I saw a couple manufacturers of video glasses at CES. One set wouldn't fit over my glasses at all. The other set was supposed to fit, but didn't. I seem to have a head on the larger end of the spectrum, but still.
And the kicker... IIRC both devices had QVGA resolution. Rather useless for hacking, and not really that good for TV anymore either.
If any manufacturers are listening... I want a set that has large image size, and high resolution. 1280x1024 is barely acceptable, and 1920x1280 would be good. Then you can watch HD, and have enough real estate for a bunch of terminal windows. And yeah, that would be expensive, but surely not nearly as expensive as a 50 inch physical display using LCD, plasma, OLED, or whatever.
The term you are looking for is 'reversible logic'. Theoretically, the only energy you really need to expend for computation is that to store the results.
... and the UK case in which Arsenal Football Club prevented a guy from selling unofficial Arsenal scarves from his front garden - using trade marks to force fans to pay for the overpriced official merchandise.
Forcing fans? No. They chose to purchase overpriced merchandise. If it ever becomes required to purchase a football t-shirt to work in a bakery, or go down to the pub and lift a pint, let me know. Then we can make a case for economic coercion.
Selling overpriced t-shirts is pure capitalism. And unless there is something else wrong (like an unusually high tendancy to catch fire compaired to other shirts) nothing should be done about the situation.
Oh my, how naive you are. (I'm not trying to be mean or snarky.) This is exactly what is supposed to happen, and that is what is written into the law.
However, this is not at all what happens in real life with the USPTO. Through plan or incompetence, almost anything can be patented. Obviousnenss is not a barrier.
Just look at all the patents where someone took an existing business process, and put it on the web. Yah, new patent! Ka-ching!
Yah, I do the same thing. XP luckily has an option under 'Performance' where you can just click 'tune for max performance' and bamf! all that stuff goes away.
I've played around with things like translucent xterms and whatall. Sure, it looks really frickin' cool. But at the end of the day, I need to focus on the foreground. So it is back to white text on a black background. And maybe a few bright colors for syntax highlighting. Its the only way I can get my work done.
I'm not swimming in cash, but I just dropped them a donation. I haven't upgraded to the very latest release, but I believe I am more secure running OpenBSD on my firewall systems. I really like the simplicity, and I like how pf is configured.
Yup, we have to understand better how minds work. Or at least enough to make a copy of them.
And then we also need the processor power equal to that of the brain too. It could well be argued that the Internet crossed that line quite some time ago. But the structure of the Internet is not even close to mind-like. Though there are possibilities...
At any rate, what gets interesting is that we've just recently crossed that same line with "single" entities like the IBM BlueGene supercomputer cluster. We'll probably have a dozen of those online by next year, and hundreds of more powerful ones in five years.
So now we really are waiting for the software. We've also got other advantages compared to what researchers 20 - 30 years ago had. Between Wikipedia and Google, we are in the process of digitizing a large percentage of human knowledge. And Wikipedia can provide a good top-level index into that knowledge.
Next it is a (highly non-trivial) task to improve the ability to map natural language into symbols accurately. Or maybe we can sucker people across the Internet into doing the mapping for us (Tom Sawyer fence painting) by making it fun somehow.
Yup, I agree with the above. In fact, I would go further and say that you have to regularly practice stuff like replacing a drive, or restoring a database to a backup server, to make sure your knowledge and procedures are up to snuff and documented.
Indeed. I've never followed a submitter link at all, until I read a post complaining about Beatles Beatles, and followed a link just to see what it was.
And I've been around a while longer than you! I don't think the submitter link is anything to get worried about.
Then there is just the simple fact of how many kids are going to do something meaningful with a computer when left to their own devices. I usually don't, my kids definitely don't, and my bet is that almost across the board kids won't.
OK, let's assume for the moment, you're completely right. In fact, let's really be pessimistic, and say that one in a thousand of the kids which get these laptops do something useful with them. That's still 100K kids (assuming the 100M laptops get distributed as widely as planned) who now have a much better chance of becoming a doctor, engineer or such. Spending $100K for each of these kids is probably worth it, for most countries. And of the other 999 kids who also got laptops, if any of them benefits from the program in _any_ way, you're in the bonus territory.
Sure when they are "in class" they will, but they will also push the envelope and try to screw around during class time. Reminds me of playing dopewars on the old 286 DOS machines in high school computer class.
Are you _sure_ you didn't learn anything useful playing DopeWars? Even that can help kids with arithmetic a little bit, especially when you need to calculate profits of the various products, vs. how much you can carry. It is a bit of a memory exercise too, remembering the relative prices in different locations.
Even your typical FPS requires you to do at least a little inventory management. You also need to learn how to read a map, and translate that into what you see, which many people (even some adults) aren't real good at doing. Heh, and the FPSs without automaps can really exercise your ability to navigate in 3-D enviroment.
And if you want to really excel at an action game like Diablo II, you'll need to learn how to use complicated formulas for determining things like hit percentage and damage for various weapons. For the rest of your equipment, you'll also need to balance resistances vs. armor rating vs. stat boosts vs. other special abilities. It can be a lot of work, and some of those skills can apply to other areas of your life.
If you can get the kids hooked on a game like SimCity... then they have to learn budgeting, forecasting, tax policy, traffic management, etc., etc., etc..
If we can motivate people to write/port some cool software for this platform (and get it translated), I think it has the potential to do a lot of good.
We can't have our expectations too high with this program. But we don't have to let the lack of mentors/teachers be the limiting factor either. I taught myself to program in BASIC and assembly (6809 rocks!) from books. No one I knew was able to teach me much, until I got to college.
So yeah, most of the kids aren't going to teach themselves math or programming with these laptops. But there are some kids out there who can teach themselves to read, even without much help. Those are the kids we want to find and develop. They'll be able to figure out how to run these laptops. And they'll be able to teach their friends too.
Well, hopefully, with better education, these kids will be able to avoid the disease, famine, and poverty routes. How long are we going to keep the poor of the world in a box, living and dying at our discretion? It is only through education that they will break the cycle of poverty and dependance.
I am not generally a fan of long term government handouts, especially when the systems are poorly designed and encourage dependancy. The one thing which is always safe and good to give out, however, is education. That is the best gift you can give anyone.
These patent troll companies remind me a lot of what LLCs and other corporations evolved into in Charles Stross's Accelerando. You can read it for free, but I encourage you to buy a copy.
It's a very good read. Easily his best work yet. I got quite a kick out of the infovore idea from his "The Atrocity Archives".
Lisp macros are automatic code generators. Actually any kind of GUI builder is a code generator, for that matter, even the ones that output XML documents. Indeed, the libraries that read them in are nothing more than interpreters for DSLs encoded in XML.
In one sense this is true. However, unlike the klunky external code generators, LISP macros are built into the programming language. And, with a few caveats, can be tightly integrated with the rest of your code. LISP macros are a mostly clean and seamless.
With the external code generators, there are always two levels you are operating at. And that's what causes problems during regular programming, but especially when you want to integrate a third-party library.
I've believed for the last few years that automatic code generators (*) are a bad idea.
I haven't spent too much time looking at RoR, except for reading some introductory docs. But when I ran across the code generation bit, that set off alarm bells.
I've used automatic code generation systems a few times. It was always inconvenient. There were always instances where you wanted to do something the code generator couldn't do by itself, so you had to drop into the programming language itself. One system I used had a sort of patch system, so that when you made changes and regenerated the code, it would re-apply your hand modifications. Still was klunky. But the programming language (**) was bad enough to work in directly, so I didn't mind the klunkiness.
Needing any kind of code generator, be it a GUI layout tool, or what's in RoR is a realization that the programming language itself isn't capable of further abstraction. I guess there is no shame in that, most languages aren't capable of abstraction beyond a fixed level.
There is a solution to this kind of thing, however. If you have a programming system which is sufficiently expressive enough, you can keep coding to higher and higher levels of abstraction. I'm talking about things like higher-order functions and macros. Not C preprocessor macros of course; I mean Common Lisp and Scheme macros.
With a good macro system (which Ruby doesn't have, but can kinda / sorta get by without) you can abstract away even the boilerplate code. But if RoR needs a code generator, then maybe you really need higher levels of abstraction that Ruby can provide.
I suspect, though that the code generator was the first solution. Someone of sufficient cleverness may be able to find a cleaner and more elegant way to do the same thing without the code generator. Ruby is not LISP, but it is pretty good.
(*) I don't mean code generators in the loosest sense of the term, so this doesn't include compilers and such. I mean programs which generate high-level language code. As opposed to C, which is a low-level language / high-level assembler.
(**) Informix 4GL, for the old fogie DBAs out there. Used FourGen's CASE tools.
Whatever you think of the Carter administration, the interesting thing to me is that Jimmy Carter has been arguably one of the most effective ex-presidents we've ever seen in recent times.
The guy is out there, doing what he believes in, and making a difference. Promoting affordable housing and democracy, a couple of the USA's core values. All I can say is: Bravo.
Compare that to the rest of the recent ones: Nixon, Ford, Regan, Bush One, Clinton. Mostly they made appearances to pay the bills. And found libraries. And play golf. Dunno, maybe I haven't studied up on them enough (which is to say, at all), but I don't recall hearing anything about any of them which impressed me.
The only exception seems to be Clinton. He seems to be following the same path as Carter, but more ego-centrically. You hear a lot of 'I' when he talks about what he's working on.
Actually, CDargs isn't as neato as I had hoped. The one nice about using it though is that you don't have to type a '$' as you do with plain shell variables. I don't like hitting shift any more than I have to.
Now if you set up aliases to create and maintain the shortcuts, and also automatically create aliases for the variables themselves... I think you'd end up with CDargs.
The comments on the dedicated CDargs article by the same author was useful, because commentors to that also talked about pushing and poping directories, which is something I want to do here and there. Of course, these days I just start a new shell, and then exit out of it to 'pop'. But it wouldn't kill me to learn a little more about bash.
Schemers don't like macros because it is possible to write buggy macros, and want their own elegant, bulletproof mechanisms to define language extensions. T
And it looks like they can get that now with SRFI-72, "Simple hygienic macros". I haven't dug into all the details and compared it to all the other macro systems for Scheme, but it looks pretty good on the surface.
Even mainstream Python won't have macros any time soon. Check out the section "Gotchas for Lisp Programmers in Python" number 12 on the Python for Lisp Programmers guide by Peter Norvig.
What's funny is that the guide is supposed to be an introduction for Lisp programmers to learn Python. But to me, it is damming evidence of why Python is so much less powerful than Common Lisp. Instead of making me want to learn Python, I wanted to learn CL even more!
More properly, Lisp macros operate on the program structure itself. Macros for languages like C and C++ operate on text. I believe this is the why most programmers don't immediately comprehend why Lisp macros are so much more powerful and useful. They are used to macro systems which are useful for for putting CONSTANT_VALUE instead of a hard-coded number in the program text, but quickly become a bear to use when it is time to do something more sophisticated.
Well, I sort of agree with the sentiment. We probably do spend too much time talk about the SCOundrels and not as much making cool stuff.
But we have to keep in mind it is the SCOundrels, M$, and the copyright cartel who are intent on taking away from us all the tools we use to make cool stuff.
So I recommend we pay at least some attention to these matters.
Can someone explain to me what's so wrong and broken with the MedicAlert bracelet system?
Sure it's old-fashioned. But because of this, anyone can read the bracelet and get the needed information. Just call up the 800-number. No special RFID reader needed or Internet access.
Anything which is going to destroy the bracelet is likely to be fatal to me. So I wouldn't be worried about penicillin allergies at this point.
And I don't see how an RFID tag is going to survive much better at that point.
Slashdot Mirror
Yeah.
I saw a couple manufacturers of video glasses at CES. One set wouldn't fit over my glasses at all. The other set was supposed to fit, but didn't. I seem to have a head on the larger end of the spectrum, but still.
And the kicker... IIRC both devices had QVGA resolution. Rather useless for hacking, and not really that good for TV anymore either.
If any manufacturers are listening... I want a set that has large image size, and high resolution. 1280x1024 is barely acceptable, and 1920x1280 would be good. Then you can watch HD, and have enough real estate for a bunch of terminal windows. And yeah, that would be expensive, but surely not nearly as expensive as a 50 inch physical display using LCD, plasma, OLED, or whatever.
The term you are looking for is 'reversible logic'. Theoretically, the only energy you really need to expend for computation is that to store the results.
Forcing fans? No. They chose to purchase overpriced merchandise. If it ever becomes required to purchase a football t-shirt to work in a bakery, or go down to the pub and lift a pint, let me know. Then we can make a case for economic coercion.
Selling overpriced t-shirts is pure capitalism. And unless there is something else wrong (like an unusually high tendancy to catch fire compaired to other shirts) nothing should be done about the situation.
Oh my, how naive you are. (I'm not trying to be mean or snarky.) This is exactly what is supposed to happen, and that is what is written into the law.
However, this is not at all what happens in real life with the USPTO. Through plan or incompetence, almost anything can be patented. Obviousnenss is not a barrier.
Just look at all the patents where someone took an existing business process, and put it on the web. Yah, new patent! Ka-ching!
But that's solvable now -- they could spin up an entire virtualized copy of XP in a VM. It'd be slow, but it'd at least work.
Doesn't even have to be that slow. With Xen, you taking only a small performance hit, 2% to 8%.
The trick, of course, is to get the VM guest OS applications to interact with the rest of your system in a seamless fashion.
Yah, I do the same thing. XP luckily has an option under 'Performance' where you can just click 'tune for max performance' and bamf! all that stuff goes away.
I've played around with things like translucent xterms and whatall. Sure, it looks really frickin' cool. But at the end of the day, I need to focus on the foreground. So it is back to white text on a black background. And maybe a few bright colors for syntax highlighting. Its the only way I can get my work done.
Contrast, contrast, contrast.
Dang, that is messed up. Thanks for the info. Hilarious.
I'm not swimming in cash, but I just dropped them a donation. I haven't upgraded to the very latest release, but I believe I am more secure running OpenBSD on my firewall systems. I really like the simplicity, and I like how pf is configured.
Keep on hacking guys!
Yup, we have to understand better how minds work. Or at least enough to make a copy of them.
And then we also need the processor power equal to that of the brain too. It could well be argued that the Internet crossed that line quite some time ago. But the structure of the Internet is not even close to mind-like. Though there are possibilities...
At any rate, what gets interesting is that we've just recently crossed that same line with "single" entities like the IBM BlueGene supercomputer cluster. We'll probably have a dozen of those online by next year, and hundreds of more powerful ones in five years.
So now we really are waiting for the software. We've also got other advantages compared to what researchers 20 - 30 years ago had. Between Wikipedia and Google, we are in the process of digitizing a large percentage of human knowledge. And Wikipedia can provide a good top-level index into that knowledge.
Next it is a (highly non-trivial) task to improve the ability to map natural language into symbols accurately. Or maybe we can sucker people across the Internet into doing the mapping for us (Tom Sawyer fence painting) by making it fun somehow.
Yup, I agree with the above. In fact, I would go further and say that you have to regularly practice stuff like replacing a drive, or restoring a database to a backup server, to make sure your knowledge and procedures are up to snuff and documented.
Indeed. I've never followed a submitter link at all, until I read a post complaining about Beatles Beatles, and followed a link just to see what it was.
And I've been around a while longer than you! I don't think the submitter link is anything to get worried about.
Then there is just the simple fact of how many kids are going to do something meaningful with a computer when left to their own devices. I usually don't, my kids definitely don't, and my bet is that almost across the board kids won't.
OK, let's assume for the moment, you're completely right. In fact, let's really be pessimistic, and say that one in a thousand of the kids which get these laptops do something useful with them. That's still 100K kids (assuming the 100M laptops get distributed as widely as planned) who now have a much better chance of becoming a doctor, engineer or such. Spending $100K for each of these kids is probably worth it, for most countries. And of the other 999 kids who also got laptops, if any of them benefits from the program in _any_ way, you're in the bonus territory.
Sure when they are "in class" they will, but they will also push the envelope and try to screw around during class time. Reminds me of playing dopewars on the old 286 DOS machines in high school computer class.
Are you _sure_ you didn't learn anything useful playing DopeWars? Even that can help kids with arithmetic a little bit, especially when you need to calculate profits of the various products, vs. how much you can carry. It is a bit of a memory exercise too, remembering the relative prices in different locations.
Even your typical FPS requires you to do at least a little inventory management. You also need to learn how to read a map, and translate that into what you see, which many people (even some adults) aren't real good at doing. Heh, and the FPSs without automaps can really exercise your ability to navigate in 3-D enviroment.
And if you want to really excel at an action game like Diablo II, you'll need to learn how to use complicated formulas for determining things like hit percentage and damage for various weapons. For the rest of your equipment, you'll also need to balance resistances vs. armor rating vs. stat boosts vs. other special abilities. It can be a lot of work, and some of those skills can apply to other areas of your life.
If you can get the kids hooked on a game like SimCity... then they have to learn budgeting, forecasting, tax policy, traffic management, etc., etc., etc..
If we can motivate people to write/port some cool software for this platform (and get it translated), I think it has the potential to do a lot of good.
We can't have our expectations too high with this program. But we don't have to let the lack of mentors/teachers be the limiting factor either. I taught myself to program in BASIC and assembly (6809 rocks!) from books. No one I knew was able to teach me much, until I got to college.
So yeah, most of the kids aren't going to teach themselves math or programming with these laptops. But there are some kids out there who can teach themselves to read, even without much help. Those are the kids we want to find and develop. They'll be able to figure out how to run these laptops. And they'll be able to teach their friends too.
Well, hopefully, with better education, these kids will be able to avoid the disease, famine, and poverty routes. How long are we going to keep the poor of the world in a box, living and dying at our discretion? It is only through education that they will break the cycle of poverty and dependance.
I am not generally a fan of long term government handouts, especially when the systems are poorly designed and encourage dependancy. The one thing which is always safe and good to give out, however, is education. That is the best gift you can give anyone.
These patent troll companies remind me a lot of what LLCs and other corporations evolved into in Charles Stross's Accelerando. You can read it for free, but I encourage you to buy a copy.
It's a very good read. Easily his best work yet. I got quite a kick out of the infovore idea from his "The Atrocity Archives".
Lisp macros are automatic code generators. Actually any kind of GUI builder is a code generator, for that matter, even the ones that output XML documents. Indeed, the libraries that read them in are nothing more than interpreters for DSLs encoded in XML.
In one sense this is true. However, unlike the klunky external code generators, LISP macros are built into the programming language. And, with a few caveats, can be tightly integrated with the rest of your code. LISP macros are a mostly clean and seamless.
With the external code generators, there are always two levels you are operating at. And that's what causes problems during regular programming, but especially when you want to integrate a third-party library.
I've believed for the last few years that automatic code generators (*) are a bad idea.
I haven't spent too much time looking at RoR, except for reading some introductory docs. But when I ran across the code generation bit, that set off alarm bells.
I've used automatic code generation systems a few times. It was always inconvenient. There were always instances where you wanted to do something the code generator couldn't do by itself, so you had to drop into the programming language itself. One system I used had a sort of patch system, so that when you made changes and regenerated the code, it would re-apply your hand modifications. Still was klunky. But the programming language (**) was bad enough to work in directly, so I didn't mind the klunkiness.
Needing any kind of code generator, be it a GUI layout tool, or what's in RoR is a realization that the programming language itself isn't capable of further abstraction. I guess there is no shame in that, most languages aren't capable of abstraction beyond a fixed level.
There is a solution to this kind of thing, however. If you have a programming system which is sufficiently expressive enough, you can keep coding to higher and higher levels of abstraction. I'm talking about things like higher-order functions and macros. Not C preprocessor macros of course; I mean Common Lisp and Scheme macros.
With a good macro system (which Ruby doesn't have, but can kinda / sorta get by without) you can abstract away even the boilerplate code. But if RoR needs a code generator, then maybe you really need higher levels of abstraction that Ruby can provide.
I suspect, though that the code generator was the first solution. Someone of sufficient cleverness may be able to find a cleaner and more elegant way to do the same thing without the code generator. Ruby is not LISP, but it is pretty good.
(*) I don't mean code generators in the loosest sense of the term, so this doesn't include compilers and such. I mean programs which generate high-level language code. As opposed to C, which is a low-level language / high-level assembler.
(**) Informix 4GL, for the old fogie DBAs out there. Used FourGen's CASE tools.
Whatever you think of the Carter administration, the interesting thing to me is that Jimmy Carter has been arguably one of the most effective ex-presidents we've ever seen in recent times.
The guy is out there, doing what he believes in, and making a difference. Promoting affordable housing and democracy, a couple of the USA's core values. All I can say is: Bravo.
Compare that to the rest of the recent ones: Nixon, Ford, Regan, Bush One, Clinton. Mostly they made appearances to pay the bills. And found libraries. And play golf. Dunno, maybe I haven't studied up on them enough (which is to say, at all), but I don't recall hearing anything about any of them which impressed me.
The only exception seems to be Clinton. He seems to be following the same path as Carter, but more ego-centrically. You hear a lot of 'I' when he talks about what he's working on.
Actually, CDargs isn't as neato as I had hoped. The one nice about using it though is that you don't have to type a '$' as you do with plain shell variables. I don't like hitting shift any more than I have to.
Now if you set up aliases to create and maintain the shortcuts, and also automatically create aliases for the variables themselves... I think you'd end up with CDargs.
The comments on the dedicated CDargs article by the same author was useful, because commentors to that also talked about pushing and poping directories, which is something I want to do here and there. Of course, these days I just start a new shell, and then exit out of it to 'pop'. But it wouldn't kill me to learn a little more about bash.
But then, I've been using most of those tools for quite some time.
The one that interested me is cdb, which I hadn't heard of before. I'll have to try that out.
Schemers don't like macros because it is possible to write buggy macros, and want their own elegant, bulletproof mechanisms to define language extensions. T
And it looks like they can get that now with SRFI-72, "Simple hygienic macros". I haven't dug into all the details and compared it to all the other macro systems for Scheme, but it looks pretty good on the surface.
There's people working on that stuff now. Check out Coyotos for information on the BitC programming language.
Even mainstream Python won't have macros any time soon. Check out the section "Gotchas for Lisp Programmers in Python" number 12 on the Python for Lisp Programmers guide by Peter Norvig.
What's funny is that the guide is supposed to be an introduction for Lisp programmers to learn Python. But to me, it is damming evidence of why Python is so much less powerful than Common Lisp. Instead of making me want to learn Python, I wanted to learn CL even more!
More properly, Lisp macros operate on the program structure itself. Macros for languages like C and C++ operate on text. I believe this is the why most programmers don't immediately comprehend why Lisp macros are so much more powerful and useful. They are used to macro systems which are useful for for putting CONSTANT_VALUE instead of a hard-coded number in the program text, but quickly become a bear to use when it is time to do something more sophisticated.
Well, I sort of agree with the sentiment. We probably do spend too much time talk about the SCOundrels and not as much making cool stuff.
But we have to keep in mind it is the SCOundrels, M$, and the copyright cartel who are intent on taking away from us all the tools we use to make cool stuff.
So I recommend we pay at least some attention to these matters.
Can someone explain to me what's so wrong and broken with the MedicAlert bracelet system?
Sure it's old-fashioned. But because of this, anyone can read the bracelet and get the needed information. Just call up the 800-number. No special RFID reader needed or Internet access.
Anything which is going to destroy the bracelet is likely to be fatal to me. So I wouldn't be worried about penicillin allergies at this point. And I don't see how an RFID tag is going to survive much better at that point.