of murders, etc. that the average child sees. I, for one, welcome^W don't plan to have an average child.
Does your keyboard have a Sleep button? That one will bluescreen my WinXP box every time.
Good post, although as you probably know, NiMH (and NiCd) cells both produce a nominal 1.2V, not 1.5V, so you'd need 23 cells. Nominal LiIon voltage is 3.6V, so can use a 1/3 fewer cells to achieve the same voltage (mobile phones today typically use a single LiIon cell, whereas three NiMH or NiCd cells in series would be required).
:)
On the downside, it was only quite recently that (ironically) Sony discovered how they could be used safely at all in consumer applications.
What distinguishes a rule from any other assertion?
Ah, yes, that does the trick, thanks!
- I've created the following constants for my cats, their sibling and parents:
- #$Comet-TheCat
- #$Rocket-TheCat
- #$Packet-TheCat
- #$Mama-TheCat
- #$GhostDad-TheCat
- I've asserted (#$isa [cat] #$Cat) about all of them.
- I've asserted (#$biologicalMother [cat] #$Mama-TheCat) about Comet, Rocket and Packet
- I've asserted (#$biologicalFather [cat] #$GhostDad-TheCat) about Comet, Rocket and Packet as well.
- I even created #$ConceptionOfKitties, asserted (#$isa #$ConceptionOfKitties #$BiologicalReproductionEvent), (#$parentActors #$ConceptionOfKitties #$Mama-TheCat) and (#$parentActors #$ConceptionOfKitties #$GhostDad-TheCat).
So why can't Cyc infer that (#$siblings #$Comet-TheCat #$Packet-TheCat)? Is it a limitation in the public subset of the ontology, or some more fundamental issue with my data?The title of this book is disingenuous! XML and PHP each come with a huge helping of nonsense that you can't get rid of. Together, they're... *shudder*
Yeah, my-first-name at my-first-and-last-name.com
Or, alternatively, my-first-name at (the-opposite-of-bogon)flux.com
I remembered reading a fawning article on a tech recycling outfit in Wired or somesuch back in the iDolatry era (1995-7?) and wondered if it was the same one you've been on about. I was a few pages deep into the Google results for "california PC recycling" before I spotted the word "Australia." :)
:)
BTW, next time you're bored of your job, drop us a line.
Is HMR the company you worked at?
This thing, from http://www.infrant.com/, can be had for around $600 US without drives, which is not much more than it would cost to build your own. It lets you start with two mirrored drives in RAID 1, then add another drive or two, and it transparently migrates to RAID 5. Pretty cool.
:)
Note I have no financial interest in Infrant, I just want their products.
It's not just game programmers who are "weird." Most people, including programmers, just aren't very good at doing things they haven't already done before. And just because you can write code and 95% of the population can't doesn't mean you have the talent or skill to handle any coding task thrown at you.
You and I have been comfortable coding for multithreaded environments for awhile, but the game industry hasn't been forced to feel that pain yet, and Gabe Newell is pissing into the wind about it when he should be spending all his time trying to figure out how to make best use of the hardware we're going to get rather than the hardware we wish we could get.
AMD, Intel and IBM would love to continue providing the exponential improvements in single-threaded, in-order execution we've been enjoying for thirty years, but it doesn't look likely this trend will continue even at geometric rates into the foreseeable future.
Of course, particular JVM implementations (in this case the outdated and increasingly unpopular MS VM) are subject to the same rules as any other software. My point is that the design of Java, including the verifier, security manager and default applet policy, provide an effective (but of course not perfect) prophylactic against most common types of mobile code security problems.
Sorry, you're right, I had XSS bugs confused with cross-domain scripting bugs, which do affect browsers.
Unfortunately, code signing, as currently implemented and (mis)understood by users, is an all-or-nothing proposition. There are certainly legitimate uses for privileged mobile code, but most users don't really read or understand security warning dialogs, they just think "I just clicked the Start Game button, and now it's asking me if I really want to Start the Game. How stupid."
Marimba actually came up with a good partial solution ages ago. When their framework loaded and executed a Java app, the framework would closely manage exactly what resources could be exploited by the app. Each application's ability to read and write files was restricted by default to its own tiny corner of the filesystem, and the amount of space it could occupy with its files was constrained as well.
Note that Java's security manager infrastructure has allowed these sorts of fine-grained controls since 1.2 (circa 1998), but no one to my knowledge has yet found a way to effectively communicate to a user:
MSIE's concept of local policies set according to centrally defined security zones was a step in the right direction; it's too bad its development stalled when the Browser War was "won."
The passive voice will be used until all are made sorry!
Art is a pretty amazingly nerdy guy. That site is about the impressive but ill-fated "Mark I" glider.
In the time since the site was put together he's built a remotely-piloted submarine and has been working on Mark II, which will probably feature a much better camera system, a modular "mission bay" and most importantly considering the way Mark I left this world, some awareness of the height of the terrain in the flight area.
Most likely he'll manually enter landmarks like the mountain that likely put paid to Mark I, but if anyone knows of any conveniently and freely accessible topographical data, especially of the us Northwest and Western Canada, please let me know and I'll pass it along to Art.